GRUNDIG SUPER COLOR 6236/30 SERIE 30F22 CHASSIS 29301-114.68(07) INTERNAL VIEW.

GRUNDIG SUPER COLOR 6236/30 SERIE 30F22 CHASSIS 29301-114.68(07)

-Right side

Frame oscillator
E/W Correction Unit

Line deflection Stabiliser circuit unit

Line deflection output with Thyristor circuit.

Line deflection transformer + EHT Tripler

- Left side

St-By supply

Safety start circuit

Luminance + Chrominance

RGB Matrix + RGB Amplifier

IF Unit

Sound Unit

Tuning drive and controls (Metal boxed Unit)

-Center

Synchronization Unit

G2 Regulators

GRUNDIG SUPER COLOR 6236/30 SERIE 30F22 CHASSIS 29301-114.68(07)

HOW THYRISTOR LINE DEFLECTION OUTPUT SCAN STAGES WORK:

INTRODUCTION:

The massive demand for colour television receivers in Europe/Germany in the 70's  brought about an influx of sets from the continent. Many of these use the thin -neck (29mm) type of 110° shadowmask tube and the Philips 20AX CRT Tube, plus the already Delta Gun CRT . 

Scanning of these tubes is accomplished by means of a toroidally wound deflection yoke (conventional 90° and thick -neck 110° tubes operate with saddle -wound deflection coils). The inductance of a toroidal yoke is very much less than that of a saddle -wound yoke, thus higher scan currents are required. The deflection current necessary for the line scan is about 12A peak -to -peak. This could be provided by a transistor line output stage but a current step-up transformer, which is bulky and both difficult and costly to manufacture, would be required. 

An entirely different approach, pioneered by RCA in America and developed by them and by ITT (SEL) in Germany, is the thyristor line output stage. In this system the scanning current is provided via two thyristors and two switching diodes which due to their characteristics can supply the deflection yoke without a step-up transformer (a small transformer is still required to obtain the input voltage pulse for the e.h.t. tripler). The purpose of this article is to explain the basic operation of such circuits. The thyristor line output circuit offers high reliability since all switching occurs at zero current level. C.R.T. flashovers, which can produce high current surges (up to 60A), have no detrimental effects on the switching diodes or thyristors since the forward voltage drop across these devices is small and the duration of the current pulses short. If a surge limiting resistor is pro- vided in the tube's final anode circuit the peak voltages produced by flashovers seldom exceed the normal repetitive circuit voltages by more than 50-100V. This is well within the device ratings.

 

Brief Basics: LINE Scan output stages operate on the same basic principle whether the active device used is a valve, transistor or thyristor. As a starting point, let's remind ourselves of this principle, which was first developed by Blumlein in 1932. The idea in its simplest form is shown in Fig. 1. The scan coils, together with a parallel tuning capacitor, are connected in series with a switch across the h.t. supply. When the switch is closed - (a) - current flows through the coils, building up linearly as required to deflect the beam from the centre to the right-hand side of the screen. At this point the switch is opened. The coils and the capacitor then form a resonant circuit. The magnetic fields generated around the coils during the preceeding forward scan as current flowed through them when the switch was closed now collapse, charging the capacitor - (b). As a result of the resonant action the capacitor next discharges, driving current through the coils in the opposite direction - (c). Once more magnetic fields are generated around the coils. This resonant action lasts for one half -cycle of oscillation, during which the beam is rapidly deflected from the right- hand side to the centre and then to the left-hand side of the screen. The flyback is thus complete. If the switch is now closed again further oscillation is prevented and, as the magnetic fields around the coils collapse, a decaying current flows through them in the direction shown at (d). This decaying current flow deflects the beam from the left-hand side of the screen back towards the centre: the period during which this occurs is often referred to as the energy recovery part of the scanning cycle. When the current has decayed to zero we are back at the situation shown at (a): the current through the coils reverses, driving the beam to the right-hand side of the screen. This is a very efficient System, since most of the energy drawn from the supply is subsequently returned to it. There is negligible resistance in the circuit, so there is very little power loss.

 Basic Transistor Circuit:

 In Blumlein's day valves had to be used to perform the switching action. Two were required since a valve is a unidirectional device, and as we have seen current must flow through the switch in both directions. Nowadays we generally use a transistor to perform the switching action, arranging the circuit along the lines shown in Fig. 2. The line output transformer T is used as a load for the transistor and as a simple means of generating the e.h.t. and other supplies required by the receiver. The scan -correction capacitor Cs also serves as a d.c. block. Capacitor Ct tunes the coils during the flyback when the transistor is cut off. During the forward scan Cs first charges, then discharges, via the scan coils, thus providing deflection from the left- hand side to the right-hand side of the screen. One advantage of a transistor is that it can conduct in either direction. Thus unless we are operating the stage from an 1.t. line of around 11V - as in the case of many small -screen portables - we don't need a second switching device. With a supply of 11-12V a shunt efficiency diode - connected in parallel with the transistor, cathode to collector and anode to emitter, is required because the linearity is otherwise unacceptable. Another advantage of a transistor compared to a valve is that it is a much more efficient switch. When a transistor is saturated both its junctions are forward biased and its collector voltage is then at little more than chassis potential. The anode voltage of a saturated pentode however is measured in tens of volts, and this means that there is considerable wasteful dissipation. Thyristor Switch If what we need is an efficient switch, why not use a thyristor??? 

Thyristors are even more efficient switches than transistors. They are more rugged, can pass heavy currents, and are insensitive to the voltage overloads that can kill off transistors. In addition, in the sort of circuit we are about to look at the power supply requirements can be simplified (a line output transistor must be operated in conjunction with a stabilised power supply: this is not necessary in the thyristor circuit since regulation can be built in). In the nature of things however there must be disadvantages as well - and there are! First, a thyristor will not act as a bidirectional switch. 

There is no great problem here however: all we need do is to shunt it with a parallel efficiency diode. More awkward is the fact that once a  thyristor has been triggered on at its gate it cannot be switched off again by any further action taken in its gate circuit. In fact it's this problem of operating the thyristor switch that is responsible for the complexity of thyristor line output circuits. 

A thyristor can be switched off only by reducing the current through it below the "hold on" value, either by momentarily removing the voltage across the device or by passing an opposing current through it in the opposite direction - this latter technique is used in practical thyristor line output circuits. Once the reverse current through the thyristor is about equal to the forward current flowing through it the net current falls below the "hold on" value and the thyristor switches off.

 Basic Thyristor Circuit:

 There is more than one way of arranging a thyristor line output stage. Only one basic circuit has been used so far however, though as you'd expect there are differences in detail in the circuits used by different setmakers. The basic circuit was first devised and put into production by RCA in the USA in the late 1960s. It was subsequently popularised in Europe by ITT, and many continental setmakers have used it, mainly in colour receiver chassis fitted with 110° delta gun c.r.t.s. They include Finlux, Grundig, Saba, Siemens and ASA. Korting use it in their 55636 chassis which is fitted with a 90° PIL tube, while Grundig continue to use it in their latest sets which use the Mullard/Philips 20AX tube. 

Amongst Japanese setmakers, Sharp use it in their Model C1831H which is fitted with a Toshiba RIS tube. 

Reduced to its barest essentials, the circuit takes the form shown in Fig. 3. To start with this looks strange indeed! The right-hand side however is simply the equivalent of the scanning section of the transistor circuit shown in Fig. 2, with TH2 and D2 replacing the transistor as the bidirectional switch.  

The tuning capacitor however is returned to chassis via the left-hand side of the circuit - in consequence there is no d.c. path between the right-hand and left-hand sides of the circuit. L1 provides a load. The efficiency diode D2 conducts during the first part of the forward scan, after which TH2 is switched on to drive the beam towards the right-hand side of the screen. The purpose of the left-hand side of the circuit, the bidirectional switch TH1/D1 and L2, together with the tuning capacitor Ct, is to switch TH2 off and to provide the flyback action.

 The output from the line oscillator consists  of a brief pulse to initiate the flyback. It occurs just before the flyback time (roughly 3µS before) and is applied to the gate of TH1, switching it on. When this happens L2 is connected to chassis and current flows into it, discharging Ct (previously charged from the h.t. line). L2 is called the commutating coil, and forms a resonant circuit with Ct. Thus when TH1 is switched on a sudden pulse builds up and this is used to switch off TH2. In addition to tuning L2, Ct tunes the scan coils to provide the usual flyback action. 

Roughly speaking therefore D2 and TH2 conduct alternately during the forward scan and are cut off during the flyback, while TH1 is triggered on just before the flyback, TH1 and D I subsequently conducting alternately during the flyback and then cutting off when the efficiency diode takes over. 

 Thyristor Line Scan Practical Circuit:

 A more practical arrangement is shown in Fig. 4. A secondary winding L3 is added to Ll to provide the trigger pulse for TH2: L4, C4 and R I provide the pulse shaping required. The tuning capacitor Ct is rearranged as a T network: this is done to reduce the voltage across the individual capacitors and enable smaller values to be used, all in the interests of economy. And finally a transformer is coupled to the circuit by C5 to make use of the flyback pulse for e.h.t. generation and to provide other supplies. In many recent chassis THUD 1 and TH2/D2 are encapsu- lated together, in pairs. In practical circuits L1 and L2 generally consist of a single transformer - often a transductor is used, for convenience rather than for the transductor characteristics. This makes practical circuits look at first glance rather different to the basic form shown in Figs. 3 and 4. A further winding is often added to the transformer to provide a supply for other parts of the receiver, making the circuit look even more confusing. In addition e.h.t. regulation, pincushion distortion correction and beam limiting circuitry is required, and protection circuits may be incorporated.

 

Scanning Sequence:  It's time to look at the basic scanning sequence in more detail, basing the description on Figs. 3 and 4. We'll start at the beginning of the flyback. TH2 and D2 have just been switched off - we'll come to how this is done later - while  TH1 which was triggered on by a pulse from the line oscillator is still conducting. Energy is stored in the scan coils in the form of magnetic fields. As these collapse, a decaying current flows via the coils, Cs, Ct, L2 and TH 1. When this current falls to zero the charge on Ct will have reversed and TH 1 will switch off. This completes the first half of the flyback. The left-hand plate of Ct is charged negatively, while its right-hand plate carries a positive charge. D1 is now biased on and Ct discharges back into the scan coils to give the second half of the flyback. Current is flowing via D1, L2, Ct, Cs and the scan coils. At the end of this period the circuit energy will have been transferred once again to the scan coils - in the form of magnetic fields. One complete half cycle of oscillation will have occurred, returning the beam to the left-hand side of the screen. With Ct discharged, D 1 switches off. The oscillation tries to continue in the negative direction, but we then get the normal efficiency diode action, i.e. D2 conducts shorting out the tuned circuit. As the fields around the coils collapse a linearly decaying current flows via the coils, Cs and D2. This gives us the first part of the forward scan. Towards the centre of the screen TH2 is switched on by the pulse obtained from L3 and the current in the scan coils reverses to complete the scan.  

 Switching the Scan Thyristor Off:  The tricky part is when it comes to switching TH2 off. As we have seen, TH1 is triggered on about 3fitS before the end of the forward scan. Prior to this Ct will have been charged to the h.t. potential via L 1 and L2. When TH1 conducts, current flows via TH1, L2, Ct and TH2 (which is on remember). Because of the tuned circuit formed by L2 and Ct, the current builds up rapidly in the form of a pulse - the commutating pulse shown in Fig. 5. When this current, which flows through TH2 in the opposite direction to the scan current, exceeds the scan current TH2 switches off. Once TH2 cuts off D2 is able to conduct - it is no longer reverse biased - which it does for a short period to provide an earth return path for the remaining duration of the commutating pulse and also to enable the scan to be completed (Cs discharging via the scan coils). When the reverse, commutating current falls below the scan current D2 switches off and we then get the flyback action as the magnetic fields around the coils collapse.

 Power Transferring ; during the forward scan Ct is charged via L1 and L2, its right-hand plate being held at little above

 through the conduction of D2 and then TH2. During the flyback, when TH1 and D1 conduct alternately, connecting the junction L1, L2 to chassis, Ct supplies energy to the scan part of the circuit. The Practical Circuit so much then for the basic circuit and its action. Turning now to a practical circuit, Fig. 6 shows the thyristor line output stage used in the Grundig SuperColor  Models 5011 and 6011. Ty511/Di511 form the flyback switch, T1 is the input/commutating transformer, C516/7/8 comprise the tuning capacitance, Di518 is the efficiency diode and Ty518 the forward scan thyristor. The scan -correction capacitor Cs is C537. As can be seen, the line output transformer circuit is quite conventional. The main complication arises because of the need to provide width/e.h.t. stabilisation. In a valve line output stage it is a simple matter to achieve stabilisation by using a v.d.r. in a feedback circuit to alter the bias on the output pentode. We can't do this with a transistor line output stage, so we have to operate this in conjunction with a stabilised supply. There is a subtle but quite simple method of applying stabilisation to a thyristor line output stage however. As we have seen, the energy supplied to the output side of the circuit is provided by the tuning capacitors when they discharge during the flyback period. During the forward scan they charge via the input coil - or transformer as it is in practice. Now if we shunt the transformer's input winding with a transductor we can control the inductance in series with the tuning capacitors and in consequence the charging time of the capacitors and hence the power supplied to the output side of the circuit.

 

EHT/Width Stabilisation:

 The stabilising transductor in Fig. 6 is Td 1, whose load windings are connected in series with R504/Di504 across the input winding of T1. The transductor's control winding is driven by transistor Tr506, which senses the h.t. voltage (via R506) and the amplitude of the signal at tag d on the line output transformer. R508 in the transistor's base circuit enables the e.h.t. to be set to the correct voltage (25kV). 

 

Other Circuit Details: A fourth winding on Ti feeds the 1.t. rectifier and stabiliser which provide the supply for the low -power circuits in the receiver. The trigger pulse winding also feeds a stabilised 1.t. supply circuit (21V). 

EW pincushion distortion correction is applied by connecting the load windings of a second transductor (Td2) across a section of the line output transformer's primary winding. By feeding a field frequency waveform to the control winding on this transductor the line scanning is modulated at field frequency. There is a simple but effective safety circuit in this Grundig line output stage. If the voltage at tag c on the line output transformer rises above 68V zener diode Di514 conducts, triggering thyristor Ty511 into conduction with the result that the cut-out operates. C517 is returned to chassis via a damped coil (L517) so that the voltage transient when the efficiency diode cuts off is attenuated. Likewise L512/C512/R512 are added to suppress the voltage transient when the flyback thyristor Ty511 cuts off. The balancing coil L516 is included to remove unwanted voltage spikes produced by the thyristors. 

 

At the end........This Grundig circuit is representative of the way in which thyristor line output circuits are used in practice. There are differences in detail in the thyristor line output stages found in other setmakers' chassis, but the basic arrangement will be found to be substantially 

 

Servicing / Throubleshooting / Repairing Thyristor Line Scan Timebases Crt Deflections circuits:

LARGELY due to advances in colour c.r.t. scan coil design, the thyristor line output stage has become obsolete laready in the 1981's.

 It  was a very good system to use where the line scan coils require large peak currents with only a moderate flyback voltage - an intrinsic characteristic of toroidally wound deflection coils.

it was originally devised by RCA. Many sets fitted with 110°, narrow -neck delta -gun tubes used a thyristor line output stage - for example those in the Grundig and Saba ranges and the Finlux Peacock , Indesit, Siemens, Salora, Metz, Nordmende, Blaupunkt, ITT, Seleco, REX, Mivar, Emerson, Brionvega, Loewe, Galaxi, Stern, Zanussi, Wega, Philco. The circuit continued to find favour in earlier chassis designed for use with in -line gun tubes, examples being found in the Grundig and Korting ranges - also,  Indesit, Siemens, Salora, Metz, Nordmende, Blaupunkt, ITT, Seleco, REX, Mivar, Emerson, Brionvega, Loewe, Galaxi, Stern, Zanussi, Wega, Philco the Rediffusion Mk. III chassis. Deflection currents of up to 13A peak -to -peak are commonly encountered with 110° tubes, with a flyback voltage of only some 600V peak  to peak. The total energy requirement is of the order of 6mJ, which is 50 per cent higher than modern 110° tubes of the 30AX and S4 variety with their saddle -wound line scan coils.   The only problem with this type of circuit is the large amount of energy that shuttles back and forth at line frequency. This places a heavy stress on certain components. Circuit losses produce quite high temperatures, which are concentrated at certain points, in particular the commutating combi coil. This leads to deterioration of the soldered joints around the coil, a common cause of failure. This can have a cumulative effect, a high resistance joint increasing the local heating until the joint becomes well and truly dry -a classic symptom with some Grundig / Emerson sets. The wound components themselves can be a source of trouble, due to losses - particularly the combi coil and the regulating transductor. Later chassis are less prone to this sort of thing, partly because of the use of later generation, higher efficiency yokes but mainly due to more generous and better design of the wound components. The ideal dielectric for use in the tuning capacitors is polypropylene (either metalised or film). It's a truly won- derful dielectric - very stable, with very small losses, and capable of operation at high frequencies and elevated temperatures. It's also nowadays reasonably inexpensive. Unfortunately many earlier chassis of this type used polyester capacitors, and it's no surprise that they were inclined to give up. When replacing the tuning capacitors in a thyristor line output stage it's essential to use polypropylene types -a good range of axial components with values ranging from 0.001µF to 047µF is available from RS Components, enabling even non-standard values to be made up from an appropriate combination. Using polypropylene capacitors in place of polyester ones will not only ensure capacitor reliability but will also lower the stress on other components by reducing the circuit losses (and hence power consumption).

       Numerous circuit designs for completely transistorized television receivers either have been incorporated in commercially available receivers or have been described in detail in various technical publications. One of the most troublesome areas in such transistor receivers, from the point of View of reliability and economy, lies in the horizontal deflection circuits.

       As an attempt to avoid the voltage and current limitations of transistor deflection circuits, a number of circuits have been proposed utilizing the silicon controlled rectifier (SCR), a semiconductor device capable of handling substantially higher currents and voltages than transistors.

       The circuit utilizes two bi-directionally conductive switching means which serve respectively as trace and commutating switches. Particularly, each of the switching means comprises the parallel combination of a silicon controlled rectifier (SCR) and a diode. The commutating switch is triggered on shortly before the desired beginning of retrace and, in conjunction with a resonant commutating circuit having an inductor and two capacitors, serves to turn off the trace switch to initiate retrace. The commutating circuit is also arranged to turn oft the commutating SCR before the end of retrace.  

Circuit Operation:

The basic thyristor line output stage arrangement used in all these chassis is shown in Fig. 1 - it was originally devised by RCA. The part to the right of the tuning capacitance acts in exactly the same manner as a transis- tor line output stage, with the scan thyristor Th2 replacing the transistor. The thyristor is switched on about half way through the forward scan, the efficiency diode D2 provid- ing the initial part of the line scan (left-hand side of the screen). The scan coils and line output transformer (used to generate the e.h.t. plus various other supply lines and pulse waveforms as required) are a.c. coupled, via the scan -correction capacitor C5 and C6 respectively. The problem with a thyristor is that it can be turned on at its gate but not off. To switch a thyristor off, the current flowing through it must be reduced below a value known as the hold -on current. This is the main function of the components on the left-hand side - the line generator, the flyback thyristor with its parallel diode and the commutat- ing coil. During the forward scan, the tuning capacitors are charged from the h.t. line via the input and commutat- ing coils. The line generator produces a pulse to trigger the flyback thyristor Th1- this occurs just before the actual flyback. When Thl1 switches on, the junction of the  input coil and the commutating coil is momentarily con- nected to chassis. The tuning capacitance and the com- mutating coil then resonate, producing a pulse which draws current via the scan thyristor. Since this current flow is in the opposite direction to the scan current flow, the two cancel and the current flowing via the scan thyris- tor falls below the hold -on current. Th2 is thus switched off, and the scan coils resonate with the tuning capaci- tance to provide the flyback action. So much for the basic action. A secondary winding coupled to the input coil produces a pulse to switch the scan thyristor on, in conjunction with the shaping/delay network Ll, C4, R1. The tuning capacitors are usually arranged in the T formation shown to reduce the values required and the voltages developed across them. In practical circuits the input and commutating coils are usually combined in a single unit which for obvious reasons is generally known as the combi coil. The main point not so far mentioned is stabilisation. There are two approaches to this. In earlier circuits a transductor was included in parallel with the input coil to vary the impe- dance in series with the tuning capacitance. This was driven by a transistor which was in turn controlled by feedback from the line output transformer. A more efficient technique is used in later circuits, with a current dumping thyristor in series with the input coil. Practical Circuit As a typical example of the earlier type of circuit, Fig. 2 shows the thyristor line output stage used in the Grundig 5010/5011/6010/6011 series. Td1 is the regulating transductor which is driven by Tr506. Ty511 is the flyback thyristor (commutating thyristor might be a better name), Ty518 the scan thyristor, Di518 the efficiency diode and C516/7/8 the tuning capacitance. The scan coils are cou- pled via C537, while C532 provides coupling between the primary winding of the line output transformer and chas- sis. A transductor (Td2) is used for EW raster correction. The combi coil also feeds 1.t. rectifiers from its secondary windings. 

Component Problems: The only problem with this type of circuit is the large amount of energy that shuttles back and forth at line frequency. This places a heavy stress on certain components. Circuit losses produce quite high temperatures, which are concentrated at certain points, in particular the combi coil. This leads to deterioration of the soldered joints around the coil, a common cause of failure. This can have a cumulative effect, a high -resistance joint increasing the local heating until the joint becomes well and truly dry -a classic symptom with some Grundig sets. The wound components themselves can be a source of trouble, due to losses - particularly the combi coil and the regulating transductor. Later chassis are less prone to this sort of thing, partly because of the use of later generation, higher efficiency yokes but mainly due to more generous and better design of the wound components. The ideal dielectric for use in the tuning capacitors is polypropylene (either metalised or film). It's a truly won- derful dielectric - very stable, with very small losses, and capable of operation at high frequencies and elevated temperatures. It's also nowadays reasonably inexpensive. Unfortunately many earlier chassis of this type used polyester capacitors, and it's no surprise that they were inclined to give up. When replacing the tuning capacitors in a thyristor line output stage it's essential to use poly- propylene types -a good range of axial components with values ranging from 0.001µF to 047µF is available from RS Components, enabling even non-standard values to be made up from an appropriate combination. Using polypropylene capacitors in place of polyester ones will not only ensure capacitor reliability but will also lower the stress on other components by reducing the circuit losses (and hence power consumption). The thyristors are also liable to fail, as are their parallel diodes. Earlier devices were less reliable than their successors. Since all thyristor line output stages operate in the same way and under similar conditions, the use of later types of thyristors and diodes in earlier circuits is a matter of mechanical rather than electrical con- siderations. One important point should be noted: the scan thyristor is a faster device and often has a higher voltage rating than the flyback thyristor. The simplest course is to keep in stock some of the later scan thyristors that incorporate an efficiency diode - suitable types are the RCA S3900SF and the Telefunken TD3-800H. The Telefunken device is in a TO66 package (and can be obtained quite cheaply) while the RCA type is in a TO220 package. Either type can be used in the scan or flyback positions and can also be used as a replacement for the regulating thyristor used in later designs instead of a transductor. Whenever replacing a thyristor in the line output stage it's good practice to replace the parallel diode at the same time. Using one of the above recom- mended devices will do this automatically, since the thyristor and its parallel diode share the same encapsulation - always remember to remove the old diode when this is a separate device however, as some can exhibit high -voltage leakage/breakdown which is not evident from a quite check with the Avo. Apart from the wound components (including the line output transformer), the thyristors and their parallel diodes and the tuning capacitors several other com- ponents are prone to failure. These include the tripler, scan/flyback rectifier diodes used to provide various supply lines, surge limiting resistors, the scan coil coup- ling/scan correction capacitor (replace with a metalised polypropylene type) and regulator components such as the thyristor in later types and the transductor driver transistor in earlier circuits. 

Basic Fault Conditions: At one time every engineer must have scratched his head and cursed the new-fangled idea of the thyristor line output stage. That they are awkward to service is a fallacy however. The usual symptom of a fault in the line output stage is the cutout tripping. All chassis that use a thyristor line timebase incorporate a trip of some sort. The type varies from chassis to chassis. Early Grundig sets have a mechanical cutout; the Saba H chassis uses a thyristor and solenoid to open the mains on/off switch; a common arrangement consists of a thyristor in series with the h.t, line and a control transistor which shorts the thyristor's gate and cathode in the event of excessive current demand (this gives audible tripping at about 2Hz). Some sets incorporate both excess current and over -voltage trips, but most have just the former. 

There are two basic fault conditions: when the excess current trip is activated and the set goes dead, or no e.h.t. with the trip not activated. The first condition is usually due to a line timebase fault, the most common being a short-circuit flyback thyristor or its parallel diode. A straightforward resistance check will sort this out. If this is not the case, short-circuit the scan thyristor by soldering a wire link between its anode and cathode. This will prevent any drive to the scan coils and the line output transformer. If the tripping stops, the fault could be due to the tripler, the line output transformer, a rectifier diode fed from a winding on the latter or a short in a circuit supplied by a scan rectifier diode. If the trip continues to operate and the flyback thyristor/diode is not the culprit, the most likely causes are incorrect drive to this thyristor - if possible check with a scope against the waveform given in the manual - or a rectifier diode fed from the combi coil. As an example of the latter, Fig. 3 shows the arrangement used in the Finlux Peacock: the electronic trip will operate if either D503 or D504 goes short-circuit, a fairly common fault on these sets. The diodes can also go open-circuit/high resistance to give the no sound with field collapse symp- tom, but that's another story ( referring to the diodes as D603/4 ). When the set is dead, h.t. is present and the trip is not activated, suspect the following: the scan thyristor, the efficiency diode, the line output transformer, the scan - correction capacitor, or lack of drive to the scan thyristor. Dry -joints can be the cause of any of these basic fault conditions, depending on the actual circuit and where the dry -joint has occurred. 

Other Symptoms: Hairline cracks in the ferrite core of a wound com- ponent can give rise to strange symptoms since this upsets the delicate balance of the tuning arrangements. There will usually be excessive current which will probably cause the trip to operate. Alternatively the fault may be incorrect line frequency which cannot be set by the line hold control. This fault can also give rise to excessive e.h.t., which can in turn produce a chain reaction of des- truction, e.g. the tripler is a common victim as are the two line output stage thyristors. Excessive e.h.t. leading to instant destruction of these components may also be due to open -circuit line scan coils or the connections to them. A quick resistance check done on the board itself will eliminate both the coils and the leads/connectors. Excessive e.h.t. with foldover in the centre of the screen and cooking in the tube's first anode supply net- work occurs in the Grundig 5010 series when L515 in the scan thyristor's trigger circuit (see Fig. 2) goes short- circuit. The reason for this situation is that the thyristor is triggered on early. Another common fault in these sets is failure of Di504/R504 - failure of one seems to affect the other, so both should be replaced. The usual symptom is fuzzy verticals and a sawtooth effect on diagonals. The trip may operate, possibly after period of operation. These components set up the transductor's operating bias. Linearity problems are usually caused by the regulator circuit, which can also be responsible for line "hunting". In the event of lack of width in the earlier type of circuit, check for dry -joints in the regulator circuit and suspect the control transistor. Foldover on the left-hand side of the screen can be caused by an open -circuit flyback diode. Foldover at the centre of the screen with greatly reduced width is the symptom when the efficiency diode goes open -circuit - the trip may or may not operate. Unusual interference patterns on the screen, best viewed with the contrast control turned to minimum and the brightness control advanced until a distinctly visible but not over bright white raster is obtained, can be due to the tripler if there's curved patterning on the extreme left- hand side of the screen, the regulator clamp diode (Di505 in Fig. 2) if there's curved interference just to the left of centre, or the flyback thyristor drive circuit if there's a single vertical line of patterning about four fifths of the way to the right of the screen.

The aim of this article has been to provide a general guide to servicing rather than to list faults common to particular models. Much useful information on individual 
chassis with thyristor line output stages has appeared in previous issues of  Obsolete Technology Tellye !- refer to the following as required: Search with the tag Thyristors at the bottom of the post to select all posts with this argument on various fabricants.

Electron beam deflection circuit including thyristors Further Discussion and deepening of knowledge, Thyristor horizontal output circuits:

1. An electron beam deflection circuit for a cathode ray tube with electromagnetic deflection by means of a sawtooth current waveform having a trace portion and a retrace portion, said circuit comprising: a deflection winding; a first source of electrical energy formed by a first capacitor; first controllable switching means comprising a parallel combination of a first thyristor and a first diode connected together to conduct in opposite directions, for connecting said winding to said first source during said trace portion, while said first switching means is turned on; a second source of electrical energy including a first inductive energy storage means coupled to a voltage supply; reactive circuit means including a combination of inductive and capacitive reactances for storing the energy supplied by said second source; second controllable switching means, substantially similar to said first one, for completing a circuit including said reactive circuit means and said first switching means, when turned on before the end of said trace portion, so as to pass through said first switching means an oscillatory current in opposite direction to that which passes through said first thyristor from said first source and to turn said first thyristor off after these two currents cancel out, the oscillatory current flowing thereafter through said first diode for an interval termed the circuit turn-off time, which has to be greater than the turn-off time of said first thyristor; wherein the improvement comprises: means for drawing, during at least a part of said trace portion, a substantial amount of additional current through said first switching means, in the direction of conduction of said first diode, whereby to perceptibly shift the waveform of the current flowing through said first switching means towards the negative values by an amount equal to that of said substantial additional current and to lengthen, in proportion thereto, said circuit turn-off time, without altering the values of the reactances in the reactive circuit which intervene in the determination of both the circuit turn-off and retrace portion time intervals.

2. A deflection circuit as claimed in claim 1, wherein said amount of additional current is greater than or equal to 5 per cent of the peak-to-peak value of the current flowing through the deflection winding.

3. A deflection circuit as claimed in claim 1, wherein said means for drawing a substantial amount of additional current through said first switching means comprises a resistor connected in parallel to said first capacitor.

4. A deflection circuit as claimed in claim 1, wherein said means for drawing an additional current is formed by connecting said first and second energy sources in series so that the current charging said reactive circuit means forms the said additional current.

5. A deflection circuit as claimed in claim 1, further including a series combination of an autotransformer winding and a second high-value capacitor, said combination being connected in parallel to said first switching means, wherein said autotransformer comprises an intermediate tap located between its terminals respectively connected to said first switching means and to said second capacitor, said tap delivering, during said trace portion, a suitable DC supply voltage lower than the voltage across said second capacitor; and wherein said means for drawing a substantial amount of additional current comprises a load to be fed by said supply voltage and having one terminal connected to ground; and further controllable switching means controlled to conduct during at least part of said trace portion and to remain cut off during said retrace portion, said further switching means being connected between said tap and the other terminal of said load.

Description:

The present invention relates to electron beam deflection circuits including thyristors, such as silicon controlled rectifiers and relates, in particular, to horizontal deflection circuits for television receivers.

The present invention constitutes an improvement in the circuit described in U.S. Pat. No. 3,449,623 filed on Sept. 6, 1966, this circuit being described in greater detail below with reference to FIGS. 1 and 2 of the accompanying drawings. A deflection circuit of this type comprises a first thyristor switch which allows the conenction of the horizontal deflection winding to a constant voltage source during the time interval used for the transmisstion of the picture signal and for applying this signal to the grid of the cathode ray tube (this interval will be termed the "trace portion" of the scan), and a second thyristor switch which provides the forced commutation of the first one by applying to it a reverse current of equal amplitude to that which passes through it from the said voltage source and thus to initiate the retrace during the horizontal blanking interval.

A undirectional reverse blocking triode type thyristor or silicon controlled rectifier (SCR), such as that used in the aformentioned circuit, requires a certain turn-off time between the instant at which the anode current ceases and the instant at which a positive bias may be applied to it without turning it on, due to the fact that there is still a high concentration of free carriers in the vicinity of the middle junction, this concentration being reduced by a process of recombination independently from the reverse polarity applied to the thyristor. This turn-off time of the thyristor is a function of a number of parameters such as the junction temperature, the DC current level, the decay time of the direct current, the peak level of the reverse current applied, the amplitude of the reverse anode to cathode voltage, the external impedance of the gate electrode, and so on, certain of these varying considerably from one thyristor to another.

In horizontal deflection circuits for television receivers, the flyback or retrace time is limited to approximately 20 percent of the horizontal scan period, the retrace time being in the case of the CCIR standard of 625 lines, approximately 12 microseconds and, in the case of the French standard of 819 lines, approximately 9 microseconds. During this relatively short interval, the thyristor has to be rendered non-conducting and the electron beam has to be returned to the origin of the scan. The first thyristor is blocked by means of a series resonant LC circuit which is subject to a certain number of restrictions (limitations as to the component values employed) due to the fact that, inter alia, it simultaneously determines the turn-off time of the circuit which blocks the thyristor and it forms part of the series resonant circuit which is to carry out the retrace. To obtain proper operation of the deflection circuit of the aforementioned Patent, especially when used for the French standard of 819 lines per image, the values of the components used have to subject to very close tolerances (approximately 2%), which results in high costs.

The improved deflection circuit, object of the present invention, allows the lengthening of the turn-off time of the circuit for turning the scan thyristor off, without altering the values of the LC circuit, which are determined by other criteria, and without impairing the operation of the circuit.

According to the invention, there is provided an electron beam deflection circuit for a cathode ray tube with electromagentic deflection by means of a sawtooth current waveform having a trace portion and a retrace portion, said circuit comprising: a deflection winding; a first source of electrical energy formed by a first capacitor; first controllable switching means comprising a parallel combination of a first thyristor and a first diode, connected together to conduct in opposite directions, for connecting said winding to said first source during said trace portion when said first switching means is turned on; a second source of electrical energy including a first inductive energy storage means coupled to a voltage supply; reactive circuit means including a combination of inductive and capacitive reactances for storing the energy supplied by the said second source; a second controllable switching means, substantially identical with the first one, for completing a circuit including said reactive circuit means and said first switching means, when turned on, so as to pass through said first thyristor an oscillatory current in the opposite direction to that which passes through it from said first source and to turn it off after these two currents cancel out, the oscillatory current then flowing through said first diode for an interval termed the circuit turn-off time which has to be greater than the turn-off time of said first thyristor; and means for drawing duing at least a part of said trace portion a substantial amount of additional current from said first switching means in the direction of conduction of said first diode, whereby said circuit turn-off time is lengthened in proportion to the amount of said additional current, without altering the values of the reactances in the reactive circuit by shifting the waveform of the current flowing through said first switching means towards the negative by an amount equal to that of said additional current.

A further object of the invention consists in using the supplementary current in the recovery diode of the first switching means to produce a DC voltage which may be used as a power supply for the vertical deflection circuit of the television receiver, for example.

The invention will be better understood and other features and advantages thereof will become apparent from the following description and the accompanying drawings, given by way of example, and in which:

FIG. 1 is a schematic circuit diagram partially in bloc diagram form of a prior art deflection circuit according to the aforementioned Patent;

FIG. 2 shows waveforms of currents and voltages generated at various points in the circuit of FIG. 1;

FIG. 3 is a schematic diagram of a deflection circuit according to the invention which allows the principle of the improvement to be explained;

FIG. 4 is a diagram of the waveforms of the current through the first switching means 4, 5 of the circuit of FIG. 3;

FIG. 5 is a circuit diagram of another embodiment of the circuit according to the invention;

FIG. 6 is a schematic representation of the preferred embodiment of the circuit according to the invention; and

FIG. 7 shows voltage waveforms at various points of the high voltage autotransformer 21 of FIG. 6.

In all these Figures the same reference numerals refer to the same components.

FIG. 1 shows the horizontal deflection circuit described and claimed in the U.S. Pat. No. 3,449,623 mentioned above, which comprises a first source of electrical energy in the shape of a first capacitor 2 having a high capacitance C ₂ for supplying a substantially constant voltage Uc ₂ across its terminals. A first terminal of the first capacitor 2 is connected to ground, whilst its second terminal which supplies a positive voltage is connected to one of the terminals of a horizontal deflection winding shown as a first inductance 1. A first switching means 3, consisting of a first reverse blocking triode thyristor 4 (SCR) and a first recovery diode 5 in parallel, the two being interconnected to conduct current in opposite directions, is connected in parallel with the series combination formed by the deflection winding 1 and the first capacitor 2. The assembly of components 1, 2, 4 and 5 forms the final stage of the horizontal deflection circuit in a television receiver using electromagnetic delfection.

The deflection circuit also includes a drive stage for this final stage which here controls the turning off of the first thyristor 4 to produce the retrace or fly-back portion of the scan during the line-blanking intervals i.e. while the picture signal is not transmitted. This driver stage comprises a second voltage source in the shape of a DC power supply 6 which delivers a constant high voltage E. The negative terminal of the power supply 6 is connected to ground and its positive terminal to one of the terminals of a second inductance 7 of relatively high value, which draws a substantially lineraly varying current from the power supply 6 to avoid its overloading. The other terminal of the second inductance 7 is connected, on the one hand, to the junction of the deflection winding 1 and the first switching means 3 by means of a second inductance 8 and a second capacitor 9 in series and, on the other hand, to one of the terminals of a second controllable bi-directionally conducting switching means 10, similar to the first one 3, including a parallel combination of a second thyristor 11 and a second recovery diode 12 also arranged to conduct in opposite directions.

The respective values of the third inductance 8 (L ₈ ) and of the second capacitor 9 (C ₉ ) are principally selected so that, on the one hand, one half-cycle of oscillation of the first series resonant circuit L ₈ - C ₉ , (i.e. π √ L ₈ ^. C ₉ ) is longer than the turn-off time of the first thyristor 4, but still is as short as possible since this time interval determines the speed of the commutation of the thyristor 4, and, on the other hand, one half-cycle of oscillation of another series resonant circuit formed by L ₁ , L ₈ and C ₉ , i.e. π √ (L ₁ + L ₈ ) ^. C ₉ , is substantially equal to the required retrace time interval (i.e. shorter than the horizontal blanking interval).

The gate (control electrode) of the second thyristor 11 is coupled to the output of the horizontal oscillator 13 of the television receiver by means of a first pulse transformer 14 and a first pulse shaping circuit 15 so that it is fed short triggering pulses which are to turn it on.

The gate of the first thyristor 4 fed with signals of a substantially rectangular waveform which are negative during the horizontal blanking intervals, is coupled to a winding 16 by means of a second pulse shaping circuit 17, the winding 16 being magnetically coupled to the second inductance 7 to make up the secondary winding of a transformer of which the inductance 7 forms the primary winding. It will be noted here that it is also possible to couple the secondary winding 16 magnetically to a primary winding connected to a suitable output (not shown) of the horizontal oscillator 13.

The operation of a circuit of this type will be explained below with reference to FIG. 2 which shows the waveforms at various points in the circuit of FIG. 1 during approximately one line period.

FIG. 2 is not to scale since one line period (t ₇ - t ₀ ) is equal to 64 microseconds in the case of 625 lines and 49 microseconds in the case of 819 lines, while the durations of the respective horizontal blanking intervals are approximately 12 and 9.5 microseconds.

Waveform A shows the form of the current i _L1 passing through deflection winding 1, this current having a sawtooth waveform substantially linear from t ₀ to t ₃ and from t ₅ to t ₇ , and crossing zero at time instants t ₀ and t ₇ , and reaching values of + I _1m and - I _1m , at time instants t ₃ and t ₅ respectively, these being its maximum positive and negative amplitudes.

During the second half of the trace portion of the horizontal deflection cycle, that is to say from t ₀ to t ₃ , the thyristor 4 of the first switching means 3 is conductive and makes the high value capacitor 2 discharge through the deflector winding 1, which has a high inductance, so that current i _L1 increases linearly.

A few microseconds (5 to 8 μ s) before the end of the trace portion, i.e. at time instant t ₁ , the trigger of the second thyristor 11 receives a short voltage pulse V _G11 which causes it to turn on as its anode is at this instant at a positive potential with respect to ground, which is due to the charging of the second capacitor 9 through inductances 7 and 8 by the voltage E from the power supply 6.

When thyristor 11 is made conductive at time t ₁ , on the one hand, inductance 7 is connected between ground and the voltage source 6 and a linearly increasing current flows through it and, on the other hand, the reactive circuit 8, 9 forms a loop through the second and first switching means 10 and 3, thus forming a resonant circuit which draws an oscillatory current i ₈ ,9 of frequency ##EQU1##

This oscillatory current i ₈ ,9 will pass through the first switching means 3, i.e. thyristor 4 and diode 5, in the opposite direction to that of current i _L1 . Since the frequency f ₁ is high, current i ₈ ,9 will increase more rapidly than i _L1 and will reach the same level at time t ₂ , that is to say i ₈ ,9 (t ₂ ) = -i _L1 (t ₂ ) and these currents will cancel out in the thyristor 4 in accordance with the well known principle of forced commutation. After time instant t ₂ , current i ₈ ,9 continues to increase more rapidly than i _L1 , but the difference between them (i ₈ ,9 - i _L1 ) passes the diode 5 (see wave form B) until it becomes zero at time instant t ₃ which is the turn off time instant of the first switching means 3, at which the retrace begins.

The interval between the time instant t ₂ and t ₃ , i.e. (t ₃ -t ₂ ), during which diode 5 is conductive and the thyristor is reverse biased will be termed in what follows the circuit turn-off time and it should be greater than the turn-off time of the thyristor 4 itself since the latter will subsequently become foward biased (i.e. from t ₃ to t ₅ ) by the retrace or flyback pulse (see waveform E) which should not trigger it.

At time instant t ₃ , the switching means 3 is opened (i ₄ and i ₅ are both zero -- see waveforms B and C) and the reactive circuit 8, 9 forms a loop through capacitor 2 and the deflection coil 1 and thus a series resonant circuit including (L ₁ + L ₈ ) and C ₉ , C ₂ being of high value and representing a short circuit for the flyback frequency ##EQU2## thus obtained.

The retrace which stated at time t ₃ takes place during one half-cycle of the resonant circuit formed by reactances L ₁ , L ₈ and C ₉ , i.e. during the interval between t ₃ and t ₅ . In the middle of this interval i.e. at time instant t ₄ , both i _L1 (waveform A) and i ₈ ,9 (waveform D) pass through zero and change their sign, whereas the voltage at the terminals of the first switching means 3 (V ₃ , waveform E) passes through a maximum. Thus, from t ₄ onwards, thyristor 11 will be reverse biased and diode 12 will conduct the current from the resonant circuit 1, 8 and 9 in order to turn the second thyristor 11 off.

At time instant t ₅ , when current i _L1 has reached - I _1m and when voltage v ₃ falls to zero, diode 5 of the first switching means 3 becomes conductive and the trace portion of scan begins.

Current i ₈ ,9 nevertheless continues to flow in the resonant circuit 8, 9 through diodes 5 and 12, which causes a break to appear in waveform D at t ₅ , and a negative peak to appear in waveform D and a positive one in waveform B in the interval between t ₅ and t ₆ , these being principally due to the distributed capacities of coil 1 or to an eventual capacitor (not shown) connected in parallel to the first switching means 3.

At time instant t ₆ , diode 12 of the second switching means 10 ceases to conduct after having allowed thyristor 11 time to become turned off completely.

The level of current i ₈ ,9 at time instant t ₅ (i.e. I _c ) as well as the negative peak I _D12 in i ₈ ,9 and the positive peak I _D5 in i ₅ depend on the values of L ₈ and C ₉ in the same way as does the turn-off time of the circuit (t ₃ - t ₂ ). If, for example, L ₈ and C ₉ , are increased I _D5 increases towards zero and this could cause diode 5 to be cut off in an undesirable fashion. I _c also increases towards zero, which is liable to cause diode 12 to be blocked and thyristor 11 to trigger prematurely.

From the foregoing it can be clearly seen that the choice of values for L ₈ and C ₉ is subject to four limitations which prevent the values from being increased to lengthen the turn-off time of the driver circuit of first switching thyristor 4 so as to forestall its spurious triggering.

Waveform F shows the voltage v _G4 obtained at the gate of thyristor 4 from the secondary winding 16 coupled to the inductor 7. This voltage is positive from t ₀ to t ₁ and from t ₆ to t ₇ and is negative between t ₂ and t ₆ i.e. while the second switching means 10 is conducting.

The present invention makes the lengthening of the turn-off time of thyristor 4 possible without altering the parameters of the circuit such as inductance 8 and capacitor 9.

In the circuit shown in FIG. 3, which illustrates the principle of the present invention, means are added to the circuit in FIG. 1 which enable the turn-off time to be lengthened by connecting a load to diode 5 so as to increase the current which flows through it during the time that it is conductive. These means are here formed by a resistor 18 connected in parallel with a capacitor 20 (which replaces capacitor 2) which is of a higher capacitance so that, in practice, it holds its charge during at least one half of the line period. FIG. 4, which shows the waveform of the current in the first switching means 3 for a circuit as shown in FIG. 3, makes it possible to explain how this lenthening of the turn-off time is achieved.

In FIG. 4, the broken lines show the waveform of the current in the first switch device 3 in the circuit of FIG. 1, this waveform being produced by adding waveforms B and C of FIG. 2. The current i ₄ above the axis flows through thyristor 4 and current i ₅ below the axis flows through diode 5. When the capacitance C ₂₀ of the capacitor in series with the deflector coil is increased to some tens of microfarads (C ₂ having been of the order of 1 μ F) and when there is connected in parallel with capacitor 20 a resistor 18 the value of which is calculated to draw a strong current I _R18 from capacitor 20, that is to say a current at least equal to 0,1 I _m (I _m being of the order of some tens of amperes), current I _R18 is added to that i ₅ which flows through diode 5 without in any way altering the linearity of the trace portion nor the oscillatory commutation of thyristor 4 which is brought about by the resonant circuit L ₈ , C ₉ .

The fact of loading capacitor C ₂₀ by means of a resistor 18 thus has the effect of permanently displacing the waveform of the current in the negative direction by I _R18 . Thus, during the trace portion of the scan, the transfer of the current from the diode 5 to the thyristor 4 begins at time t ₁₀ instead of t ₀ , that is to say with a delay proportional to I _R18 . The effect of the triggering pulse delivered by the horizontal oscillator (13 FIG. 1) to the second thyristor 11 at time instant t ₁ , will be to start the commutation process of the first thyristor 4 when the current it draws is less by I _R18 than that i ₄ (t ₁ ) which it would have been drawing had there been no resistor 18. Because of this, the turn-off time of the thyristor 4 proper, which as has been mentioned increases with the maximum current level passing throught it, is slightly reduced. Moreover, because the oscillatory current i ₈ ,9 (FIG. 2) from circuit L ₈ , C ₉ which flows through thyristor 4 in the opposite direction is unchanged, it reaches a value equal to that of the current i _L1 (FIG. 1) flowing in the coil 1 in a shorter time, that is to say at time t ₁₂ . Diode 5 will thus take the oscillatory current i ₈ ,9 (FIG. 2) over in advance with respect ro time instant t ₂ and will conduct it until it reaches zero value at a time instant t ₁₃ later than t ₃ , the amounts of advance (t ₂ - t ₁₂ ) and delay (t ₁₃ - t ₃ ) being practically equal.

It can thus be seen in FIG. 4 that the circuit turn-off time T _R of a circuit according to the invention and illustrated by FIG. 3 is distinctly longer than that T _r of the circuit in FIG. 1. This increase in the turn-off time (T _R - T _r ) depends on the current I _R18 and increases therewith.

It should be noted at this point that the current I _R18 produces a voltage drop at the terminals of the resistor the only effect of which is to heat up the resistor since the level of this voltage (40 to 60 volts) does not necessarily have a suitable value to be used as a voltage supply for other circuits in an existing transistorised television receiver.

In accordance with one embodiment of the invention, illustrated in FIG. 5, an application is proposed for the additional current which is to be drawn through diode 5. In FIG. 5, the positive terminal of capacitor 20 is connected by a conductor 19 to the negative pole of the power supply 6 and the voltage at the terminals of capacitor 20 is thus added to that E from the source 6.

In the preferred embodiment of the present invention, which is shown in FIG. 6, it is possible to cause a supplementary current of a desired value to flow through the first diode 5 while obtaining a voltage which has a suitable value for use in another circuit in the television receiver.

If the voltage at the terminals of capacitor 20 in FIG. 3 is not a usable value, it is possible to connect in parallel with the series circuit comprising the deflector coil 1 and the capacitor 2 in FIG. 1, i.e. in parallel with the terminals of the first switching means 3, a series combination of an autotransformer 21 and a high value capacitor 22 (comparable with capacitor 20 in FIGS. 3 and 5). The autotransformer 21 has a tap 23 is suitably positioned between the terminal connected to capacitor 22 at the tap 24 connected to the first switching means 3. This autotransformer 21 may be formed by the one conventionally used for supplying a very high voltage to the cathode ray tube, as described for example in U.S. Pat. No. 3,452,244; such a transformer comprises a voltage step-up winding between taps 24 and 25, which latter is connected to a high voltage rectifier (not shown).

The waveform of the voltage at the various points in the autotransformer is shown in FIG. 7, in which waveform A shows the voltage at the terminals of capacitor 22, waveform B the voltage at tap 24 and waveform C the voltage at tap 23 of the autotransformer 21.

The voltage V _c22 at the terminals of capacitor 22 varies slightly about a mean value V _cm . It is increasing while diode 5 is conducting and decreasing during the conduction of the thyristor 4.

The voltage v ₂₄ at tap 24 follows substantially the same curve as waveform E in FIG. 2, that is to say that during the retrace time interval from t ₁₃ to t ₅ to a positive pulse called the flyback pulse is produced and, during the time interval while the first switching means 3 is conducting, the voltage is zero. The mean valve of the voltage v ₂₄ at tap 24 of the auto-transformer 21 is equal to the mean value V _cm of the voltage at the terminals of capacitors 2 and 22.

Thus, there is obtained at tap 23 a waveform which is made up, during the retrace portion, of a positive pulse whose maximum amplitude is less than that of v ₂₄ at tap 24 and, during the trace portion, of a substantially constant positive voltage, the level V of which is less than the mean value V _cm of the voltage v _c22 at the terminals of capacitor 22. By moving tap 23 towards terminals 24 the amplitude of the pulse during fly-back increases while voltage V falls and conversely by moving tap 23 towards capacitor 22 voltage V increases and the amplitude of the pulse drops.

In more exact terms, the voltage V at tap 23 is such that the means value of v ₂₃ is equal to V _cm . It has thus been shown that by choosing carefully the position of tape 23, a voltage V may be obtained during the trace portion of the scan, which may be of any value between V _cm and zero.

This voltage V is thus obtained by periodically controlled rectification during the trace portion of the scan. For this purpose an electronic switch is used to periodically connect the tap 23 of trnasformer winding 21 to a load. This switch is made up of a power transistor 26 whose collector is connected to tap 23 and the emitter to a parallel combination formed by a high value filtering capacitor 27 and the load which it is desired to supply, which is represented by a resistor 28. The base of the transistor 26 receives a control voltage to block it during retrace and to unblock it during the whole or part of the trace period. A control voltage of this type may be obtained from a second winding 29 magnetically coupled to the inductance 7 of the deflection circuit and it may be transmitted to the base of transistor 26 by means of a coupling capacitor 30 and a resistor 31 connected between the base and the emitter of transistor 26.

It may easily be seen that the DC collector/emitter current in transistor 26 flows through the first diode 5 of the first switching means 3 via a resistor 28 and the part of the winding of auto-transformer 21 located between taps 23 and 24.

Experience has shown that a circuit as shown in FIG. 6 can supply 24 volts with a current of 2 amperes to the vertical deflection circuit of the same television set, the voltage at the terminals of capacitor 22 being from 50 to 60 volts.

It should be mentioned that, when the circuit which forms the load of the controlled rectifier 26, 27 does not draw enough current to sufficiently lengthen the circuit turn-off time T _R , an additional resistor (not shown) may be connected between the emitter of transistor 26 and ground or in parallel to capacitor 22, which resistor will draw the additional current required.

INTEGRAL THYRISTOR-RECTIFIER DEVICEA semiconductor switching device comprising a silicon controlled rectifier (SCR) and a diode rectifier integrally connected in parallel with the SCR in a single semiconductor body. The device is of the NPNP or PNPN type, having gate, cathode, and anode electrodes. A portion of each intermediate N and P region makes ohmic contact to the respective anode or cathode electrode of the SCR. In addition, each intermediate region includes a highly conductive edge portion. These portions are spaced from the adjacent external regions by relatively low conductive portions, and limit the conduction of the diode rectifier to the periphery of the device. A profile of gold recombination centers further electrically isolates the central SCR portion from the peripheral diode portion.
That class of thyristors known as controlled rectifiers are semiconductor switches having four semiconducting regions of alternate conductivity and which employ anode, cathode, and gate electrodes. These devices are usually fabricated from silicon. In its normal state, the silicon controlled rectifier (SCR) is non-conductive until an appropriate voltage or current pulse is applied to the gate electrode, at which point current flows from the anode to the cathode and delivers power to a load circuit. If the SCR is reverse biased, it is non-conductive, and cannot be turned on by a gating signal. Once conduction starts, the gate loses control and current flows from the anode to the cathode until it drops below a certain value (called the holding current), at which point the SCR turns off and the gate electrode regains control. The SCR is thus a solid state device capable of performing the circuit function of a thyratron tube in many electronic applications. In some of these applications, such as in automobile ignition systems and horizontal deflection circuits in television receivers, it is necessary to connect a separate rectifier diode in parallel with the SCR. See, for example, W. Dietz, U. S. Pat. Nos. 3,452,244 and 3,449,623. In these applications, the anode of the rectifier diode is connected to the cathode of the SCR, and the cathode of the rectifier is connected to the SCR anode. Thus, the rectifier diode will be forward biased and current will flow through it when the SCR is reverse biased; i.e., when the SCR cathode is positive with respect to its anode. For reasons of economy and ease of handling, it would be preferable if the circuit function of the SCR and the associated diode rectifier could be combined in a single device, so that instead of requiring two devices and five electrical connections, one device and three electrical connections are all that would be necessary. In fact, because of the semiconductor profile employed, many SCR's of the shorted emitter variety inherently function as a diode rectifier when reverse biased. However, the diode rectifier function of such devices is not isolated from the controlled rectifier portion, thus preventing a rapid transition from one function to the other. Therefore, it would be desirable to physically and electrically isolate the diode rectifier portion from that portion of the device which functions as an SCR.

GRUNDIG SUPER COLOR 6236/30 SERIE 30F22 CHASSIS 29301-114.68(07)
Gating circuit for television SCR deflection system AND REGULATION / stabilization of horizontal deflection NETWORK CIRCUIT with Transductor reactor / Reverse thyristor energy recovery circuit.
In a television deflection system employing a first SCR for coupling a deflection winding across a source of energy during a trace interval of each deflection cycle and a second SCR for replenishing energy to the source of energy during a commutation interval of each deflection cycle, a gating circuit for triggering the first SCR. The gating circuit employs a voltage divider coupled in parallel with the second SCR which develops gating signals proportional to the voltage across the second SCR.


1. In a television deflection system in which a first switching means couples a deflection winding across a source of energy during a trace interval of each deflection cycle and a second switching means replenishes energy to said source of energy during a commutation interval of each deflection cycle, a gating circuit for said first switching means, comprising:
capacitive voltage divider means coupled in parallel with said second switching means for developing gating signals proportional to the voltage across said second switching means; and
means for coupling said voltage divider means to said first switching means to provide for conduction of said first switching means in response to said gating signals.
2. A gating circuit according to claim 1 wherein said voltage divider includes first and second capacitors coupled in series and providing said gating signals at the common terminal of said capacitors. 3. A gating circuit according to claim 2 wherein said first and second capacitors are proportional in value to provide for the desired magnitude of gating signals. 4. A gating circuit according to claim 3 wherein said means for coupling said voltage divider means to said first switching means includes an inductor. 5. A gating circuit according to claim 4 wherein said inductor and said first and second capacitors comprise a resonant circuit having a resonant frequency chosen to shape said gating signal to improve switching of said first switching means.

Description:

BACKGROUND OF THE INVENTION
This invention relates to a gating circuit for controlling a switching device employed in a deflection circuit of a television receiver.

Various deflection system designs have been utilized in television receivers. One design employing two bidirectional conducting switches and utilizing SCR's (thyristors) as part of the switches is disclosed in U.S. Pat. No. 3,452,244. In this type deflection system, a first SCR is

employed for coupling a deflection winding across a source of energy during a trace interval of each deflection cycle, and a second SCR is employed for replenishing energy during a commutation interval of each deflection cycle. The first SCR is commonly provided with gating voltage by means of a separate winding or tap of an input reactor coupling a source of B+ to the second SCR.

Various regulator system designs have been utilized in conjunction with the afore described deflection system to provide for uniform high voltage production as well as uniform picture width with varying line voltage and kinescope beam current conditions.
One type regulator system design alters the amount of energy stored in a commutating capacitor coupled between the first and second SCR's during the commutating interval. A regulator design of this type may employ a regulating SCR and diode for coupling the input reactor to the source of B+. With this type regulator a notch, the width of which depends upon the regulation requirements, is created in the current supplied through the reactor and which notch shows up in the voltage waveform developed on the separate winding or tap of the input reactor which provides the gating voltage for the first SCR. The presence of the notch, even though de-emphasized by a waveshaping circuit coupling the gating voltage to the first SCR, causes erratic control of the first SCR.
SUMMARY OF THE INVENTION
In accordance with one embodiment of the invention, a gating circuit of a television deflection system employing a first switching means for coupling a deflection winding across a source of energy during a trace interval of each deflection cycle and a second switching means for replenishing energy to said source of energy during a commutation interval of each deflection cycle includes a voltage divider means coupled in parallel with the second switching means for developing gating signals proportional to the voltage across the second switching means. The voltage divider means are coupled to the first switching means to provide for conduction of the first switching means in response to the gating signals.
A more detailed description of a preferred embodiment of the invention is given in the following description and accompanying drawing of which:
FIG. 1 is a schematic diagram, partially in block form, of a prior art SCR deflection system;
FIG. 2 is a schematic diagram, partially in block form, of an SCR deflection system of the type shown in FIG. 1 including a gating circuit embodying the invention;
FIG. 3 is a schematic diagram, partially in block form, of one type of a regulator system which employs an SCR as a control device and which is suitable for use with the SCR deflection system of FIG.2;
FIG. 4 is a schematic diagram, partially in block form, of another type of a regulator system suitable for use with the deflection circuit of FIG. 2; and
FIG. 5 is a schematic diagram, partially in block form, of still another type of a regulator system suitable for use with the SCR deflection system of FIG. 2.
DESCRIPTION OF THE PREFERRED EMBODIMENT
FIG. 1 is a schematic diagram, partially in block form, of a prior art deflection system of the retrace driven type similar to that disclosed in U.S. Pat. No. 3,452,244. This system includes a commutating switch 12, comprising a silicon controlled rectifier (SCR) 14 and an oppositely poled damper diode 16. The commutating switch 12 is coupled between a winding 18a of an input choke 18 and ground. The other terminal of winding 18a is coupled to a source of direct current voltage (B+) by means of a regulator network 20 which controls the energy stored in the deflection circuit 10 when the commutating switch is off, during an interval T3 to T0' as shown in curve 21 which is a plot of the voltage level at the anode of SCR 14 during the deflection cycle. A damping network comprising a series combination of a resistor 22 and a capacitor 23 is coupled in parallel with commutating switch 12 and serves to reduce any ringing effects produced by the switching of commutating switch 12. Commutating switch 12 is coupled through a commutating coil 24, a commutating capacitor 25 and a trace switch 26 to ground. Trace switch 26 comprises an SCR 28 and an oppositely poled damper diode 30. An auxiliary capacitor 32 is coupled between the junction of coil 24 and capacitor 25 and ground. A series combination of a horizontal deflection winding 34 and an S-shaping capacitor 36 are coupled in parallel with trace switch 26. Also, a series combination of a primary winding 38a of a horizontal output transformer 38 and a DC blocking capacitor 40 are coupled in parallel with trace switch 26.
A secondary of high voltage winding 38b of transformer 38 produces relatively large amplitude flyback pulses during the retrace interval of each deflection cycle. This interval exists between T1 and T2 of curve 41 which is a plot of the current through windings 34 and 38a during the deflection cycle. These flyback pulses are applied to a high voltage multiplier (not shown) or other suitable means for producing direct current high voltage for use as the ultor voltage of a kinescope (not shown).
An auxiliary winding 38c of transformer 38 is coupled to a high voltage sensing and control circuit 42 which transforms the level of flyback pulses into a pulse width modulated signal. The control circuit 42 is coupled to the regulator network 20.
A horizontal oscillator 44 is coupled to the gate electrode of commutating SCR 14 and produces a pulse during each deflection cycle slightly before the end of the trace interval at T0 of curve 21 to turn on SCR 14 to initiate the commutating interval. The commutating interval occurs between T0 and T3 of curve 21. A resonant waveshaping network 46 comprising a series combination of a capacitor 48 and an inductor 50 coupled between a winding 18b of input choke 18 and the gate electrode of trace SCR 28 and a damping resistor 52 coupled between the junction of capacitor 48 and inductor 50 and ground shapes the signal developed at winding 18b (i.e. voltage waveform 53) to form a gating signal voltage waveform 55 to enable SCR 28 for conduction during the second half of the trace interval occurring between T2 and T1' of curve 41.
The regulator network 20, when of a type to be described in conjunction with FIG. 3, operates in such a manner that current through winding 18a of input choke 18 during an interval between T4 and T5 (region A) of curves 21, 53 and 55 is interrupted for a period of time the duration of which is determined by the signal produced by the high voltage sensing and control circuit 42. During the interruption of current through winding 18a a zero voltage level is developed by winding 18b as shown in interval T4 to T5 of curve 53. The resonant waveshaping circuit 46 produces the shaped waveform 55 which undesirably retains a slump in region A corresponding to the notch A of waveform 53. The slump in waveform 55 applied to SCR 28 occurs in a region where the anode of SCR 28 becomes positive and where SCR 28 must be switched on to maintain a uniform production of the current waveshape in the horizontal deflection winding 34 as shown in curve 41. The less positive amplitude current occurring at region A of waveform 55 may result in insufficient gating current for SCR 28 and may cause erratic performance resulting in an unsatisfactory raster.
FIG. 2 is a schematic diagram, partially in block form, of a deflection system 60 embodying the invention. Those elements which perform the same function in FIG. 2 as in FIG. 1 are labeled with the same reference numerals. FIG. 2 differs from FIG. 1 essentially in that the signal to enable SCR 28 derived from sampling a portion of the voltage across commutating switch 12 rather than a voltage developed by winding 18b which is a function of the voltage across winding 18a of input choke 18 as in FIG. 1. This change eliminates the slump in the enabling signal during the interval T4 to T5 as shown in curve 64 since the voltage across the commutating switch 12 is not adversely effected by the regulator network 20 operation.
A series combination of resistor 22, capacitor 23 and a capacitor 62 is coupled in parallel with commutating switch 12, one terminal of capacitor 62 being coupled to ground. The junction of capacitors 23 and 62 is coupled to the gate electrode of SCR 28 by means of the inductor 50. The resistor 52 is coupled in parallel with capacitor 62.
Capacitors 23 and 62 form a capacitance voltage divider which provides a suitable portion of the voltage across commutating switch 12 for gating SCR 28 via inductor 50. The magnitude of the voltage at the junction of capacitors 23 and 62 is typically 25 to 35 volts. It can, therefore, be seen that the ratio of values of capacitors 23 and 62 will vary depending on the B+ voltage utilized to energize the deflection system. Capacitors 23 and 62 and inductor 50 form a resonant circuit tuned in a manner which provides for peaking of the curve 64 between T4 and T5. This peaking effect further enhances gating of SCR 28 between T4 and T5.
Since the waveshape of the voltage across commutating switch 12 (curve 21) is relatively independent of the type of regulator system employed in conjunction with the deflection system, the curve 64 also is independent of the type of regulator system.
When commutating switch 12 switches off during the interval T3 to T0' curve 21, the voltage across capacitor 62 increases and the voltage at the gate electrode of SCR 28 increases as shown in curve 64. As will be noted, no slump of curve 64 occurs between T3 and T5 because there is no interruption of the voltage across commutating switch 12.



















FIG. 3 is a schematic diagram, partially in block form, of one type of a regulator system which may be used in conjunction with the invention. B+ is supplied through a regulator network 20 which comprises an SCR 66 and an oppositely poled diode 68. The diode is poled to provide for conduction of current from B+ to the horizontal deflection circuit 60 via winding 18a of input choke 18. Current flows through the diode during the period T3 to T4 of curve 21 FIG. 1 after which current tries to flow through the SCR 66 from the horizontal deflection circuit to B+ since the commutating capacitor 25 is charged to a voltage higher than B+.
The horizontal deflection circuit 60 produces a flyback pulse in winding 38a of the flyback transformer 38 which is coupled to winding 38c. The magnitude of the pulse on winding 38c determines how long the signal required to switch SCR 66 on is delayed after T4 curve 21 FIG. 1. If the flyback pulse is greater than desirable, the SCR 66 turns on sooner than if the flyback pulse is less than desirable and provides a discharge path for current in commutating capacitor 25 back to the B+ supply. In this manner a relatively constant amplitude flyback pulse is maintained.
FIG. 4 is a schematic diagram, partially in block form, of another well-known type of a regulator system which may be used in conjunction with the invention shown in FIG. 2. B+ is coupled through winding 18a of input choke 18 and through a series combination of windings 70a and 70b of a saturable reactor 70 and a parallel combination of a diode 72 and a resistor 74 to the horizontal deflection circuit 60. Diode 72 is poled to conduct current from the horizontal deflection circuit 60 to B+.
Flyback pulse variations are obtained from winding 38c of the horizontal output transformer 38 and applied to a voltage divider comprising resistors 76, 78 and 80 of the high voltage sensing and control circuit 42. A portion of the pulse produced by winding 38c is selected by the position of the wiper terminal on potentiometer 78 and coupled to the base electrode of a transistor 82 by means of a zener diode 84. The emitter electrode of transistor 82 is grounded and a DC stabilization resistor 85 is coupled in parallel with the base-emitter junction of transistor 82. When the pulse magnitude on winding 38c exceeds a level which results in forward biasing the base-emitter junction of transistor 82, current flows from B+ through a resistor 86, a winding 70c of saturable reactor 70 and transistor 82 to ground. Due to the exponential increase of current in winding 70c during the period of conduction of transistor 82, the duration of conduction of transistor 82 determines the magnitude of current flowing in winding 70c and thus the total inductance of windings 70a and 70b. The current in winding 70c is sustained during the remaining deflection period by means of a diode 88 coupled in parallel with winding 70c and poled not to conduct current from B+ to the collector electrode of transistor 82. A capacitor 90 coupled to the cathode of diode 88 provides a bypass for B+. Windings 70a and 70b are in parallel with input reactor 18a and thereby affect the total input inductance of the deflection circuit and thereby controls the transfer of energy to the deflection circuit. The dotted waveforms shown in conjunction with a curve 21' indicate variations from a nominal waveform provided at the input of horizontal deflection circuit 60 by the windings 70a and 70b.













FIG. 5 is a schematic diagram of yet another type of a regulator system which may be used in conjunction with the invention. B+ is coupled through a winding 92a and a winding 92b of a saturable reactor to the horizontal deflection circuit 60. Windings 92a and 92b are used to replace the input choke 18 shown in FIGS. 1 and 2 while also providing for a regulating function corresponding to that provided by regulating network 20.
Flyback pulse variations are obtained from winding 38c and applied to the high voltage sensing and control circuit 42 as in FIG. 4. Current flows from B+ through resistor 86, a winding 92c and transistor 82 to ground. As in FIG. 4 the duration of the conduction of transistor 82 determines the energy stored in winding 92c and thus the total inductance of windings 92a and 92b which control the amount of energy transferred to the deflection circuit during each horizontal deflection cycle. The variations in waveforms of curve 21', shown in conjunction with FIG. 4, are also provided at the input of horizontal deflection circuit 60 by windings 92a and 92b.
For various reasons including cost or performance, a manufacturer may wish to utilize a particular one of the regulators illustrated in FIGS. 3, 4 and 5. Regardless of the choice, the gating circuit according to the invention may be utilized therewith advantageously by providing improved performance and the possibility of cost savings by eliminating taps or extra windings on the wound components which heretofore normally provided a source of SCR gating waveforms.

THE TBA800, TBA810 AUDIO integrated circuits:

AUDIO integrated circuits are being increasingly used in television chassis and certainly represent the simplest approach to improving the audio side of a TV set. A number of such i.c.s have appeared during the 70's.
Here describes the use of two fairly recent ones, the SGS-ATES TBA800 and TBA8I0S. Both devices can provide reasonably high outputs into a suitable loudspeaker-the TBA800 will give up to 5W and the TBA810S up to 7W.

The main difference between them being that the TBA800 is a somewhat higher voltage, lower current device. The TBA800 is used in the current Grundig and ASA 110° colour chassis while the Finlux 110' colour chassis uses a TBA810. In each of these chassis the audio i.c. is driven from a TBA120 intercarrier sound i.c. The TBA800 and TBA810S can also be used as the field output stage in 110' monochrome chassis with c.r.t.s of up to l7in. and as the field driver stage in larger screen monochrome sets.
The TBA800 is designed to provide up to 5W into a 16 Ohm load when operated from a 24V supply. It is encapsulated in the type cf quad -in -line case shown in Fig. I: the tabs at the centre are to assist in cooling the device and must be earthed. The TBA800 can be operated from power supply voltages up to the absolute maximum permissible value of 30V. It is best to regard 24V as being the upper limit however in order to provide an adequate safety margin and prevent possible damage during voltage surges. The minimum power supply voltage recommended by the manufacturers is 5V, but the power output is then less than 0-5W. The quiescent current taken by the TBA800 is typically 9mA from a 24V supply-no device of this type should draw more than 20mA. When an input signal is applied the current increases considerably- up to about 1.5A at full power. Two circuits for use with the TBA800 are shown in Figs. 2 and 3 and give comparable performance. The circuit shown in Fig. 2 is somewhat simpler but that
shown in Fig. 3 enables one side of the loudspeaker to be connected to chassis. The input resistance of the TBA800 is quite high (typically 5 MOhm) but a resistor must be connected between the input pin 8 and chassis otherwise the out- put stage will not operate with the correct bias. In the circuits shown the volume control VR1 provides this function: the bias current that flows through it is typically 1 microA (maximum 5 microA). The average voltage at the output pin 12 is half the supply potential. The loudspeaker must be capacitively coupled therefore and the low frequency response will be worse as this capacitor is decreased in value. The output coupling capacitor C4 in Fig. 2 also provides the bootstrap connection to pin 4. In Fig. 3 an additional capacitor (C9) is required for this purpose.
In both circuits the value of R1 controls the amount of feedback and thus the gain. The output signal is fed back to pin 6 via an internal 7 kOhm resistor. If R1 is reduced in value the gain will increase but the frequency response will be affected and the distortion will rise. With the component values shown the voltage gain of both circuits is typically 140 (43dB) which is quite adequate for most audio applications. R3 in Fig. 3 is necessary only if the power supply voltage is fairly low (less than about 14V).
C2 smooths the power supply input and C1 is connected between pin 1 and chassis to provide r.f. decoupling and help prevent instability. If mains hum is present on the supply line with the circuit shown in Fig. 3 capacitor C8 should be included between pin 7 and chassis. The circuits shown have a level frequency response (within ±3dB) between about 40Hz and 20kHz. If you wish to reduce the upper 3dB level to about 8kHz C5 can be increased to about 560pF. The total harmonic distortion provided by these circuits remains fairly constant at about 0.5% until the power output reaches 3W: it then rises rapidly with power level as shown in Fig. 4.
The TBA800 can be operated from a 13V supply to feed up to 2.5W into an 80 load or from a 17V supply to feed the same power into a 160 load without an additional heatsink. If more output power is required the cooling tabs must be connected to a heatsink. Two methods of mounting the TBA800 are shown in Figs. 5 and 6. In Fig. 5 the device is inserted into a circuit board and a heatsink is soldered to the same points as the tabs: this has the disadvantage that the heatsink extends above the board though on the other hand the whole board can be used for the construction of the circuit. In Fig. 6 the tabs are soldered directly to a suitable area of copper on the board: this method has the disadvantage that about two square inches of the board are not available for component mounting. It is generally best to make soldered connections to the pins of the device since this ensures good heat dissipation with minimum unwanted feedback. Observe the usual heat precautions when soldering. The pins can however be carefully bent so that they will fit into a 16 -pin dual -in -line socket.
The TBA810S has the same type of encapsulation as the TBA800 and the connections are also as shown in Fig. 1 except that there is no internal connection to pin 3. An alternative version, the TBA810AS, has two horizontal tabs with a hole in each (see Fig. 7) so that a heatsink can be bolted on. Some readers may find it easier to bolt a heatsink to a TBA810AS than to solder the TBA810S tabs. TBA810 devices can provide 7W of audio power to a 40 loudspeaker when operated from a I6V supply. Fig. 8 shows the change in maximum output power with different supply voltages. As a 4.5W output can be obtained with a 12V supply the TBA810 is much more suitable than the TBA800 for use with battery operated equipment. The TBA810 can provide output currents up to 2.5A.
Two circuits for use with TBA810 devices are shown in Figs. 9 and 10: they are very similar to the circuits shown in Figs. 2 and 3 though some of the capacitor values are larger because of the lower output impedance. The two circuits have comparable performance but that shown in Fig. 10 gives somewhat better results at low supply voltages (down to 4V). In either circuit R2 may be replaced with a 100k0 volume control. The bias current flowing in the pin 8 circuit is typically
0-4 microA and the input resistance 5M 0 (the value of R2 must be much less however to ensure correct bias.
 The gain decreases as the value of R1 is increased for the same reason as with the TBA800. The values of R1, C3 and C7 affect the high -frequency response. With the values shown the response is level within ±3dB from about 40Hz to nearly 20kHz. Fig. 11 shows values of C3 plotted against R1 where the frequency is 3dB down at 10kHz and 20kHz and C7 is five times C3. The output distortion with these circuits is about 0.3% for outputs up to 3W rising to about 1% at 4W, 3% at 5W and 9% at 6W with a 14.4V supply voltage. The voltage gain is typically 70 times (37dB). Although this value is half that obtained with the TBA800 the input voltage required to produce a given output power is about the same for both types. This is because a smaller output voltage is required to drive a 40 load at a certain power level than is required to drive a 160 load.

The TBA810S may be mounted in the same way as the TBA800. One way of mounting the TBA810AS is shown in Fig. 12. It is simpler however to bolt flat heatsinks to the tabs.
Devices of this type will be destroyed within a fraction of a second if the power supply is accidentally con- nected with reversed polarity. When experimenting therefore it is wise to include a diode in the positive power supply line to prevent any appreciable reverse current flowing in the event of incorrect power supply connection. The diode can be removed once the circuit has been finalised. The TBA800 is likely to be destroyed if the output is accidentally shorted to chassis. The TBA810S and TBA810AS however are protected from damage in the event of such a short-circuit even if this remains for a long time (but note that the earlier TBA8I0 and  TBA810A versions did not contain internal circuitry to provide this protection). The TBA800 is not protected against overheating but the TBA810S and TBA810AS incorporate a thermal shutdown circuit.
For this reason the heat- sinks used with the TBA810S and TBA810AS can have a smaller safety factor than those used with the TBA800. If the silicon chip in a TBA810S or TBA810AS becomes too hot the output power is temporarily reduced by the internal thermal shutdown circuit. As with all high -gain amplifiers great care should be taken to keep the input and output circuits well separated otherwise oscillation could occur. The de- coupling capacitors should be soldered close to the i.c. -especially the 0 1pF decoupling capacitor in the supply line (this should be close to pin I).
Field Output Circuit:

 Fig. 13 shows a suggested field output stage for monochrome receivers with 12-17in. 110° c.r.t.s using the TBA81OS. For safe working up to 50°C ambient temperature each tab of the device must be soldered to a square inch of copper on the board. The peak -to - peak scanning current is 1.5A, the power delivered to the scan coils 0.47W, power disspipation in the TBA810S 1 8W, scan signal amplitude 4.1V, flyback amplitude 5V and the maximum peak -to -peak current available in the coils 1.75A

TUNING E REMOTE CONTROL SYSTEM (SL SUCH-LAUF ABSTIMM BAUSTEIN 29502.003.22) VIEW.

GRUNDIG SUPER COLOR 6236/30 SERIE 30F22 CHASSIS 29301-114.68(07)
TUNING E REMOTE CONTROL SYSTEM (SL SUCH-LAUF ABSTIMM BAUSTEIN 29502.003.22) VIEW.


The Tuning system in this set is a voltage synthesized tuner controlled by a sophisticated
and complex unit called Abstimm Baustein 29502.003.22 (Tuning Unit) which features all the functions of the set
via ASIC ICs and a uC (Microcontroller) from Texas Instruments TMS1100. Additional ASICs TMS1100P1072B (Uc Masked) TMS 3755 . ICs SN29799N , 2x TMS3529nl (Channel Memory), TMS3731bnl SN29762N.













Tuning system in this set is a voltage synthesized tuner controlled by a sophisticated
and complex unit called Abstimm Baustein 29502.003.22 (Tuning Unit) which features all the functions of the set
via ASIC ICs and a uC (Microcontroller) from Texas Instruments TMS1100. Additional ASICs TMS1100P1072B (Uc Masked) TMS 3755 . ICs SN29799N , 2x TMS3529nl (Channel Memory), TMS3731bnl SN29762N.

TMS1000General

General Information
Texas Instruments was locked in a race with Intel to create the first microprocessor. By most accounts Intel won with the 4004, but there are a few die hard TI fans who say the TMS1000 was first, because it was the first “computer on a chip” and that the 4004 was just a calculator chip.

Texas Instruments followed the Intel 8080 with the 4-bit TMS1000. So, while Intel was leading the industry in microprocessors, TI led with this industry unique design "a computer on a chip", specifically designed for control and automation purposes. The 1000 was the first MCU (MicroComputer Unit) , which is an MPU (MicroProcessor Unit) with other support chips (such as RAM, ROM, counters, timers, I/O interfaces) integrated on to the same silicon chip.

The original 1000 family consists of 6 chips the TMS1000 and TMS1200 are basic chips, the TMS1070 and TMS1270 are high voltage versions to interface to displays, the TMS1100 and TMS1300 provide twice the on-board ROM and RAM. The TMS1000, TMS1070, and TMS1100 are 28-lead packages, the TMS1200, TMS1270, and TMS1300 are 40-lead versions of the same chips (just 200 to the 28-lead chip numbers).

In the 80's TI added to the 1000 family. The 28-lead TMS1170 started with a TMS1100 base and added fluorescent display drive capability and expanded memory (2KB ROM). The TMS1370 was the same as the TMS1170 and added 27 I/O lines. An expanded memory group based on the original TMS1000 chips was also created. They were the TMS1400, TMS1470, and TMS1700 (64 Bytes RAM, 4KB ROM). There were 40-lead versions of the TMS1400 and TMS1470, which because the TMS1600 and TMS1670. CMOS versions were also added, denoted with a "C" suffix, such as TMS1200C.

The TMS1000 also had system evaluator chips. The original evaluator chips were the TMS1098 and TMS1099. These 64-lead evaluator chips were ROM-less versions of their corresponding standard chips. The TMS1099 supported the TMS1000/TMS1200 and the TMS1070/1270. The TMS1098 supported the TMS1100/1300. Later evaluators were introduced to support the entire TMS1000 family, they were the SE1000P (supports TMS1000,1070,1200,1700), SE2200P (supports TMS1100,1170,1300,1370), and the SE1400P (supports 1400, 1470, 1600, 1670).

The success of the the TMS1000 is demonstrated by its long lifecycle (over 20 years) and its expanded product line. The TMS1000 is found in many appliances, control systems, and games. Most of these chips were sourced by companies for direct use in their products and will have custom or house numbers on the chips (not the standard numbers listed above). Even TI used custom numbers in its products. The TMS1000 was used as a customized chip in the Texas Instruments "Speak and Spell" educational toy line (See Pictures at bottom).

Production
Early 1975

 GRUNDIG SUPER COLOR 6236/30 SERIE 30F22  CHASSIS 29301-114.68(07)  Microcomputer processing approach for a non-volatile TV station memory tuning system:


A television tuning system having a non-volatile memory for storing digital tune words is electrically updated by a microcomputer type architecture control circuitry. A ROM memory matrix is provided for the storage of VHF minimum and maximum binary tune words corresponding to each of twelve VHF channels in addition to a UHF minimum and maximum binary tune word encompassing all possible 72 UHF channels. Tuning of individual VHF and UHF chanels is accomplished by incrementing or decrementing a given tune word within the minimum and maximum limits established in the ROM memory matrix by means of a microcomputer processing approach.

TMS1000  General

General Information:
Texas Instruments was locked in a race with Intel to create the first microprocessor. By most accounts Intel won with the 4004, but there are a few die hard TI fans who say the TMS1000 was first, because it was the first “computer on a chip” and that the 4004 was just a calculator chip.

Texas Instruments followed the Intel 8080 with the 4-bit TMS1000. So, while Intel was leading the industry in microprocessors, TI led with this industry unique design "a computer on a chip", specifically designed for control and automation purposes. The 1000 was the first MCU (MicroComputer Unit) , which is an MPU (MicroProcessor Unit) with other support chips (such as RAM, ROM, counters, timers, I/O interfaces) integrated on to the same silicon chip.

The original 1000 family consists of 6 chips the TMS1000 and TMS1200 are basic chips, the TMS1070 and TMS1270 are high voltage versions to interface to displays, the TMS1100 and TMS1300 provide twice the on-board ROM and RAM. The TMS1000, TMS1070, and TMS1100 are 28-lead packages, the TMS1200, TMS1270, and TMS1300 are 40-lead versions of the same chips (just 200 to the 28-lead chip numbers).

In the 80's TI added to the 1000 family. The 28-lead TMS1170 started with a TMS1100 base and added fluorescent display drive capability and expanded memory (2KB ROM). The TMS1370 was the same as the TMS1170 and added 27 I/O lines. An expanded memory group based on the original TMS1000 chips was also created. They were the TMS1400, TMS1470, and TMS1700 (64 Bytes RAM, 4KB ROM). There were 40-lead versions of the TMS1400 and TMS1470, which because the TMS1600 and TMS1670. CMOS versions were also added, denoted with a "C" suffix, such as TMS1200C.

The TMS1000 also had system evaluator chips. The original evaluator chips were the TMS1098 and TMS1099. These 64-lead evaluator chips were ROM-less versions of their corresponding standard chips. The TMS1099 supported the TMS1000/TMS1200 and the TMS1070/1270. The TMS1098 supported the TMS1100/1300. Later evaluators were introduced to support the entire TMS1000 family, they were the SE1000P (supports TMS1000,1070,1200,1700), SE2200P (supports TMS1100,1170,1300,1370), and the SE1400P (supports 1400, 1470, 1600, 1670).

The success of the the TMS1000 is demonstrated by its long lifecycle (over 20 years) and its expanded product line. The TMS1000 is found in many appliances, control systems, and games. Most of these chips were sourced by companies for direct use in their products and will have custom or house numbers on the chips (not the standard numbers listed above). Even TI used custom numbers in its products. The TMS1000 was used as a customized chip in the Texas Instruments "Speak and Spell" educational toy line (See Pictures at bottom).



 1. A broadcast receiver tuning system for tuning said broadcast receiver to a selected frequency comprising:
first means for storing digital tune words responsive to a binary address for outputting a selected said digital tune word,
second means for storing said selected digital tune word and said binary address operably associated with said first means for storing,
a microcomputer operable for selectively changing said digital tune words in said first and second means for storing, and
means for converting said digital tune word stored in said second means for storing into an analog voltage operative to tune said broadcast receiver to said selected frequency.


2. A tuning system of claim 1 wherein said microcomputer comprises: means for incrementing and decrementing said digital tune word stored in said second means in updating said digital tune word,
means for providing a plurality of operating instructions and logic functions operative of said microcomputer,
means for storing binary data responsive to said binary address and said instructions operative for incrementing and decrementing said digital tune word stored in said second means for storing, and
means for inputting control functions operably associated with said means for providing a plurality of operating instructions.


3. A tuning system of claim 1 wherein said means for converting comprises: a pulse width modulator generator for outputting a digital signal proportional to said digital tune word, and
a digital to analog converter for converting said digital signal into said analog voltage for tuning said broadcast receiver to said selected frequency.


4. A tuning system of claim 1 wherein said broadcast receiver comprises a television set.

5. A tuning system of claim 1 wherein said first means for storing digital tune words comprises a nonvolatile random access memory.

6. A tuning system of claim 1 wherein said second means for storing said digital tune word and said binary address comprises a shift register.

7. A tuning system of claim 2 wherein said means for incrementing and decrementing comprises an arithmetic logic unit.

8. A tuning system of claim 7 wherein said arithmetic logic unit comprises: a plurality of shift registers,
a one bit full adder operably associated with said plurality of shift registers for adding and subtracting said digital tune words and said binary data stored in said plurality of shift registers, and
means for storing said digital tune words and said binary data operably associated with said plurality of shift registers and said one bit full adder.


9. A tuning system of claim 2 wherein said means for providing a plurality of operating instructions and logic functions comprises: a program counter,
an instruction memory responsive to said program counter for outputting binary instructions, and
a program logic array responsive to said binary instructions for outputting a plurality of said logic functions.


10. A tuning system of claim 9 further including a microprogram counter operably associated with said program counter.

11. A broadcast receiver tuning system for tuning said broadcast receiver to a selected frequency comprising: a first memory matrix for storing digital tune words corresponding to said selected frequency,
means for generating a binary address for addressing said digital tune word from said first memory,
means for storing said binary address and said addressed digital tune words operably associated with said first memory and said means for generating said binary address,
means connected to said address and tune word storing means for incrementing and decrementing said addressed digital tune word for updating said digital tune word,
means responsive to said binary address for outputting selected binary data from a second memory matrix, said binary data used for incrementing and decrementing said addressed digital tune word,
means connected to said incrementing and decrementing means for providing a plurality of operating instructions and logic functions operative for updating said digital tune word,
means for inputting control functions operably associated with said means for providing a plurality of operating instructions, and
means for converting said addressed digital tune word into an analog voltage operative to tune said broadcast receiver to said selected frequency.


12. A tuning system of claim 11 wherein said digital tune words further correspond to a plurality of VHF and UHF television channels.

13. A tuning system of claim 11 wherein said means for storing said binary address and said addressed digital tune word comprises a shift register.

14. A tuning system of claim 11 wherein said means for incrementing and decrementing comprises: a plurality of shift registers,
a one bit full adder operably associated with said plurality of shift registers for adding and subtracting said digital tune words and said binary data stored in said plurality of shift registers, and
means for storing said digital tune words and said binary data operably associated with said plurality of shift registers and said one bit full adder.


15. A tuning system of claim 11 wherein said means for providing a plurality of operating instructions and logic functions comprises: a program counter,
an instruction memory responsive to said program counter for outputting binary instructions, and
a program logic array responsive to said binary instructions for outputting a plurality of said logic functions.


16. A tuning system of claim 11 wherein said means for inputting logic control functions comprises an input logic status switch.

17. A tuning system of claim 11 wherein said means for converting comprises: a pulse width modulator generator for outputting a digital signal proportional to said digital tune word, and
a digital to analog converter for converting said digital signal into said analog voltage for tuning said broadcast receiver to said selected frequency.


18. A tuning system of claim 15 further including a microprogram counter operably associated with said program counter.

19. A tuning system of claim 14 further including an automatic channel shift encode for normalization of a binary VHF increment value comprising one of said binary data stored in said second memory matrix.

20. A television tuning system for tuning said television to a selected VHF and UHF channel comprising: a first memory matrix for storing digital tune words corresponding to said VHF and UHF channels,
means for generating a binary address on a multibus line for outputting said digital tune words from said first memory,
a shift register operably associated with said first memory and said means for generating a binary address for storing said digital tune word and said binary address,
an arithmetic logic unit for incrementing and decrementing said digital tune word stored in said serial shift register in updating said digital tune word,
means for providing a plurality of operating instructions and logic functions operative in updating said digital tune word,
a second memory matrix for storing binary data used in incrementing and decrementing said digital tune word, said second memory matrix responsive to said binary address and said operating instructions, said second memory matrix also operably associated with said arithmetic logic unit,
an input logic status switch for inputting control functions operably associated with said means for providing a plurality of operating instructions,
a pulse width modulator responsive to said digital tune word stored in said shift register for outputting a digital signal proportional to said digital tune word, and
means for converting said digital signal to an analog voltage operative to tune said television to said selected UHF or VHF channel.


21. A tuning system of claim 20 further including an automatic channel shift encode for normalization of a binary VHF increment value comprising one of said binary data stored in said second memory matrix.

22. A tuning system of claim 20 wherein said means for providing a plurality of operating instructions and logic functions comprises: a program counter,
an instruction memory responsive to said program counter for outputting binary instructions, and
a program logic array responsive to said binary instructions for outputting a plurality of said logic functions.


23. A tuning system of claim 22 further including a microprogram counter operably associated with said program counter.

Description:

BACKGROUND OF THE INVENTION
This invention relates in general to the tuning of a broadcast receiver, and more particularly relates to the tuning of a television receiver using a non-volatile memory for storing binary tuning words that are electrically updated by a microcomputer type architecture control circuitry.
Previously developed electronic channel tuning systems have not been sufficiently flexible to enable wide-spread use for a variety of different types of television sets in applications. For example, certain previously developed systems have required extremely uniform varactor tuning diodes to enable channel tuning, thereby allowing insufficient tolerances for conventional variances between varactor diodes. Other previously developed systems have not been sufficiently modular to enable a selection of various types of channel access or displays. Moreover, previously developed electronic channel tuning systems have not been sufficiently economical to fabricate and have required uneconomical printed circuit boards or other uneconomical fabrication techniques for construction. For example, certain prior systems have required expensive potentiometers for each channel desired to be tuned. In addition, previously developed electronic television tuning systems have not satisfactorily satisfied recent regulatory requirements which call for a television tuner to provide a comparable capability and quality of tuning for both VHF and UHF stations. Specifically, such prior tuning systems have not enabled selection of precise UHF channels, nor have the prior systems provided means for easily changing selected UHF channels.
A major disadvantage in the channel tuning sections of television receivers has been the inability to electronically program and store tune voltages under all operating and non-operating conditions without using an auxiliary power source or a mechanically programmed memory. Existing electronically operable tuners are dedicated electronic circuitry to program tune voltage information in volatile memories where the volatile memories require batteries to provide standby power when the main power source is removed. The batteries are undesirable because they represent an additional cost to the manufacturer and a present a long-term tune voltage jeopardy if they fail when the main power source is removed. Memory loss due to battery failure can occur if there are poor battery connections, battery corrosion, or excessive battery drain. Other tuning systems use potentiometers to retain the channel tune voltage, but are also undesirable because they are not electronically alterable, and require a potentiometer for each channel to be tuned.

In accordance with the present invention, the undesirable characteristics are eliminated by using a non-volatile DIFMOS memory matrix to store the channel tune voltages. The DIFMOS memory (dual injection floating gate MOS technology) is electronically alterable and has a projected memory retention capability of over 100 years with power removed. The control circuitry for the system uses a microcomputer type architecture to integrate the user control inputs and to generate the signals needed to access and alter the DIFMOS memory matrix. A principal advantage of this type of control compared to the dedicated control circuit approach is the ease with which different manufacturers' system requirements can be satisfied by simply reprogramming the algorithm of the instruction memory.
Accordingly, an object of the present invention is to provide an electronically programmable television tuning system having a non-volatile memory matrix for the storage of binary tune words.
Another object of the present invention is to provide electronic alterable tuning means for a broadcast receiver using a microcomputer approach, thereby eliminating the need for dedicated control circuitry.
Yet another object of the present invention is to provide means for electronically updating binary tune words of a selected channel in the tuning of a television receiver and for storing the updated binary words in a non-volatile memory matrix.
Still a further object of the present invention is to provide a means for generating a binary tune word corresponding to a selected UHF or VHF channel within the limits of a binary minimum and maximum word stored in a memory matrix.

SUMMARY OF THE INVENTION
A television tuning system is taught having a non-volatile RAM memory for storing digital tune words that are electronically updated by a microcomputer type architecture control circuitry. A five-bit binary address word is provided for addressing a 15-bit binary word from a non-volatile memory matrix. The 15-bit binary word comprises 14 bits corresponding to a tune word for the channel selected and a 15th MSB as a skip toggle indicator. The 20 bits are stored in three shift registers in the data in/out circuit in a 5-bit address buffer, a 1-bit skip toggle buffer, and a 14-bit data buffer register. The 14-bit tune word is placed in a data latch comparator for the PWM generator. An analog circuit provides the voltage conversion of the digital output of the PWM generator proportional to the tune word for applying to the varactor tuner of the TV at a selected frequency.
The binary tune word is incremented or decremented to provide an updated tune word in tuning the system by means of a microcomputer approach. The binary tune word is written and read from the non-volatile memory by the same microcomputer system.
The 14-bit binary tune word is updated either by external user control or AFC tuning. In either mode of operation, the tune word is incremented or decremented within a minimum and maximum binary tune word that is stored in a ROM memory matrix. In addition, increment values and tuning time limits are also stored in the ROM memory matrix. An arithmetic logic unit comprising a temporary storage RAM file, two 14-bit working registers, and a 1-bit full adder provide the means for performing the system's computations.
An 8-bit program counter provides the binary address of instructions in the 8 × 256 instruction ROM which addresses the PLA decode providing for an instruction generator. The PLA decode provides 26 "and" functions and 12 "or" functions. In addition, a 12 to 1 input logic status switch provides the necessary status indication for the 12 external controls. These input signals are detected by a 1-bit status latch.
The system is partitioned into two major functions: the non-volatile memory and the digital to analog converter and control circuits. The channel addressing and varactor diode band selection is generated with a rotary switch assembly. While a rotary switch assembly was used to implement the embodiment, non-volatile memory designs have been generated for addressing and band selection and could be easily implemented. The tune voltage interface between the digital to analog converter and the varactor diodes use standard oscillator and amplifier buffer circuits to provide the AFC summing and UHF tuning functions.

BRIEF DESCRIPTION OF THE DRAWINGS
The novel features believed characteristic of the invention are set forth in the appended claims. The invention itself, however, as well as further objects and advantages thereof, will best be understood by reference to the following detailed description of an illustrated embodiment taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a functional block diagram employed to illustrate the present invention in a TV receiver.
FIGS. 2, 2A-2B are detailed circuit diagrams of the input buffer registers in the data in/out circuit.
FIGS. 3, 3A-3B are detailed circuit diagrams of the ROM constant file and its addressing circuitry.
FIG. 4 is a detailed circuit diagram of the automatic channel shift encode.
FIGS. 5, 5A-5D are detailed circuit diagrams of the instruction ROM, program counter, and microprogram counter.
FIGS. 6, 6A-6D are detailed circuit diagrams of the instruction PLA.
FIGS. 7, 7A-7B are detailed circuit diagrams of the input logic status switch.
FIGS. 8, 8A-8D are detailed circuit diagrams of the arithmetic logic unit.
FIGS. 9, 9A-9B are detailed circuit diagrams of the PWN generator.
FIGS. 10, 10A-10B are detailed circuit diagrams of the analog circuitry.
FIGS. 11A-11H are detailed architecture diagrams of the microcomputer system.
FIG. 12 represents the tune voltage amplifier diagram and related equations for calculating binary words corresponding to tune voltages.
FIGS. 13, 13A-13L are detailed drawings of the instruction set algorithm for the non-volatile stationary memory tuning system.

DETAILED DESCRIPTION
A more complete understanding of the detailed embodiment will be understood by a brief description of the requirements of the system. The fine tune up or down is accomplished by a rocker switch with center off position. A closed position on the switch will increment the tune voltage at the rate of 2 to 8 steps per second. The fine tune control is operative on VHF and UHF tuning modes.
UHF programming is accomplished by pushing a potentiometer control knob and turning the knob pointer to the desired channel number. When the knob is pushed, a contact is switched to ground. The knob is spring loaded in the out position and cannot be turned unless pushed in. The UHF programming potentiometer has approximately 30 turns. The user is able to fine tune a UHF station with this potentiometer and also with a fine tune rocker switch. The UHF fine tune limit is said to be plus or minus 128 steps from the binary word stored in the non-volatile memory RAM matrix only when the fine tune rocker switch is used. If the user continues to hold the rocker switch in the same mode after 128 steps, the tune voltage reverses direction and increments in the other direction for 256 steps until it hits the other limit where it reverses direction again.
Storage and memory requires approximately 240 milliseconds. The binary tune voltage word and skip signal is stored when the set is turned off. If any tuning control for the channel skip button has been engaged while addressing the channel, the tune voltage and skip will also be stored in the memory when a channel change occurs.
An interchannel AFC defeat pulse occurs between each adjacent channel position. The pulse occurs when a switch contact is momentarily shorted to ground. The duty cycle of the pulse is approximately constant versus the rate of rotation of the channel select knob. The duty cycle is about 25% contact closed and 75% contact open. The binary input address is sampled and latched at the end of a write time or 48-68 milliseconds after receipt of the last interchannel pulse, whichever occurs last. A user programmable skip channel signal output is utilized. The operator uses a pushbutton to change the state of the signal.
The system has been designed for a 20 channel capacity. This includes 12 dedicated VHF channels plus 8 undedicated UHF channels. In VHF mode, a ROM plus non-volatile RAM approach is used to limit fine tuning. The ROM plus RAM make up a 14-bit tuning word plus a 1-bit skip flag. The RAM is 8-bits tuning word plus skip flag. The system is designed such that the LSB of the 8-bit tuning word can be reprogrammed for each VHF channel to occur anywhere from the LSB position to the 7th bit of the 14-bit tuning word. In the UHF mode the RAM shall be 14-bits for the tuning word plus 1 bit for skip flag.
Referring now to the block flow diagra
m of FIG. 1, the TV tuning microcomputer approach flow diagram is indicated. The television receiver 2 has a selector switch 26 for generating an address for the non-volatile memory matrix contained in the microprocessor circuitry 4. A more detailed block diagram of the non-volatile memory architecture and address architecture is indicated in FIG. 11A. In one embodiment the non-volatile memory comprises a DIFMOS memory matrix (dual injection floating gate metal oxide semiconductor). Data retention without power is achieved by storing charge on an array of floating gates. Any floating gate in the memory array can be charged or discharged by the injection of electrons or holes from an avalanche plasma formed in two special injector structures within each bit. Once a floating gate has been charged, it will stay charged almost forever, unless it is intentionally discharged by reprogramming. The decay rate of a charge from a floating gate has been measured at less than 1% of the initial value per decade of time at 85° C. In the embodiment described the Texas Instruments X-929A decoded 32 bit non-volatile RAM semiconductor memory is used. However, other non-volatile memories may additionally be used in the present invention.

Digital tune words corresponding to the UHF and VHF channels are read from the memory and written into the memory by way of the data in/out circuitry. The data in/out circuitry contains temporary storage registers for the 5 bit channel address, the 1 bit skip toggle indicator, and the 14 bit tune word. The tune word is loaded into the PWM comparator where a PWM counter and PWM generator produce digital output signals proportional to the binary tune word. These digital output signals are fed to an analog circuit comprising an op-amp for the conversion to the analog voltage required to be applied to the varactor tuner of the television for tuning at the selected channel.
The channel shift encode is provided to normalize the bit weighting of the increment value for selected VHF channels. The normalized binary word is applied to the microprogram counter to provide shift controls to the various shift registers of the tuning circuitry in the VHF mode.
Input commands by the user is read into the system by means of the input logic and status latch. This provides a means of detecting a change of state on the input switches during a tuning function so that the system may be changed to the latest input command. The change of state is detected by a status latch which loads a new address of the instruction ROM into the program counter.
The program counter provides any one of 256 instruction addresses of the instruction ROM. The instruction ROM addresses the constant memory matrix which contains the upper and lower limits for the VHF and UHF channels, increment values for both VHF and UHF tuning, time increments, write time, and maximum times. In addition, the instruction ROM addresses the instruction PLA which contains decoding for 26 "and" functions and 12 "or" functions.
The ROM constant memory matrix transfers the data to the arithmetic logic unit which contains 2 working registers, a 1 bit full adder and a RAM temporary storage file. The arithmetic logic unit provides for the operation of incrementing and decrementing tune words, providing for write times, and time out functions. The new binary tune word from the arithmetic logic unit is loaded into the data in/out circuitry or read from it. In all aspects of the operation of the present invention, the binary tune word corresponding to the individually selected channel in both the VHF and UHF mode are stored in the non-volatile RAM memory matrix for addressing upon channel selection.

Referring now to FIGS. 2A-2B the data in/out circuitry comprising the input buffer registers is indicated in greater detail. A 5 bit address buffer serial register 230 is provided in addition to 2 D flip flops 232 and 254. A 3 to 1 encode 236 is provided for transmitting of data to the 14 bit input data buffer serial register 234A-234B. Data stored in the input data buffer is parallel loaded into the 14 bit data latch serial register 238A-238B the output of which is parallel loaded into a 14 bit pulse width modulated logic latch serial register. The D flip flop 232 provides as a 1 bit skip toggle buffer for the MSB of the tune word.
A more detailed circuit diagram of the address decode and ROM

constant file is indicated in FIGS. 3A-3B. Five bits from the address generator and 4 bits from the instruction ROM are decoded to address the 32 by 14 bit ROM constant file 264A-264B. The output of the ROM constant file is loaded into a 14 bit B working register.

The automatic channel shift encode for normalization in VHF tuning is indicated in greater detail by the circuit diagram in FIG. 4. The 4 LSB's of output is applied to an encode of the microprogram counter. Two serial shift registers 100 and 102 are provided for transfer of data in the decode operation.
The instruction ROM, program counter, and microprogram counter circuitry are indicated in greater detail in FIGS. 5A-5D. The 8 by 256 bit instruction ROM 286A-286B is addressed by the 8 bit program counter 290A-290B. The 8 bit program counter is divided into two serial registers comprising 4 MSB's and 4 LSB's. The LSB's are loaded directly from the 8 bit instruction program word from the instruction ROM. The 4 MSB's are loaded into the program counter by means of the 4 bit page latch 294. In addition, 6 bits of the instruction program word are applied to a PLA decode and 4 LSB's of the instruction program word are applied to a 9 by 32 address decode of a ROM constant file. The 8 to 4 encode 302A and 302B is addressed by 4 LSB's from the 8 bit instruction word and 4 bits from an automatic channel shift encode. These 8 bits are encoded to 4 bits which addresses the 4 bit microprogram counter 400. Also, the 4 LSB's from the instruction ROM addresses a 4 to 12 decode for an input logic status switch. Two of the 4 LSB's addresses a 2 to 4 decode of a temporary storage RAM file in the arithmetic logic unit.

FIGS. 6A-6B is a more detailed circuit diagram of the instruction PLA. Six bits of address from the instruction ROM are used to address the 6 by 28 by 12 bit PLA decode. The output of the PLA comprise 26 "and" functions and 12 "or" functions.
FIGS. 7A-7B is a more detailed circuit diagram of the input logic status switch. The 12 inputs to the status switch are read by decoding 4 LSB's of instruction word from an instruction ROM. An indication of a match between the decode and the 1 of 12 inputs is indicated by the setting of a status latch 282. This status latch is loaded to the one state in the presence of any of the 12 input functions and a matching code.

FIGS. 8A-8D are a more detailed circuit diagram of the arithmetic logic unit and temporary storage RAM file. The 14 bit word from a ROM constant file is parallel loaded into the 14 bit B working serial register 274A-274B. A 4 by 14 bit temporary storage RAM file 276A-276H is provided for temporary storage of the data from the ROM constant file and working registers. The temporary storage RAM file has four memory locations that are selected by the 4 to 1 decode 308. Access to working register B is by means of a 2 to 1 encode 304 and access to the 14 bit A working shift register 266A-266B is by means of the 4 to 1 encode 270. The temporary storage RAM file is accessed by means of the 3 to 1 encode 278. A 1 bit full adder 288 is provided for addition and subtraction of the A and B working registers. Two LSB's of instruction word are used to address the temporary storage RAM file.

FIGS. 9A-9B is a more detailed circuit diagram of the pulse width modulator (PWM) generator. A 2¹⁴ PWM counter 250A-250B is provided. The binary word output is parallel loaded into a 14 bit PWM logic latch. When the 14 bit binary word from the PWM counter matches the 14 bit tune word stored in the 14 bit data latch the PWM logic latch is tripped and the PWM digital output is generated.

FIGS. 10A-10B are a more detailed circuit diagram of the analog circuit for converting the digital output of the PWM generator of an analog voltage to be applied to the varactor tuner of the television. In addition, circuits for PWM power up clear, AFC defeat, interchannel pulse, and UHF up/down circuitry are provided.














Referring now to the system diagram of FIGS. 11A-11H, the TV tuning microprocessor architecture is indicated in greater detail than the block diagram of FIG. 1. The 5 bit binary channel address is read off the 20 position selector switch 202 by means of the address generator 204 in FIG. 11A. The binary address corresponds to any one of 20 channels, 12 of which are VHF channels and 8 of which are UHF channels. In addition to the channel addressing the selector switch has means for channel interrupt selection 224, means to select the varactor band of the TV tuner 226, and means to program AFC bias on and off 228. The channel address is read directly into the 5 bit address latch 206 in the non-volatile RAM circuitry. Information in the 5 bit address latch 206 is used to address the 32 bit addressable non-volatile RAM matrix and also provides a parallel input into a 5 bit address shift register 208. The 5 bit address on a multibus line from the selector switch is used to address one of the 20 locations in the non-volatile memory circuitry used to retain the binary tune word. Provided in the memory circuitry are 12 VHF binary tune words and 8 UHF binary tune words.

In series with the shift register 208 is a 15 bit data out shift register 210. These two shift registers 208 and 210 are in a read mode when not programmed to shift out. Therefore they are always looking at and reading the address latch 206 and the 15 bits of the memory matrix 212. Fourteen bits of the non-volatile RAM matrix are used for representing the binary tune word and the 15th MSB is used for a skip toggle indicator. The 20 bits comprising 5 from the address register and 15 bits from the data out register are serially shifted out when we read the non-volatile memory 212. As the bits are serially shifted out they are also fed back into the stack in a serial manner by loop 222 so that the 5 bit address and the 15 bit data tune word are restored into the registers.
The address and data tune word as they are shifted out of the registers into the control chip are fed into a 20 bit input data buffer comprising a 5 bit address buffer 230, a 1 bit skip toggle buffer 232, and a 14 bit input data buffer register 234 indicated

 in FIG. 11B. The address buffer register 230 contains the last bits shifted out of the non-volatile memory block which comprises the 5 address bits. The 6th MSB is the skip toggle bit and resides in the skip toggle buffer register 232 immediately following the address buffer. The 14 bit data tune word is steered through a selector switch encode 236 into the 14 bit input data buffer register 234. The selector switch encode 236 has 3 select states comprising load input data buffer (LIDB), read non-volatile memory (RNVM), and read input data buffer (RIDB). The 14 bit tune word is loaded into the data buffer register 234 by selecting the read non-volatile memory mode of the selector switch encode 236.
The binary tune word in data buffer register 234 is loaded parallel into the 14 bit PWM logic latch 248 when there is a load PWM (LPWM) signal on the 14 bit data latch 238. The 14 bit tune word in the PWM logic latch 248 is used as a compare word for the 14 bit pulse width modulator counter 250. The pulse width modulator operates with a 1 MHz input clock from the PWM buffer and oscillator 252 that is fed into the 2¹⁴ PWM counter 250 and runs continuously.
The PWM counter 250 counts from binary 0 in a binary manner until it reaches one of two conditions. First, if the binary word of the counter 250 compares with the 14 bit tune word in the PWM logic latch 248 then the PWM logic latch which is performing a magnitude compare will provide an output signal and trip a flipflop which will then remain in that state until the counter completes its count-out cycle. The second condition is when the PWM logic latch 248 is at an all 1 state whereby the PWM counter would count up to an all 1 state that also corresponds to the runover point of the counter. Therefore, the PWM counter will always count up to 2¹⁴ and then run over where 2¹⁴ and a 1 MHz input corresponds to a writeout at 16 milliseconds.
In the PWM generator we therefore have a magnitude compare of the PWM counter with the 14 bit tune word stored in the 14 bit PWM logic latch 248 and when the first time there is a match of the counter and the binary magnitude we receive an output signal from the PWM logic latch 248 proportional to the tune word. To tune the television we alter the pulse width modulated signal from the PWM logic latch 248. We alter the pulse width modulated signal by changing the bit value of the binary tune word contained in the PWM logic latch thereby giving us a modulated pulse width at a duty cycle of 16 milliseconds.
The skip toggle bit in the skip toggle buffer register 232 may be altered by means of the skip toggle inputs through the MAND gate 256. In the program algorithm when the skip toggle is altered we read the information out and write it into memory once the function is complete. Altering of the skip toggle information is achieved by first reading the state of the skip toggle buffer 232 which contains an MSB that was read out of memory, loading that bit into a D register 254, and changing that information if we have a program input to change the state of the skip toggle. A skip toggle output 258 is provided to give an indication that the skip toggle has been altered and the present program condition of the skip toggle. The skip toggle is not applicable to a mechanical rotary type selector switch system as indicated in this embodiment whereas the selector switch 202 is of a rotary type. However, by replacing the rotary selector switch with an electronically addressable circuit as disclosed in U.S. Pat. No. 3,968,443 issued on July 16, 1976, assigned to Texas Instruments Incorporated, the same assignee of the present patent application, then a skip function would be applicable in the present tuning circuitry.
After loading the address buffer 230 with the 5 bits of address from the address generator 204 these 5 address bits are transferred in a parallel mode to the 9 by 32 address decode 260 and the 5 to 4

automatic channel shift encode (VHF only) 262 indicated in FIG. 11C. The automatic channel shift encode is used to determine whether the system is functioning in the UHF or VHF mode. If the system is functioning in the VHF mode the automatic channel shift encode provides one of four possible codes for incrementing the VHF tune word. The four codes corresponding to the particular incrementing bit value that applies to the VHF channel that has been selected by the address generator 204. Since there are only four increment rate values and 12 VHF tune words, the encode 262 selects depending upon which channel the system is on one of the four incrementing rate values to be applied to the given tune word.
The 5 bit address from the address buffer 230 is also parallel applied to the 9 by 32 address decoder 260 to select a 14 bit data word stored in the ROM constant file 264. The 5 bit address which determines the VHF or UHF channel is decoded by the 9 by 32 address decoder into a 32 bit address word to address the 32 by 14 ROM constant file. The four LSB's of the instruction code determines which of the 32 words we are addressing. These 32 fourteen bit words in the ROM constant file comprise upper and lower limits for the VHF channels, UHF channel limits, increment values for both VHF and UHF tuning, time increments, maximum times, and write time.
When we have read a tune word into the input data buffer 234 and want to perform a tuning function upon it, we transfer the 14 bits of data out of the input data buffer register and into the 14 bit A working register 266 by means of a read input data buffer (RIBD) command at the 4 to 1 encode switch 270. Also, the 14 bit word is serially transferred back into itself by means of loop 272. After loading register A with the 14 bit tune word, the tune limit and increment value is outputted from the 14 bit ROM constant file and loaded into the 14 bit B working register 274. These values are now loaded into the temporary storage RAM file 276 by selection of the LBMX command on the selector switch encode 278. The temporary storage file comprises a 4 by 14 bit RAM file. Tuning is now performed by adding an increment value which is stored in register B to the 14 bit tune word stored in register A if the system is in a tuned upmode and subtracting them if the system is in a tuned downmode.

 The incremented or decremented tune word is restored into the A working register by means of the "A" normalize command on the selector switch encode 270 indicated in FIG. 11D.
After the restore operation the updated 14 bit tune word is transferred into the input data buffer 234 by performing a load input data buffer (LIDB) command on selector switch encode 236. The updated tune word is now stored into the input data buffer and also restored into register A. The updated tune word is now loaded into the 14 bit PWM logic latch 248 whereby the PWM counter 250 can compare its out to updated tune word.
Whenever a tuning function is performed the system goes through a sequence whereby it performs an addition and a time out routine in the arithmetic logic unit by decrementing our timing word until a negative number is reached. In each case information is read from the ROM constant file and stored into the temporary storage RAM file. This information is a function of the particular channel and whether the channel is a UHF or VHF channel. In the sequence the system always goes through reading the input switches so if there is a change of state on our input switches during a tuning function it will be detected and the system function will be changed to the latest input command.

These input control functions are read into the system by means of the 12 to 1 input logic status switch 280 having 12 inputs indicated in FIG. 11F. A 61 kilohertz slow clock is provided to perform the write function which in the case of the non-volatile memory comprising DIFMOS memory cells takes in the order of 100 milliseconds to write a 0 into the memory, therefore requiring a clock running at a slower rate then the control or processing clock that is normally used. The slow clock is also used to provide dampening when in the power up mode or after we have already completed a write command in writing into memory so that the system doesn't read the new word while it is still settling.
Another input is the UHF/VHF control line that is a function of the particular address that has been detected from our selector switch 202. A third and fourth input is an AFC high and an AFC low select. The function of the AFC high/low is to provide a digital AFC control function. The means of achieving the digital AFC control is not indicated in the figures or represented in the algorithm. However, the digital AFC control system could be incorporated into the architecture by means of a couple of comparator windows and the appropriate addition of control logic to the present algorithm.
A fifth input is a UHF up/down control that is a control from a comparator 282 that determines whether the tune voltage is above or below the corresponding potentiometer setting of the UHF channel coarse tune potentiometer 284. An additional input is a power on/off select. Upon a power down input the 14 bit tune word stored in the input data buffer register 234 is written into the addressable non-volatile memory.
The seventh input is a skip toggle input which is not incorporated into the present system. This skip function if made available would allow for the skipping over of selected channels but is not applicable to a mechanical rotary switch as noted above.
The eighth and ninth inputs comprise the rocker arm fine tuneup and fine tunedown for the control voltage. The UHF tune on/off control places the tuning function into a coarse UHF tuning mode. The AFC on/off switch is used to activate the internal AFC tuning function or to allow for the external manual tuning mode. The final switch on the input logic status switch is the interchannel pulse that is inputted from the selector switch 202 by means of the channel interrupt line 224. The interchannel pulse provides an indication that the selector switch is in between channels in a changing mode and also detects the completed change.
The twelve inputs are read into the input logic status switch. If one of the twelve logic status switches is activated it is compared with a particular select code and if there is any indication of a match on the read command of that given instruction to the particular switch being closed or opened as the case may be, the status latch flip-flop 282 is set. The status switch inputs are decoded by the 4 to 12 decoder 284 which is addressed by four LSB's of instruction from the instruction ROM 286. The status latch 282 provides an indication that the system has received an input corresponding to one that has been coded in the instruction table of the decoder 284.
The second input to the status latch 282 is the carry input from the one bit full adder when the system is in a subtract routine and if the subtract routine results in a negative number. The negative number indication is used to perform compare tests to determine whether an upper or lower tuning limit or timing limit has been reached. The setting of the status latch 282 provides an input to the instruction ROM 286 to load the program counter with a new page of instruction address.

The eight bit program counter 290 indicated in FIG. 11C receives its count clock input from the 250 kilohertz clock 292 which is a one quarter division of the one megahertz clock from the PWM counter 250. The program counter gives any one of 256 instruction addresses for the instruction ROM 286. The location of the program counter in its counting sequence may be altered by loading in a new eight bit word into the program counter. The four LSB's from the instruction ROM are parallel loaded into the four LSB positions of the eight bit program counter and parallel loaded into a four bit page batch 294. If the status latch is set by a subtract operation from reading an input from the logic status switch, then upon a load page command (LPD) applied to the NAND gate 296 the four LSB bits of address with be loaded from the page latch into the program counter in the MSB position.
The output of the instruction ROM 286 feeds into the instruction PLA circuitry 298 which outputs 26

"and" functions and 12 "or" functions indicated in FIG. 11E. The instruction PLA decode comprises the 6 by 28 by 12 bit memory. These "and" and "or" functions correspond to the instruction set that is used to program the system.
The four bit microprogram counter 300 is used to provide shift controls to the various shift registers of the tuning circuitry. And more particular, the microprogram counter allows for the shifting of the 14 bit data word in working register A into the input data buffer register. In addition it allows for the addition and subtraction of working registers A and B and also allows for the transfer of data to the non-volatile memory.
The maximum number of serial shifting by the microprogram counter is 14 bits. When the shifting produced by the microprogram counter is completed, the system operation is returned to the eight bit program counter where it is indexed to the next address in the program. The eight bit microprogram instruction is selected by the microprogram address select encode 302. Four bits from the automatic channel shift encode 262 and four LSB's from the instruction ROM are loaded parallel into the address select encode to provide four bits of instruction address for the microprogram counter 300.

Referring to FIG. 11D a switch encode 304 is provided to allow for a restore operation whereby the 14 bit word in the B working register is shifted back into itself. In addition the switch encode provides for a shifting of the 14 bit word from the temporary storage RAM file 276 into the working register. Switch encode 270 allows for the shifting of a 14 bit word into the A working register from the temporary storage RAM file, the input data buffer register 234, a sum product from the addition of working register A and register B by means of the one bit full adder 288, and finally the restore of the word in the A register into itself.
The temporary storage RAM file 276 is addressed by two bits from the instruction ROM through a 2 to 4 encode 306. The four bit word from the encode is used to select one of four 14 bit storage files in the RAM file 276 by means of the 4 to 1 selector encode 308.
The pulse width modulated output 310 from the PWM logic latch is fed into the PWM buffer 252.

 The PWM signal from the PWM buffer is fed into a driver buffer 312 that is referenced to +5 volts in FIG. 11H. The PWM output continues through a three stage PWM filter to provide the IC filtering required for the resolution and ripple voltage needed for a pulse width modulated signal of the longest duration to an acceptable level in the UHF mode. The VHF mode would not need as much filtering to generate a PWM at an acceptable ripple level. However, at least three stages are required for UHF filtering.
The output of the three stage filter is a DC voltage that is proportional to the pulse width modulated signal, the pulse width modulated signal being proportional to the 14 bit tune word that has been loaded into the 14 bit input data buffer register and PWM logic latch. The tune voltage is amplified by inverting voltage amplifier 316 and subsequently filtered by an additional single stage filter 318. The final DC analog tune voltage is passed to the television varactor tuner for tuning to the selected channel.
A second comparator 282 comprises a UHF up/down comparator which receives its inputs from the three stage PWM filter and a UHF course tune potentiometer 284. The potentiometer is referenced to the same +5 volts as the driver buffer 212. The comparator 282 provides an indication as to whether or not the system is in the coarse tune mode of UHF, whether or not the system is above or below the desired tune voltage for the particular channel setting, and provides a coarse tuning signal for the controller.

FIG. 320 represents the power supply required for the operation of the tuning system. The +5 volts used to provide the upper voltage for the tuning amplifier for the varactor tune voltage output. The +17 volts is used for biasing of the MOS circuitry of the non-volatile memory. The +10 volts is used for biasing the CMOS logic in the system. The +5 volts is used for the TTL and I² L logic in the system. Finally, the 0 to -35 switch voltage is used for programming the non-volatile memory when the system is in a write mode.
In performing a tuning function using the microcomputer approach in the VHF mode a binary tune word that is stored in the input data buffer register is incremented or decremented within the limits for the minimum and maximum tune voltages for the selected channel that is stored in the ROM constant file. The ROM constant file contains a binary word for the maximum tune voltage and minimum tune voltage for each of the 12 VHF channels. These limits establish the range of tuning permitted by the system. These values are individually selected for each of the VHF channels. In a similar manner minimum and maximum limits are established for the UHF channel. However, due to the large number of UHF channels the minimum and maximum limit are established so as to encompass all 72 UHF channels with tuning for the selected UHF channel falling therebetween.

Referring now to FIG. 12 the schematic diagram used for calculating the minimum and maximum tune voltages is indicated in addition to the equations used. Equation 3 is the input voltage as a function of the output tune voltage. Given the desired tune voltage E_O the input voltage E_i may be calculated. In addition, by using equation 4 the bits corresponding to the input voltage is calculated thereby resulting in the binary tune word corresponding to the calculated input voltage.
In this regard, Table I indicates the VHF ROM constants for the minimum limits as established by equations 3 and 4. Each of the VHF channels have a unique binary word corresponding to the minimum voltage limit. In a similar manner Table II indicates the VHF ROM constant for the maximum tune voltage. As noted the nominal tune voltage for tuning the television will lie somewhere between these two established limits. The binary words comprise 14 data bits and are addressed by the 5 bit binary address from the selector switch.

TABLE I

__________________________________________________________________________

VHF ROM CONSTANTS (MINIMUM) NOM MAX MIN CH# EAFC Eo Ei BITS 8192 4096 2048 1024 512 256 128 64 32 16 8 4 2 1

__________________________________________________________________________

2 1.549 2.149 1.442 4725 0 1 0 0 1 0 0 1 1 1 0 1 0 1 3 1.814 4.320 1.367 4479 0 1 0 0 0 1 0 1 1 1 1 1 1 1 4 2.310 7.158 1.444 4731 0 1 0 0 1 0 0 1 1 1 1 0 1 1 5 4.589 23.000 1.301 4263 0 1 0 0 0 0 1 0 1 0 0 1 1 1 6 4.528 29.885 0- 0- 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 2.299 7.550 1.361 4459 0 1 0 0 0 1 0 1 1 0 1 0 1 1 8 2.490 9.440 1.249 4092 0 0 1 1 1 1 1 1 1 1 1 1 0 0 9 2.726 11.790 1.107 3627 0 0 1 1 1 0 0 0 1 0 1 0 1 1 10 3.021 14.710 0.934 3060 0 0 1 0 1 1 1 1 1 1 0 1 0 0 11 3.594 18.370 0.955 3129 0 0 1 1 0 0 0 0 1 1 1 0 0 1 12 4.049 23.000 0.665 2179 0 0 1 0 0 0 1 0 0 0 0 0 1 1 13 4.621 29.920 0.103 337 0 0 0 0 0 1 0 1 0 1 0 0 0 1

__________________________________________________________________________

TABLE II

__________________________________________________________________________

VHF ROM CONSTANTS (MAXIMUM) NOM MIN MAX CH# EAFC Eo Ei BITS 8192 4096 2048 1024 512 256 128 64 32 16 8 4 2 1

__________________________________________________________________________

2 1.549 0.400 1.754 5747 0 1 0 1 1 0 0 1 1 1 0 0 1 1 3 1.814 0.888 1.979 6484 0 1 1 0 0 1 0 1 0 1 0 1 0 0 4 2.310 2.149 2.339 7664 0 1 1 1 0 1 1 1 1 1 0 0 0 0 5 4.589 4.682 4.572 14,980 1 1 1 0 1 0 1 0 0 0 0 1 0 0 6 4.528 12.260 3.147 10,311 1 0 1 0 0 0 0 1 0 0 0 1 1 1 7 2.299 4.624 1.884 6173 0 1 1 0 0 0 0 0 0 1 1 1 0 1 8 2.490 6.010 1.861 6098 0 1 0 1 1 1 1 1 0 1 0 0 1 0 9 2.726 7.550 1.865 6111 0 1 0 1 1 1 1 1 0 1 1 1 1 1 10 3.021 9.440 1.875 6144 0 1 1 0 0 0 0 0 0 0 0 0 0 0 11 3.594 11.790 2.130 6979 0 1 1 0 1 1 0 1 0 0 0 0 1 1 12 4.049 14.710 2.145 7028 0 1 1 0 1 1 0 1 1 1 0 1 0 0 13 4.621 18.370 2.166 7097 0 1 1 0 1 1 1 0 1 1 1 0 0 1

__________________________________________________________________________

The data including the VHF minimum and maximum limits are stored in the ROM constant file as indicated in Table III. As noted the ROM constant file has 32 separate data values stored therein. The VHF and UHF increment values are also stored in the ROM constant file. The maximum time for both VHF tuning and UHF tuning are also stored therein. In addition, the time increment value is also stored. Finally, the UHF minimum tune word and the UHF maximum tune word including the write time is stored in the ROM constant file.

TABLE III

______________________________________

ROM CONSTANT FILE Add- Prom ress Code Instruction No. Binary Msb Lsb

______________________________________

# 2 VHF MIN 0 00000 0 010 0100 0 111 0101 # 2 VHF MAX 1 00001 0 010 1100 0 111 0011 # 3 VHF MIN 2 00010 0 010 0010 0 111 1111 # 3 VHF MAX 3 00011 0 011 0010 0 101 0100 # 4 VHF MIN 4 00100 0 010 0100 0 111 1011 # 4 VHF MAX 5 00101 0 011 1011 0 111 0000 # 5 VHF MIN 6 00110 0 010 0001 0 010 0111 # 5 VHF MAX 7 00111 0 111 0101 0 000 0100 # 6 VHF MIN 8 01000 0 000 0000 0 000 0000 # 6 VHF MAX 9 01001 0 101 0000 0 100 0111 # 7 VHF MIN 10 01010 0 010 0010 0 110 1011 # 7 VHF MAX 11 01011 0 011 0000 0 001 1101 # 8 VHF MIN 12 01100 0 001 1111 0 111 1100 # 8 VHF MAX 13 01101 0 010 1111 0 101 0010 # 9 VHF MIN 14 01110 0 001 1100 0 010 1011 # 9 VHF MAX 15 01111 0 010 1111 0 101 1111 #10 VHF MIN 16 10000 0 001 0111 0 111 0100 #10 VHF MAX 17 10001 0 011 0000 0 000 0000 #11 VHF MIN 18 10010 0 001 1000 0 011 1001 #11 VHF MAX 19 10011 0 011 0110 0 100 0011 #12 VHF MIN 20 10100 0 001 0001 0 000 0011 #12 VHF MAX 21 10101 0 011 0110 0 111 0100 #13 VHF MIN 22 10110 0 000 0010 0 101 0001 #13 VHF MAX 23 10111 0 011 0111 0 011 1001 VHF INCRE- 24 11000 0 000 0000 0 100 0000 MENT TIME INCRE- 25 11001 0 000 0000 0 000 0001 MENT MAX TIME 26 11010 0 000 0101 0 000 1100 (ROCKER) UHF INCRE- 27 11011 0 000 0001 0 001 0101 MENT UHF MIN V 28 11100 0 000 0000 0 010 0000 UHF MAX V 29 11101 0 111 0100 0 101 0110 WRITE TIME 30 11110 0 000 1100 0 001 1011 MAX TIME 31 11111 0 000 0000 0 010 1000 (UHF)

______________________________________

Table IV indicates the PLA logic for automatic right shift addressing (VHF only) of the microprogram counter. As noted from Table III the VHF increment value has a 1 in the 7th bit position. During the incrementing or decrementing of the VHF word it is desired to increment or decrement at a particular bit weight value unique to each of the VHF channels. To accomplish this, the automatic channel shift encode provides for right shifting of the increments value so as to normalize it to provide a different bit weight for each of the 12 UHF channels. The number of right shift in the VHF mode for each of the selected channels is indicated in Table IV in addition to the encode word for the microprogram counter preset. Since the UHF tuning is performed by a process which provides for the increasing of the bit weight of the increment value, the normalization by right shifting the increment value is not needed.

TABLE IV

______________________________________

PLA LOGIC FOR AUTOMATIC RIGHT SHIFT ADDRESSING (VHF ONLY) OF μ PROGRAM COUNTER No. of Right Encode Word For VHF Channel Shifts For For μ Program No. Binary Word Normal Ration Counter Preset

______________________________________

2 00000 4 1011 3 00001 3 1100 4 00010 2 1101 5 00011 1 1110 6 00100 0 1111 7 00101 3 1100 8 00110 3 1100 9 00111 3 1100 10 01000 3 1100 11 01001 2 1101 12 01010 2 1101 13 01011 2 1101 UHF 01100 10100 0 N.A.

______________________________________

The instructional logic from the PLA decode is indicated in Tables V and VI. In Table V the instruction ROM decode outputs comprising 12 NOR gate outputs is indicated. These 12 outputs provide the "OR" logic functions for the microcomputer program. Table VI indicates the 28 instruction ROM decode outputs of the PLA decode. These outputs comprise 28 "AND" logic functions for the microcomputer programming.
TABLE V
INSTRUCTION
ROM DECODE OUTPUTS
(NOR GATE OUTPUTS)
1. lpc = ubrn + brn s/l
2. μpc enable = rnvm + lnvm + rsax + rsbx + add + suba + subb + rmxa + rmxb + lamx + lbmx + nora + norb + resa + resb + ridb + lidb + wro + wri + 2⁷ bit (of p.c.)
3. ld μpc = rnvm + lnvm + rsax + rsbx + nora + norb + resa + resb
4. data in sel idb = shift cont. in code = rnvm + lnvm
5. shift cont. idb = ridb + rnvm + lnvm + lidb
6. shift cont. ram = rmxa + rmxb + lamx + lbmx
7. ram data restore sel = rmxa + rmxb
8. shift cont. reg a = rmxa + lamx + rsax + add + suba + subb + nora + resa
9. reg a sel restore a = resa + lamx + lidb
10. reg a sel Σ = add + suba + subb
11. shift cont. reg b = rmxb + lbmx + rsbx + add + suba + subb + norb + resb
12. reg b sel restore b = resb + lbmx
table vi
instruction
rom decode outputs
(nand gate outputs)
0. unconditional branch (ubrn)
a. if s/l = 1 or 0 (i.e. DON'T CARE)
1. parallel loads program counter on clk with contents of page latch and 4 lsb's of ubrn code.
2. clears s/l on clk
1. branch (brn)
a. if s/l = 1
1. parallel loads program counter on clk with contents of page latch and 4 lsb's of brn code.
2. clears s/l on clk
b. if s/l = 0
1. do nothing
2. load page (ldp)
a. load 4 bit address code into page latch (4 msb's of address).
3. read rom (rrom)
a. parallel loads reg. b with contents of rom stored at location defined by 4 bit address code and/or channel select code (if vhf).
4. read inputs (rin)
a. enables input defined by 4 bit address to be gates thru to the s/l flip/flop. if input is "0" then s/l is loaded with a "1" on clk. if input is a "1" then s/l is loaded with a "0" on clk.
5. read channel code sw. (rccs)
a. sets input strobe to a "1" level so that the channel select switch can be parallel loaded into the nvm latch buffer on the memory ic.
6. load pwm (lpwm)
a. parallel loads the contents of the input data buffer register into the pwm data register during clk.
7. sense nvm (snvm)
a. sets sense line to "0" so that the contents of the nvm can be parallel loaded into the data buffer registers on the memory ic during clk.
***the following functions enable the μ program counter***
8. read nvm buffer (rnvm)
a. selects 4 bit address code to be loaded into μ program counter.
b. loads 4 bit address (0101) into μ program counter to reset it for 10 right shift functions. load occurs during clk.
c. on clk the program counter is disabled, and the μ program counter is enabled.
d. a right shift command is provided for the input buffer registers and the nvm buffer registers until the μ program counter reaches the 1111 state (10 shifts).
e. when the μpc is 1111 then on the next clock pulse (which sets the μpc to 0000) the clk input to the μpc is disabled, and the clk input to the program counter is restored.
f. note: this command sets the data select line to the memory ic to a logic "1" level, so that data can be sequenced out of the ic into the dac.
9. load nvm buffer (lnvm)
a. same as read nvm buffer except sequence (f) is: note: this command sets the data select line to the memory ic to a logic "0" level, so data can be sequenced into the memory ic from the dac ic.
10. right shift a (r.s.a. x)
11. right shift b (r.s.b. x)
a. selects 4 bit variable address code to be loaded into μ program counter.
b. loads μ program counter during clk.
c. disables program counter on clk.
d. enables μ program counter on clk.
e. enables data shift in (a) or (b) register. data in register (a) or (b) is with leading zero's during μ program count time then right shifted according to the following shift code.

______________________________________

code right shift operations

______________________________________

0000 14 0001 14 0010 13 0011 12 0100 11 0101 10 0110 9 0111 8 1000 7 1001 6 1010 5 1011 4 1100 3 1101 2 1110 1 1111 0

______________________________________

f. when μpc is 1111 THEN ON NEXT CLK PULSE THE μPC CLK INPUT IS DISABLED AND THE PC CLK IS ENABLED.
12-1 add (a & b) (add)
a. program counter is disabled on clk.
b. μpc is enabled on clk.
c. shift controls to reg. a and reg. b are enabled.
d. reg. b restore sel. is enabled.
e. reg a Σ input select is enabled.
f. when μpc is 1111 THE μPC CLK IS DISABLED AND THE PC IS ENABLED.
12-2,3 subtract (b-a) (suba) subtract (a-b) (subb)
a. enables inverter input to adder (e) from reg. a or reg. b.
b. sets carry bit in e₁ to "1".
c. pc is disabled on clk.
d. μpc is enabled on clk.
e. reg a and reg b shift control is enabled.
f. data is shifted serially from reg a and reg b into Σ.
g. data in reg b is restored into reg b without change.
h. Σ data out is stored in reg a.
i. when μpc is 1111 THEN PC IS ENABLED AND μPC IS DISABLED.
13-1 read ram ➝ a (rmxa)
read ram ➝ b (rmxb)
a. address bits (2 bits) select ram storage location.
b. control enables ram read storage gate.
c. control enabled ram ➝ a or ram ➝ b select.
d. pc is disabled on clk.
e. μpc is enables on clk.
f. ram & reg a or reg b shift gates are enabled.
g. data is shifted from ram to reg a or b until μpc is 1111 then μpc clk is disabled and pc clk is enabled.
13-3 load a ➝ ram (lamx) 1,2,3,4
load b ➝ ram (lbmx) 1,2,3,4
a. program counter is disabled on clk.
b. μpc is enabled on clk.
c. address bits (2 bits) select ram storage location.
d. shift controls for selected memory location are enabled.
e. reg a ➝ m sel. and REG A DATA RESTORE.
14-1 normalize a (nora)
14-2 normalize b (norb)
a. selects 4 bit channel encode address to be loaded into μpc (normalized code).
b. loads 4 bit channel encode address into μpc during clk.
c. pc is disabled on clk.
d. μpc is enabled on clk.
e. control enable shift gates of reg. a or b.
f. control sets reg. a and reg. b inputs (serial) to "0".
g. data is shifted in reg. a or b until μpc is 1111 then μpc clk is disabled & pc clk is enabled.
14-3 unused code
14-4 slow clock enable (sloc) switches the clock input to the μpc from t1 clock line to the slow clk line (16ms PERIOD).
15-1 read input data buffer (ridb)
a. enables buffer data select into reg a.
b. enables input buffer data restore select gate into data buffer.
c. enables input data buffer and reg. a data shift control.
d. pc is disabled on clk.
e. μpc is enabled on clk.
f. data is shifted from data buffer into reg. a until μpc is 1111, μpc is disabled and pc is enabled.
15-2 load input data buffer (lidb)
a. enables reg. a data select gate into data buffer reg.
b. enables restore select gate into reg. a.
c. control enables reg. a and INPUT DATA BUFFER SHIFT GATE.
D. pc is disabled on clk.
e. μpc is enabled on clk.
f. data is shifted from reg. a to data buffer until μpc is 1111, then μpc is disabled and pc is enabled.
15-3 clear write (cwro) resets the sense line to the memory ic to a "1"; switching the memory cells from a write mode to a read mode.
15-4 write (wro) sets the sense line to "0" on clk; thereby permitting data to be written into the memory ic cells.
15-5 dummy (unused) unused code used for dummy operations.
the op-codes for the single clock instructions including their address are indicated in Table VII.

TABLE VII

______________________________________

OP-CODES SINGLE CLOCK INSTRUCTIONS FUNCTION OP-CODE ADDRESS

______________________________________

UBRN 0000 1/0 1/0 1/0 1/0 BRN 0001 1/0 1/0 1/0 1/0 LDP 0010 1/0 1/0 1/0 1/0 RROM 0011 1/0 1/0 1/0 1/0 RIN 0100 1/0 1/0 1/0 1/0 RCCS 0101 X X X X LPWM 0110 X X X X SNVM 0111 X X X X

______________________________________

The 4 bit and 6 bit op-codes for the microprogram control instructions including their addresses are indicated in Table VII.

TABLE VIII

______________________________________

OP-CODES MICROPROGRAM CONTROL INSTRUCTIONS OP- FUNCTION CODE ADDRESS

______________________________________

RNVM 1000 1/0 1/0 1/0 1/0 4 BIT LNVM 1001 1/0 1/0 1/0 1/0 OP-CODE RSAX 1010 1/0 1/0 1/0 1/0 RSBX 1011 1/0 1/0 1/0 1/0 LIDB 110000 -- -- 0 0 ADD 110001 -- -- X X SUBA (B-A) 110010 -- -- X X SUBB (A-B) 110011 -- -- X X RMXA 110100 -- -- 1/0 1/0 RMXB 110101 -- -- 1/0 1/0 6 BIT LAMX 110110 -- -- 1/0 1/0 OP-CODE LBMX 110111 -- -- 1/0 1/0 NORA 111000 -- -- X X NORB 111001 -- -- X X UNUSED 111010 -- -- X X SLOC 111011 -- -- 1 1 RIDB 111100 -- -- 0 0 CLR WRO 111101 -- -- 1 1 WRO 111110 -- -- X X DUMMY 111111 -- -- 1 1

______________________________________

The input control line read codes are indicated in Table IX. The input functions each have the same 4 MSB's (0100) and differ only in the 4 LSB's.

TABLE IX

______________________________________

INPUT CONTROL LINE READ CODES RIN CODE INPUT FUNCTION 4 MSB'S 4 LSB'S

______________________________________

UNUSED 0100 0000 CHANNEL INTERRUPT 0100 0001 AFC ON/OFF 0100 0010 UHF ON/OFF 0100 0011 FINE TUNE UP 0100 0100 FINE TUNE DWN 0100 0101 SKIP TOGGLE 0100 0110 UHF/VHF 0100 0111 POWER ON/OFF 0100 1000 UHF UP/DWN 0100 1001 AFC HI 0100 1010 AFC LO 0100 1011 SLOW CLOCK 0100 1100 UNUSED 0100 1101 UNUSED 0100 1110 UNUSED 0100 1111

______________________________________

The ROM constant address codes are indicated in Table X. It is to be noted that the 12 VHF channels are encoded to 12 minimum limits and 12 maximum limits and use the first 24 ROM addresses (00000 to 10111). The remaining 8 words are located in the last 8 ROM addresses (11000 to 11111), thus using a 32 word by 14 bit ROM structure. A greater understanding of the tuning system and the information contained in Tables I-X is gained by referring to the instructions set algorithm.

TABLE X

______________________________________

ROM CONSTANT ADDRESS CODES RROM CODE STORED WORD (14 BITS) 4 MSB'S 4 LSB'S

______________________________________

VHF MIN LIMIT 0011 00XX VHF MAX LIMIT 0011 01XX VHF INCREMENT 0011 1000 TIME/UHF INCREMENT 0011 1001 MAX FINE TUNE TIME 0011 1010 UHF CHANNEL LIMIT 0011 1011 UHF MIN BAND LIMIT 0011 1100 UHF MAX BAND LIMIT 0011 1101 WRITE TIME 0011 1110 MAX UHF COARSE TUNE TIME 0011 1111

______________________________________

The instruction set algorithm for the nonvolatile station memory tuning system as indicated in FIGS. 13A-13L. The algorithm can be divided into a series of four operating modes. The first operating mode comprises the non-tuning mode, FIGS. 13A-13C, the second mode is the start of the AFC off loop which comprises the tuning mode select and initialization FIGS. 13D-13F, the third mode comprises the start of the

 Rocker Tune loop that is the channel fine tuning mode FIGS. 13G-13H, and the fourth loop is the UHF Pot Tune loop comprising the UHF coarse tuning mode FIGS. 13I-13L.

The tuning system is activated by a power up entry 1 in FIG. 13A followed by a load page command (LDP) where a 4 bit address code is loaded into the page latch 294 to address one of the 16 pages in the instruction ROM 286. A clear write (SWRO) operation is performed to reset the sense line to the memory IC to switch the memory cells from a write mode to a read mode. A read channel interrupt loop is performed whereby the system reads the channel interrupt switch until the channel interrupt indication is no longer present.

A channel interrupt indication 224 from the selector switch 202 is applied to the input logic status switch 280. When the input signal matches a 12 bit binary decode 284 the status latch 282 is set. The read channel interrupt loop continuously reads the status latch to determine whether or not it has been set. As long as the status latch has been set from an indication of a channel interrupt the system will loop back into the read mode and will continue until the signal in the status latch is eliminated by the completion of a channel selection at the selector switch 202. The purpose of the channel interrupt read operation is to prevent the system from reading the channel tune word when the selector switch is being changed from one channel to another.
Upon jumping out of the read channel interrupt loop a read channel code switch (RCCS) is performed whereby the five bit address from the selector switch and address generator is parallel loaded into the nonvolatile memory address latch buffer 206 of the memory IC. A slow clock enable signal provides dampening to offset any electromechanical bouncing that may occur during the switching operation. Additional dampening is provided by right shift of the B working register 274 (RSB3) where at a slow clock rate the working register is right-shifted 14 times into itself followed by an unconditional branch command (UBRN) to the next address in the instruction ROM.
The next instruction is to sense the nonvolatile memory (SNVM) where the 15 bit word stored in the memory is parallel loaded into the data out buffer register 210 during a clock pulse. At the same time the 5 bit address from the address latch 206 is parallel loaded into the 5 bit address register 208. Two successive 10 right shift operations are performed by the registers 208 and 210 upon a read nonvolatile memory buffer command (RNVM). Upon completion of the 20 bit right shifts the 14 bit tune word is located in the 14 bit input data buffer register 234, the 15th bit indicating the skip indication is located in the skip toggle buffer 232, and the 5 bit select address code is located in the address buffer 230. The load PWM command (LPWM) parallel loads the 14 bit binary tune word from the 14 bit data latch 238 into the 14 bit PWM logic latch 248. This provides a binary word which is proportional to the analog voltage and sets a binary compare word for the PWM counter that is continuously counting. When the PWM counter reaches the 14 bit binary word that matches it in the 14 bit PWM logic latch, a signal trips the latch and sends the output to the PWM buffer and oscillator 252.

The digital signal is converted to an analog voltage by means of the drive buffer 312, the three-stage PWM filter 314, the tune voltage amplifier 316, and the final PWM filter 318. The analog output comprising the channel tune voltage is sent to the varactor tuner of the television.
If the system power is turned off the 14 bit tune word in the input buffer data register 234 is written into the nonvolatile memory 212. After reading the on/off power switch a 4 bit page address is loaded into the page latch 294 that corresponds to the page of instruction that the system will branch to if the system detects a power off state. In the algorithm this is page 7. The power on/off indication is read into the input logic status switch 280 which if present in the status latch the system branches to the power off write routine 4 in the algorithm at address 70.
The branch statement loads a "0" into the first four bits resulting in a page 7 instruction 0 address. Two successive 10 bit shifts are required to load the 14 bit tune word, the 1 bit skip indication, and the 5 bit select address into the nonvolatile memory. Upon two successive right shift operations the 14 bit tune word and the 1 bit skip indication is loaded into the 15 bit data in register 216 and the 5 bit select address is loaded into the 5 bit shift register 218. A write 0 command (WRO) sets the voltages in the nonvolatile memory for the subsequent write operation. The nonvolatile memory will remain in a write mode until the system is commanded to change by a clear (CWRO) command. The system is programmed for the duration of time it is to remain in the right mode by a read ROM (RROM) command. Fourteen bits of data comprising the right time is read from the ROM constant file 264 and parallel loaded into the 14 bit B working register 274. In the next operation the 14 bit word comprising the write time is serially loaded from the B register into the temporary storage RAM file in the third memory file location by a LBM3 command. This command also restores the data into the working register. The 14 bit binary write time word is now read out of the temporary storage RAM file into the 14 bit A working register 226 by a RM3A command. The 14 bit binary right time word is now stored in working registers A and B.
The instruction ROM 264 is again read by a read ROM command (RROM) whereby a 14 bit word comprising the UHF/time increment is parallel loaded into the B working register 274 from the ROM file. In the next operation the increment value stored in the B working register is subtracted from the magnitude word that is stored in the A working register. In a loop routine the increment value of the B register is subtracted from the decremented magnitude word in the A register until a test condition is reached where the word in the B register is greater than the word in the A register. This condition is detected by reading the status latch 282 as when the condition is satisfied a 1 will be detected. When the 1 is detected in the status latch the system will jump out of the loop and perform a clear write operation (CWRO) which takes the system out of the write mode after the time out operation.
The system now performs a loop routine of reading the channel interrupt and upon either a 0 or 1 indication in the status latch the system will loop back into the read channel interrupt mode. This provides a fixed loop to prevent the system from jumping to another part of the algorithm during a power down routine. When the system is powered up again the algorithm will begin at the 1 power up entry location beginning with a load page command at address 00.
The write routine was entered into by an indication of a power off signal represented by a 1 in the status latch. If the power remains on the system will continue by a reading of the skip toggle input in the logic status switch 280. A load page (LDP) command will load a new page address into the page latch 294. If a skip toggle indication is detected a 1 state will be entered into the status latch in which it will trigger the page latch and enter the new address into the 8 bit program counter 290 that will perform a branch operation to the new page address. In the branch operation a slow clock enable is performed with a second read skip toggle and another load page. If the skip toggle indication has been removed the system will load a new page into the program counter and perform an unconditional branch (UBRN) to 6 of the algorithm at address 12. This branch operation tests the skip toggle to insure that it was not an accidental input. If the skip toggle indication is still present the system will read the skip toggle switch again and perform a looping operation until the skip toggle signal has been removed. When the skip toggle input is removed the program will jump to 3 at address 87 which is the skip toggle write routine. The write operation now performed is identical to the power off write routine at 4 address 70. The nonvolatile memory is sensed by an SNVM command followed by two 10 right shift commands in loading the nonvolatile memory (LNVM). A write command (WRO) sets the voltages for a write operation in the nonvolatile memory matrix. A 14 bit binary word comprising the write time is read into the B working register 274 and is then read out into the temporary storage RAM file in memory location 3. It is then read into the 14 bit A working register 266. The ROM constant file 264 is read again for the UHF/time increment value and is loaded into the 14 bit B working register. In decrementing by subtracting, a loop operation is performed until the contents of the decremented A register is less than the stored increment value in the B register. At this point the system jumps back to page 0 instruction 0 at the power up entry level of the algorithm 1.
If no skip toggle input had been detected in instruction 0 F the program would perform a read channel interrupt command at instruction 12 and a load page operation to page 0. If the channel interrupt has been detected by a 1 in the status latch the algorithm will branch to 1 which is the power up entry position at the beginning of the algorithm at address 00. If no channel interrupt has been detected the system will read the AFC on/off switch at the input logic status switch 280 and branch to 5 at address 04 which comprises a read channel code switch if an AFC on/off indication has been detected. If there is no AFC on/off indication the system will branch to 7 at address 18 which is the start of the AFC loop for the tune mode select and initialization.
With the AFC switch in the off position, the television may be tuned in a manual mode. Since the system is programmed for and has memory storage for only 8 UHF channels and there are a possible 72 UHF channels in existence, the tuning minimum and maximum limits must be set so that they may be utilized with any one of the possible 72 UHF channels. To accomplish this, a UHF channel limit is read from ROM at instruction address 5B and loads the contents into the B working register 274. In the UHF mode fixed channel limits may be set up about any one of the 72 possible channels because the varactor tuning curve has a linear transfer characteristic. This allows for the setting of fixed limits around any desired channel. The tune word from the input data buffer 234 is read into the A working register 266. The UHF tune limit in the B working register is added to the tune word in the A register and the result is stored in the temporary storage RAM file 276 in memory location 1. The original tune word is again read from the input data buffer 234 and stored into the A working register. The UHF channel limit from the B working register is now subtracted from the tune word in the A working register and loaded into the temporary storage RAM file 276 in memory location 0. This sequence of operations now stores in the temporary storage RAM file a lower tune word limit and an upper tune word limit for the particular UHF channel that has been selected.
The UHF/time increment value is read from the ROM at instruction address 62 and loaded into the B working register 274. This 14 bit binary word has a 1 in the LSB position and is used for incrementing the UHF as well as for incrementing time. This incrementor value is loaded into the temporary storage RAM file 276 in memory location 2. The temporary storage RAM file now has in its memory location 0, 1, and 2 the necessary information to perform UHF tuning if necessary. The system now reads whether the selector has been set for a UHF or a VHF channel by reading the input to the input logic status switch 280. The UHF/VHF input is directed from decoding of the 5 bit address at the address decode 260. The page latch 294 is loaded with page E and after the read operation if a UHF signal is detected by a 1 in the status latch the system will continue at address EB where the UHF on/off input is read. If a UHF tuning mode is detected the system after a load page operation will branch to 9 at address 94 which is the start of the UHF pot tune loop in the UHF coarse tune mode. If the UHF is in the off position, then the system will continue at address EE and branch to instruction 22.
This loop is applicable for both VHF and UHF tuning. If at instruction 64 a VHF mode was detected, the system would branch to page 1 instruction B where the VHF increment value would be read from the ROM constant file and loaded into the B working register 274. In the VHF mode each channel has one of four possible increment values. The increment value that has been read into the B working register is normalized by a NOR B command where a right shift operation serially shifts the VHF increment value until the weighted bit value is reached for the VHF channel that has been selected. The VHF increment value is loaded from the B working register into the temporary storage ramp file 276 in memory location 2. The ROM constant file 264 is read and the VHF maximum tune word is loaded into the B working register 274 and then loaded into the temporary storage ramp file in memory location 1. The ROM constant file is read again for the VHF minimum tune word and is read into the B working register and then read into the temporary storage RAM file in memory location 0.
The temporary storage RAM file now has in memory location 1 the VHF maximum tune word, in memory location 2 the weighted VHF increment value, and in memory location 0 the VHF minimum tune word. Each of the twelve VHF channels has a unique minimum tune value and a maximum tune value in contrast to the UHF channel limit.
The system at address 22 now operates in either the UHF or VHF mode. The increment value is read from the memory 2 location of the RAM file into the B working register 274. The tune word in the input data buffer 234 is read into an A working register 266. Before updating the tune word a series of read operations is performed. First the AFC on/off switch is read. If the AFC is at an on state, the program will jump to 2 at address 88 where a normal write routine is performed to write the tune word from the input data buffer into the nonvolatile memory. If the AFC control is in the off state, the system will read the power on/off input. If the power is off, the system will jump to 4 at address 70 for a power off write routine to again write the word in the input data buffer into the nonvolatile memory location. If the power is in the on state, the system will read skip toggle, and if a skip toggle is present will go into a skip toggle loop similar to that previously discussed. If no skip toggle is indicated, the system will read the channel interrupt input and if present will go to 11 at address 30 which is a rocker tune loop. If there is a channel interrupt, the system will go to 2 at address 88 for a normal write routine to write the tune word from the input data buffer 234 into the nonvolatile memory 212.
Address 30 is the start of the rocker tune loop for fine tuning in either the UHF or VHF mode. A read tune-up command is used to determine whether the system is in a tune-up mode or a tune-down mode. In a tune-up mode the system branches to address 4B where the contents of working registers A and B are added together comprising the tune word and the increment value with the result being stored in the input data buffer 234. The maximum tune limit is read out of the temporary storage RAM file into the B working register. A subtract operation at address 4E subtracts the updated incremented tune word from the maximum tune limits. If the resulting operation results in a positive number, the system will branch to address 3E and will load the updated tune word into the PWM logic latch 248. This address is additionally reached by reading the tune-down indicator at address 33 and upon an indication of a tune-down input checking the PWM tune word against the lower tune limit. The PWM logic latch 248 will be loaded where upon subtracting the lower tune limit from the tune word results in a positive number at address 3D.
After loading the PWM, a timing routine is performed. This routine provides a fixed timing sequence for incrementing the tuned word. The present system is designed for eight pulses per second. However, other timing sequences may be employed; for example, 16 pulses per second or 32 pulses per second. The timing varies as a function of how fast it is desired to perform the updating of the tune word in the fine tuning mode.
The maximum time limit is read from the ROM constant file into the B working register 274 and additionally loaded into the temporary storage RAM file 276 in memory location 3. The ROM constant file is read again for the UHF/time incrementing value and loaded into the B working register. Memory location 3 containing the maximum time limit is read into the A working register 266. Subtracting the contents of the B working register from the A working register is performed in a loop routine until register A has been decremented to a negative number. At this point the system will jump to 8 at address 22 where the increment value is again read from the temporary storage RAM file into the B working register and reading the latest updated tune words from the input data buffer 234 into the A working register. The A working register now contains the latest incremented or decremented tune word. The system would also jump out of the loop during the timing cycle if a channel interrupt indication was present at address 4A where the system would jump to 2 at address 88 to perform a normal write routine of the updated tune word into the nonvolatile memory.
The program will continue through the incrementing or decrementing of the tune word. In addition in the tune-up mode at address 50 the updated tune word is compared with the upper limit and in the tune-down mode at address 3D the updated tune work is compared with the lower tune limit. At these addresses if the system has gone beyond the upper tune limit or gone beyond the lower tune limit, a branch operation will be performed to read the tune-up input at address 53 or the tune-down input at 57 depending upon whether the system has been operating in a tune-up or tune-down mode. In either case, the system will unconditionally branch to 8 at address 22.
During the sequence of operation the tune-up or tune-down indication is read twice. The purpose of the double read operation is to change the function of the input switch. Where the system was reading up we want the system to read down and where the system was reading down before we want the system to read up. This provides for an automatic reversing of direction during the AFC/off tuning of the TV. If during the AFC/off tuning in either the up mode or down mode exceeds the upper or lower limits, the system will automatically reverse direction and tune in the other direction.
If during a tuning mode the UHF tune switch is read in the on state at address EB, the system at address ED will branch to 9 which is the beginning of the UHF pot tune loop at address 94 at the beginning of the UHF coarse tuning mode. The ROM constant file is read for the UHF/time increment value and loaded into the B working register 274 and into the temporary storage RAM file 276 in the memory location 0. The ROM constant file is read again for the maximum time value which is loaded into the B working register and into the temporary storage RAM file at memory location 3. The ROM constant file is read again for the UHF/time increment value which is loaded into the B working register and into memory location 2 of the temporary storage RAM file.
In the UHF tuning mode a different rate of tuning is programmed into the system. The system starts off at a slow tuning rate then accelerates the tuning rate as a function of time to allow tuning from one end of the tuning band to the other end of the tuning band within a reasonable time yet allow for the contingency of tuning initially at a slow rate if the desired tune voltage is close to the starting point. The UHF/time increment value is read from memory location 2 of the temporary storage RAM file into the A working register and also the UHF/time increment value is additionally read from the ROM constant file into the B working register. These two UHF/time increment values in the A and B working register are added together and the results stored in memory location 2 of the temporary storage RAM file. The maximum time for UHF tuning is read from memory location 3 of the temporary storage RAM file into the B working register. After a subtract operation of the updated UHF/time increment value from the maximum tune time in the B working register, a test operation at address A 1 is performed. If the maximum time value is greater than the lapsed time, the system will read the input data buffer register 234 which contains the tune word that is going to be updated into the A working register. The input data buffer acts as a temporary storage file during the tuning operation. In addition, the UHF/time increment value is read out of memory location 0 of the temporary storage RAM file into the B working register. The UHF/time increment value has a 1 in the LSB position and zeros in all the other bit positions. This is used to increment the tune word by 1 in the LSB position. The UHF up/down control is read to determine the direction of the UHF tuning.
The UHF up/down input is fed from a comparator 282. The output of the comparator is compared with the actual tune voltage in the system. If the system is tuning in an up mode at address 86, the system will perform an add function which will add the contents of the B working register with the A working register and load it into the input data buffer 234 which will now contain an incremented UHF tune word. The UHF maximum tune value is read from the ROM constant file into the B working register and the updated tune word in the A working register is subtracted from it to determine if the updated tuning word has exceeded the UHF tuning limit. At this point it would be desired to stop the tuning in the up mode and reverse direction. If at address BF the test condition indicates that the maximum UHF tuning limit has been reached, the system will branch to 5 at address 04 which is a non-tuning mode.
If the maximum UHF tuning limit has not been reached, the system at address C0 will load the updated UHF tune word into the PWM logic latch 248. The next operation performed is a time out for UHF tuning. The maximum time value is read from the ROM constant into the B working register and read into memory location 1 of the temporary storage RAM file. The UHF/time increment value is read from the ROM and stored in the B working register. The maximum time value is read from the memory 1 of the temporary storage RAM file into the A working register 266. The UHF/time increment value from the B working register is subtracted from the maximum time value in the A working register. A test loop routine is provided at address C5 to continue the subtract operation of the UHF/time increment value from the maximum time until it is decremented to a negative number.
During the loop operation of the time out the channel interrupt select switch is read for a positive input whereby the system would branch to 2 at address 88 which is a normal write routine. If no channel interrupt is detected, the system will continue looping until a negative value is reached during the subtract operation, and the system will branch to 14 at address E6 which is the start of the pot up loop.
In an identical routine the system will perform UHF tuning in the down mode at address A7. Here the system checks the PWM word against the minimum UHF tuning limit that has been read from ROM and stored in working register B at address A9.
A time out routine is provided at address AE identical to the time out routine performed during the uptuning in the UHF mode. At the completion of the time out for the UHF tuning in the down mode, the system will branch to 15 at address DB which is the start of the pot down loop.
At address E6 the system will read the UHF up/down switch to determine whether the system is tuning in the up mode or down mode. If the system has detected that it is tuning in a down mode, it will branch to 7 at address 18 which is the start of the AFC off loop. If the system is continuing the UHF tuning in the up mode, the system will continue by reading the AFC on/off switch and branch to 2 at address 8A for normal write routine if the AFC is in the on position. If the AFC is in the off position the system will read the power on/off switch and branch to 4 at address 70 for a power off write routine if the power has been turned off. If the power has remained on, the system will branch to 13 at address 9A which is the start of the lapse time counter.
In a similar manner at address DB for the pot down loop the system will read the UHF up/down switch and if an up tuning mode is detected, the system will branch to 7 at address 18. If the system is continuing to tune in the down mode, then the system will continue with the sequence at address DE.
On completion of this loop in branching back to address 9A a complete cycle of updating the tune word has been completed for either the UHF up tuning of UHF downtuning.
In continuing a second loop at address 9A the UHF/time increment value will be incremented by its own value to provide a new UHF/time increment value which is double the original value. This new UHF/time increment value will be added to the previously updated UHF tune word which is stored in the input data buffer 234. Prior to the addition a test condition will determine whether or not the updated UHF/time increment value has exceeded the UHF maximum time value. If the UHF maximum time value has not been reached, the system will continue updating the tune word in a closed loop routine starting at address 9A. The updated tune word will be incremented by an increasing UHF/time increment value during each loop until the maximum time value has been reached. When the UHF maximum time limit has been reached at the test operation at address A1, the system will branch to address CD. This new loop routine provides a continuous incrementing or decrementing of the UHF tune word first at a slow rate and then upon each successive loop at an increased rate by increasing the value of the UHF/time increment value in the ROM constant file.
Once the UHF maximum time limit has been reached, the system will change the UHF tuning rate beginning as address CD. The subsequent routine provides a high speed bit weight update of the UHF/time increment value for increasing the rate of tuning. In addition, the maximum tuning time value is also increased. The UHF/time increment value is read from memory location 0 of the temporary storage RAM file into the A working register. At address CD an RSA 14 operation is performed. This function performs a right shift operation of the A working register by 1 bit. The UHF/time increment value is read from memory location 0 of the temporary storage RAM file into the B working register and is also right shifted by 1 bit. The contents of the A and B working registers are added together, and the result stored in memory location 0 of the temporary storage RAM file. This additional operation increases the UHF/time increment value by a factor of 2. This operation now provides a high speed bit weight update by employing the increased UHF/time increment value.
The maximum time value is read from memory location 3 of the temporary storage RAM file into the A working register. Working registers A and working registers B are right shifted 14 bits with a restore operation. The content of the A and B working registers are added together and stored in memory location 3 of the temporary storage RAM file. This addition of the two working registers results in a doubling of the UHF maximum time limit. Upon completion of the doubling of the UHF/time increment value and the UHF maximum time value, the system branches back to 12 at address 98 of the program. At this point tuning will be at double the UHF/time increment value and for double the maximum tune time.
The system will continue tuning through the program by testing for an up or down tuning direction at address 86 and checking the PWM word against the maximum UHF tuning limit during an up tuning mode and checking the PWM word against the minimum UHF tuning limit in a down tuning mode. If neither limits have been exceeded, the system will load the PWM data latch with the updated tuning word and perform a time out operation followed by a branch to either the pot up loop for up tuning or pot down loop for down tuning. In either case the loop will detect a change in tuning direction and if detected branch to 7 at address 18. If the system has not received a change in direction indication, it will continue by reading the AFC on/off and power on/off controls with a branch to 13 at address 98 if neither of the controls have been activated.
At this point the system will continue the loop routine through the lapse time counter and incrementing or decrementing the updated tune word until the lapse time counter has been decremented to a negative number where the system will branch to address CD. In this new loop the UHF/time increment value and the maximum tune time value will again be doubled, that is quadruple the original values stored in the ROM, and the system will continue tuning by branching to 12 at address 98. The sequence will be as previously described where the lapsed time counter routine will be performed with a new UHF/time increment value and maximum time value which will be double the previous value. This provides for a high speed bit weight update in the tuning of the UHF channel.
A binary word corresponding to each of the individual VHF and UHF channels selected may be programmed into the nonvolatile memory matrix by the manufacturer of the system prior to sale. The binary tune word would correspond to the nominal tune voltage of the corresponding channel. In the alternative, however, the manufacturer may allow the ultimate user to perform this function. In this instance the user would initially go through the tuning mode for each channel selected until the tune voltage is arrived at that satisfies the user's requirement. At this point the tune word would be stored into the nonvolatile memory by one of the write modes defined in the instruction set.
Whereas the present invention has been described with respect to a specific embodiment thereof, it will be understood that various changes and modifications will be suggested to one skilled in the art, and it is intended to encompass such changes and modifications as fall within the scope of the appended claims.

MATRIX + RGB AMPLIFIERS (TDA2800)


















- NF Baustein / Audio Amplifier: 29301-004.01 with TBA800

- Horizontal-Baustein / Line Osc + Synch : 29301-008.02 with TBA920

- Sicherung Bst / Safety Unit 29301-410.01

- OW Dioden Baustein / E/W Correction diode Mod. Unit: 29301-041.01

- Regelbaustein / regulator unit: 20301-035.01

- Thyristor heatsink with RCA17052 AND RCA17051


TBA920 line oscillator combination

DESCRIPTION
The line oscillator combination TBA920 is a monolithic
integrated circuit intended for the horizontal deflection of the black and white
and colour TV sets
picture tube.

FEATURES:
SYNC-PULSE SEPARATION
OPTIONAL NOISE INVERSION
GENERATION OF A LINE FREQUENCY VOL-
TAGE BY MEANS OF AN OSCILLATOR
PHASE COMPARISON BETWEEN SYNC-
PULSE AND THE OSCILLATOR WAVEFORM
PHASE COMPARISON BETWEEN THE OS-
CILLATOR WAVEFORM AND THE MIDDLE OF
THE LINE FLY-BACK PULSE
AUTOMATIC SWITCHING OF THE VARIABLE
TRANSCONDUCTANCE AND THE VARIABLE
TIME CONSTANT TO ACHIEVE NOISE SUP-
PRESSION AND, BY SWITCHING OFF, POS-
SIBILITY OF TAPE-VIDEO-REGISTERED RE-
PRODUCTION
SHAPING AND AMPLIFICATION OF THE OS-
CILLATOR WAVEFORM TO OBTAIN PULSES
FOR THE CONTROL OF DRIVING STAGES IN
HORIZONTAL, DEFLECTION CIRCUITS
USING EITHER TRANSISTORS OR THYRISTORS



In the future they developed a CHASSIS Called CUC (Compact Universal Chassis which wasn't very compact anyway). The CUC is entirely based on Transistor (BU208A) technology in all the power stages.

With the coming of the CUC Chassis, GRUNDIG abandoned completely the Thyristor ERA !!!!

(All the successive chassis were calle CUC followed by a code version type number).

TDA2521 synchronous demodulator for PAL

GENERAL DESCRIPTION
The TDA2521 is a monolithic integrated circuit designed as a synchronous demodulator for PAL color television receivers. It includes an 8.8 MHz oscillator and divider, to generate two 4.4 MHz reference signals, and provides color difference output.
The TDA2521 is intended to interface directly with the TDA251O with a minimum of external components and is constructed on a single silicon chip using the Fairchild Planar
epitaxial process.

ABSOLUTE MAXIMUM RATINGS
Supply Voltage 14 V
Internal Power Dissipation 600 mW ORDER INFQRMATIQN
Operating Temperature Range —2O°C to +6O°C TYPE PART NO.
Storage Temperature Range —55°C to +125°C 2521 TDA2521
Pin Temperature iSo|dering 10 si 260°C

Planar is a patented Fairchild process













TDA2510 CHROMINANCE COMBINATION

GENERAL DESCRIPTION —
The TDA2510 is a monolithic integrated circuit designed for the function of a color television receiver. It Is designed to Interface directly with the TDA2521, using a minimum number of external components.
TDA251O is constructed on a single silicon chip using the Fairchild Planar‘ epitaxial process.

ABSOLUTE MAXIMUM RATINGS

supply Voltage 15 V
Collector voltage of chroma output transistor (pin 7) 20 V
(PD I 100 mW max)
Collector current of chroma output transistor (pin 7) 20 mA
Collector current of color killer output transistor (pin 11) 10 mA
Power dissipation 500 mW
Operating temperature range —25°C 10 +6O°
Storage temperature range *55°C to +12!-3°C