CRT TUBE TELEFUNKEN A66-140X.
DEFLECTION UNIT TELEFUNKEN AS110K213 
As is well known to those skilled in this art, the delta-gun, color cathode ray tube (CRT) differs most significantly from conventional CRT's having a single gun in that the delta-gun has three guns, each of which generate a distinct electron beam for exciting its corresponding phosphor dot of the red, green and blue dot triad matrix constituting the CRT display surface.
Each of the delta-arranged gun beams must land at precisely the same opening in the shadow mask at every point on the screen in order to produce a perfect rendition of a desired color and symbol shape. To achieve the desired rendition on the screen, the beams, substantially forward of where they leave the guns, are magnetically deflected, as a set, by excitation of a main deflection yoke comprising an electromagnet arrangement mounted around the neck of the CRT so as to control the angle at which the beams, as a set, approach the desired landing site on the display screen.
Registration of all three beams at the landing site cannot be accomplished through all deflection angles without some individual deflection of each beam in addition to the main deflection. The auxiliary deflection system is termed convergence deflection because of the function it performs in causing each beam to converge to a single point for all deflected angles.
Aside from the geometry errors (pin cushion effect) which necessitate a convergence system, there are errors due primarily to alignment tolerances associated with the deflection yoke and electron gun assemblies. These errors have the effect of distorting the ideal convergence function and consequently, each CRT assembly must be uniquely converged.
In most, if not all delta-gun CRT's, convergence deflection is accomplished by electrical current excitation of a small deflection yoke for each yoke is usually part of the electron gun assembly .
Each of these yokes produce a magnetic field which is applied to the beam emanating from their respective gun in such a way that deflection of the beam is along a single axis, that is, the axis which extends radially from the center of the CRT, through the center of each of the three guns. These axes are called red radial, green radial and blue radial, for a typical CRT and are then, by the nature of the delta-gun, 120 angular degrees apart.
It is customary to align the blue radial with the vertical deflection of the main deflection yoke. Thus, as facing the front of the CRT, the blue gun converges along a vertical axis, red along an axis 120° left and green along an axis 120° to the right (240° to the left).
The full function of convergence is not possible without a fourth degree of freedom of movement associated with one of the electron beams. This is generally applied to the blue gun along the lateral axis of the CRT at 90° to the blue radial and is consequently termed blue lateral convergence.
As is the case with radial convergence, the blue lateral convergence can be partitioned into two components, a dynamic portion and a static, or constant portion. The cynamic portion of blue lateral correction (unlike that of the dynamic portion of the radial correction) can be, in general, performed by the design of the main deflection yoke. When this is the case, only an adjustable external static magnet is required to position the blue beam laterally to a reference location at the center of the screen. This magnetic adjustment is termed the static blue lateral adjustment.
CRT TUBE TELEFUNKEN A66-140X Assembly and adjustments
 The following sequence must be followed in the adjustments.
 
 a)
 Deflection yoke, radial converging assembly, lateral converging device and purifying magnet must be fitted on the tube neck in the prescribed position.
 
 b)
 Apply supply voltages and adjust focusing to Optimum.
 
 c)
 Demagnetize the tube for several seconds. Thus local colour impurity zones will be eliminated, which are due to magnetization of the metal shield and the mask. Take care with magnetized chassis or other iron parts. 
 
 d)
 Adjust the static convergence. Using a cross-hatch or dot pattern the electron beams must converge in a dot at screen centre. This adjustment is performed with the aid of radial permanent or electromagnets on the radial converging assembly and the lateral converging device. If the blue beam is blocked, Optimum convergence is obtained once the pattern or dots in the screen centre are yellow. Subsequently the blue beam must be applied and brought horizontally to the side of the yellow dot using the corresponding radial magnet, and then brought to coincidence with it adjusting the lateral converging magnet.
 
 e)
 Adjust colour purity. With a red area raster switched on and the deflection unit placed as for back as possible (12 mm) the field of the colour purifying magnet must be adjusted as to intensity and direction until as uniform as possible c red area appears at screen centre. (Optimum beam centering to red phosphor dots in screen centre.)
 
)
Push the deflection unit forwards until the entire screen is illuminated with o uniform red. Subsequently the colour purity of the green and blue rasters must be checked Over the whole screen and, if necessary, a compromise must be found in the adjustment for all colours. Prior to, and following, each colour purity adjustment the static convergence must be checked. Centre a test pattern by means of DC pre deflection. Check static convergence as well as colour purity and readjust if necessary.
Adjustment of dynamic convergence.
Use a bright cross-hatch or dot pattern for this purpose. By adjusting the alternating currents in the converging coils the three coloured pattern must be brought to coincidence Over the entire screen (without the Corners) in such a manner that white dots or pattern lines are produced. After blocking the blue beam adjust the red a n d green roster to coincidence. During this procedure the static convergence must be readjusted and the colour purity checked several times. By regulating the corner convergence currents in the deflection coils raster coverage is performed in the corners as next Step. In conclusion pincushion distortion must be eliminated. Care must be token that at no time o considerable amount of beam current reaches the walls of the tube or ports of the electron gun.
Adjusting colour purity with o microscope
Observation of register through a microscope (magnification approx. 40) is recommended for  colour purity adjustment. For this purpose a white raster is used. The phosphor dots must be illuminated from the side with a light source (e. g. lamp). In the screen centre the beam position is influenced by means of the purifying magnet in such manner that the centres of the two triangles, which are formed by a phosphor triplet and the excited surfaces, are in coincidence. Afterwards the convergence must be checked and readjusted if necessary, the colour purity likewise being corrected once more. After switching to red raster o uniform red screen surface must be adjusted as described in e) by sliding the deflecting yoke axially.
Purifying magnet
Permanent magnet with field perpendicular to tube axis. The intensity and direction of the field must be adjustable. Lateral converging device Permanent- or electromagnet with field perpendicular to tube axis, whose direction causes a horizontal movement of the blue beam in opposite direction to red and green beams. The field intensity must be adjustable. Radial converging assembly Permanent magnets or electromagnet fed with DC having a field perpendicular to tube axis are used to adjust static convergence. The field intensity must be adjustable. The dynamic convergence is achieved by alternating magnetic fields, which are obtained through superposed AC currents in the coils attached to the permanent- or electromagnets.
Deflection unit
The axes of the deflection unit and tube must coincide. The deflection unit has to be movable along the neck for a distance of min. 12 mm and a slight turning has to be allowed. Beam centering is effected by DC exclusively, pincushion distortion correction exclusively by superposing AC. 
   METHOD AND APPARATUS FOR STATIC AND DYNAMIC CONVERGENCE IN A DELTA GUN  CRT COLOR TUBE:
The   three electron beams in a color  television cathode ray tube are  
  statically converged at the beginning  and end of vertical and 
horizontal    scan lines, and are dynamically  converged in the center 
of the  raster.   Each external convergence  assembly, mounted adjacent a
 pair of   internal  pole pieces for a  corresponding electron beam, 
includes a   single coil  passing both  vertical and horizontal 
convergence currents   which have  been combined  by a semiconductor 
circuit. The convergence   coil for the  blue beam  is split into two 
sections which are separately   excited to   independently control 
lateral movement on o
pposite
    sides of the raster.  The convergence method uses active circuits,  
and   simplified passive  circuits which require a reduced number of  
input   waveforms from the  receiver scanning stages, such as the  
waveform   across the S shaping  capacitor.
1. In a color  television   receiver having a cathode ray tube with  
plural electron  beams and   scanning means for deflecting said plural  
electron beams to  produce   scanning lines, a convergence system,  
comprising:                                          2. The convergence 
system of claim 1  wherein said   static  convergence means statically 
converges said  plural electron   beams at  opposite edges of said 
scanning lines, said  dynamic   convergence means  causes said plural 
electron beams when not  coincident   to have an arcuate  path with a 
maximum deviation from a  straight path   in the vicinity of  said 
center portion.                                          3. The 
convergence system of claim 2 wherein  said adjustable   means  includes
 vertical line adjustable means for  laterally moving   vertical  
scanning lines and horizontal line  adjustable means for   laterally 
moving  horizontal scanning lines, the  pair of line adjustable   means 
producing  maximum correction deflection  during said center   portion 
of scan and  progressively lesser amounts  of correction   deflection in
 opposite  directions away from said center  portion of   scan.         
                               4. The  convergence system   of claim 1 
wherein said scanning means  comprises  vertical deflection   means for 
producing vertical signals  synchronized  with the vertical   deflection
 of said electron beams and  horizontal  deflection means for   
producing horizontal signals  synchronized with  the horizontal   
deflection of said electron beams, said  dynamic  convergence means   
including vertical generator means having an  input  coupled to said   
vertical deflection means for converting the  vertical  signals into a  
 generally parabolic waveform having a maximum   deviation at the center
   of each vertical scanning line and vertical   adjustable means for   
varying the amount of said maximum deviation,   horizontal generator   
means having an input coupled to said horizontal   deflection means for 
  converting the horizontal signals into a  generally  parabolic 
waveform   having a maximum deviation at the center  of each  horizontal
 scanning   line and horizontal adjustable means for  varying the  
amount of said   maximum deviation, and exciter means  responsive to 
said  parabolic   waveforms for producing corresponding  deflections in 
the  scanning   lines.                                        5. The 
convergence system   of claim 4 wherein at least one of said   generator
 means includes a pair   of active devices each having an  input  and an
 output and connected as   amplifiers, input means coupled  to the  
inputs for driving said pair  of  devices oppositely by a  generally  
sawtooth shaped waveform  occurring in  synchronism with said  
synchronized  signal, and output  means coupled to  the outputs of said 
 pair of devices  for deriving said  generally  parabolic waveform.     
                                     6. The  convergence system of claim
 5  wherein said amplifiers  include   capacitive feedback means coupled
  between the output and the  input of   said pair of devices to cause  
said devices to convert the  generally   sawtooth shaped waveform into  
the generally parabolic  waveform.                                      
    7. The convergence system of claim 4   wherein said exciter means   
comprises a convergence assembly external to   said cathode ray tube and
   adjacent one of said plural electron beams   for applying convergence
   correction thereto, said assembly includes a   coil for generating a 
  magnetic flux field which produces said   correction, and said dynamic
   convergence means includes mixer means for   combining the generally 
  parabolic waveform from said vertical   generator means and the  
generally  parabolic waveform from said   horizontal generator means to 
 produce a  combined convergence waveform   coupled to said coil.       
                                  8. The   converge
nce
  system of claim 1 wherein the cathode ray tube  has two   internal 
pole  pieces defining an axis equidistant therebetween  through   which 
 passes one of said plural electron beams, said dynamic    convergence  
means includes core means having a first leg adjacent one of    said two
  pole pieces and a second leg adjacent the other of said two    pole  
pieces, convergence coil means wound about said core means for     
generating a magnetic flux field which produces the deflection of the   
  scanning lines in the vicinity of the center portion, and unbalance   
  means for producing a more concentrated magnetic flux field in one of 
    said legs to cause said electron beam to be deflected at a skew with
     respect to the equidistant axis.                                   
       9. The convergence system of claim 8 wherein said unbalance means
     comprises means mounting said coil means about only one of said 
first     and second legs.                                       10. The
    convergence system of claim 1 wherein said dynamic  convergence 
means    includes initial convergence means for producing an  initial 
correction    waveform occurring during an initial portion of a  
scanning period and    final convergence means for producing a final  
correction waveform    occurring during a final portion of a scanning  
period, a convergence    coil for one of said electron beams having a 
first  winding section and a    second winding section, initial 
adjustable means  coupled to said    initial convergence means for 
adjustably dividing said  final correction    waveform between said 
first and second winding  sections.                                     
     11. The convergence system of claim 10   wherein  each of said  
adjustable means comprises a potentiometer having   a pair  of end  
terminals with a fixed resistance located therebetween   and a  wiper  
movable across said fixed resistance, means separately   coupling  the 
end  terminals to said first and second winding sections,   and said  
wiper  being coupled to the corresponding convergence means.            
                               12. In a color television   receiver 
having a  cathode ray tube with  at least one electron beam   which is 
angularly  deflected both vertically  and horizontally by   vertical 
deflection means  and horizontal deflection  means,   respectively, a 
dynamic convergence  system, comprising:                                
          13. The  dynamic convergence system of claim 12   wherein said
  combining means  includes a summing junction for  combining  the 
horizontal  and vertical  correction signals, and active  means  coupled
 between said  summing  junction and said single coil.                  
                       14.  The dynamic convergence system  of claim  12
 in which said  cathode ray  tube includes a second electron  beam  
which is angularly  deflected both  vertically and horizontally  by said
  vertical deflection  means and said  horizontal deflection  means,  
respectively, a second  convergence  assembly external to said  cathode 
 ray tube and adjacent said  second  electron beam and including  second
  core means having a pair of  second  core legs and a second  single 
coil  wound about only one of said  pair of  second core legs for  
generating a  magnetic flux field which  produces  convergence  
correction for said  second electron beam, said  combining  means  
includes a summing junction  for combining the horizontal  and  vertical
  correction signals, and  divider means coupled between said   summing 
 junction and said first and  second coils for adjusting the   ratio of 
 the combined signals which  are impressed on said first and   second  
coils.                                        15. 
Dynamic
   convergence correction is thus required at the   beginning and end of
  the  horizontal and vertical scan periods. Thus,   the maximum 
amplitude  of  convergence correction current occurs at the   edges of 
the CRT  screen.  Since the convergence yoke is a current   integrator, 
and the  rate of  change of current level with time causes a   
deterioration in  the current  waveform, this prior approach creates   
problems. Since the  convergence  current typically increases at a   
parabolic rate, the  integration effect  degrades convergence   
performance. Furthermore, the  convergence exciter  assembly must also  
 perform as a transducer in  that it must generate a  magnetic field   
proportional to the drive  currents in its exciter coils.  This creates a
   considerable problem  since a ferrite core is a poor  magnetic  
conductor  at high flux  densities which are required to deflect  the  
electron beam  for proper  convergence at the edge of the picture  tube.
In   prior systems,  maximum correction for dynamic  convergence is 
required   at the edges  of the picture tube. If a  considerable amount 
of   correction is needed  at the right side of the  tube, for example, 
and   only a small amount  of correction is needed at  the left side, 
the   convergence exciter  assembly is forced to change  radically in 
output   during the retrace  period. The memory of the ferrite  core 
combined with   the current  integration effect of the exciter coil  
often prevents  this  rapid  change from taking place. In order to make 
 such a rapid  change,  an  extreme amount of driving power may be  
necessary. This  requires not   only high currents, but also forces an  
interrelationship  between the   convergence controls on opposite sides 
of  the tube. Thus a  change in   convergence level at the right side of
 a  picture tube may  affect the   convergence effect on the left side. 
Such an   interrelationship adds   many steps to the convergence process
 since   correcting one area may   degrade another area.
The prior art  has  recognized in general   that initial beam 
convergence may be at any   desired deflected  position,  and thus the 
beams could be initially   converged at one  corner of the  raster. Such
 a general recognition is   contained, for  example, in U.S.  Pat. No. 
3,048,740 to Nelson, issued   Aug. 7, 1962,  and entitled  "Electron 
Beam Convergence Apparatus."   However, no  apparatus has been  provided
 for effecting static  convergence  at the  edges, and dynamic  
convergence at the center of a  scan, nor has  the  unexpected 
advantages  of such a method of  convergence been   recognized, as 
discussed in the  following section.
Generally,    the convergence exciter assembly  for each beam has 
included a   horizontal  convergence coil, and a  separate vertical 
convergence coil   which  usually was of a greater  number of turns. A 
parabolic or  similar   non-linear current signal for  coupling to the 
horizontal  convergence   coil is developed by a  horizontal convergence
 circuit,  and similarly a   vertical convergence  circuit develops a 
parabolic or  similar  non-linear  current for coupling  to the separate
 vertical  convergence  coil. In  addition to requiring  two pairs of 
coils for  each convergence  exciter  assembly, this  arrangement has 
the further  disadvantage of  increasing  the length of  the ferrite 
core, thereby  increasing the  overall  dimensions of the  convergence 
yoke assembly.  The physical size  of the  yoke is an  important 
consideration in  commercial television  receivers  since the  amount of
 space available  for CRT neck components  is of a  limited  nature. The
 size of the  ferrite core is also  controlled by the  amount  of power 
necessary to  effect convergence.  Both of these  considerations  have 
resulted in a  convergence yoke  assembly of  substantial size.
  Independent  lateral adjustment of  the blue  beam over the entire  
length of the  raster has not been  possible. In a  typical convergence 
 procedure, a  single dynamic control  adjusts the  lateral positions of
  the blue  vertical lines. If the  vertical blue lines  are outside of 
 the converged  red and green  vertical lines, which  condition is known
  as a wide blue  field, then a  known adjustment moves  the blue lines 
 inward on both  sides of the  raster in order to converge  with the red
  and green  vertical lines.  Conversely, if the blue vertical  lines 
are  inside the  converged red  and green vertical lines, which  
condition is  known as a  narrow blue  field, a known adjustment 
laterally  moves the  lines into  coincidence.  However, if one half of 
the raster has  a  wide blue field  while the  other half exhibits a 
narrow blue field, it   has been  necessary to  reject the cathode ray 
tube or deflection yoke   since known  convergence  apparatus have not 
been capable of providing   this type of  correction.
SUMMARY OF THE INVENTION
In    accordance with  the present invention, the problems noted above 
with    prior convergence  methods and apparatus have been overcome. 
Initial    static convergence  correction for both the vertical and 
horizontal    scanning lines is  effectuated at the edges of the picture
 tube.    Thereafter, dynamic  convergence correction is applied to 
cause the   beams  to coincide in the  center of the picture tube. Using
 this   method, the  convergence exciter  assembly output is never 
forced to   make a radical  change during the  retrace period. Also, 
with dynamic   convergence applied  in the center of  the picture tube, a
 considerable   time period is  provided for the  correction current to 
build up in the   exciter coils. As  a result, the  convergence system 
requires much  lower  energy levels. A  further  advantage is the 
elimination of  interaction  between convergence   controls for the 
right and left side  of the  picture tube, simplifying   the convergence
 procedure.
Unlike   many prior convergence   systems, it is unnecessary to utilize 
in the   exciter assembly separate   coils for the vertical and 
horizontal   convergence functions. A reduced   number of parabolic or 
similar   non-linear waveform generators supply   currents to matrixing 
circuitry   which feed a single combined output to a   common coil 
associated with   each color beam, which common coil can be   mounted on
 only one of two   core legs of a generally U-shaped  convergence  core.
 The circuitry   includes separate excitation and  adjustment for two  
sections of the   blue coil, allowing independent  lateral control over 
 the beginning and   end of a horizontal scanning  period. Thus, a CRT 
and  deflection yoke   which exhibits a wide blue  field on one half of 
the  picture tube,  and a  narrow blue field on the  other half, can be 
 corrected. The  circuitry  necessary to implement the  present 
invention  can be either  active or  passive.
A comparison  between the  applicant's  convergence  method and 
apparatus and prior  convergence  systems  reveals a  substantial 
savings in components and a  substantial   simplification in  the 
convergence procedure. For example,  for a prior   color television  
receiver which used a conventional  convergence  system,  40 steps were 
 required in the alignment procedure.  With the  applicant's  present  
design, only 18 steps were required. A   substantial savings can  also 
be  effected in the design of a  convergence  exciter assembly. A  prior
  design required a total of 12  coils and 12  bobbins, whereas the   
present apparatus permits a total  of four coils  and four bobbins. This
   represents a significant savings  in material and  labor costs, and 
also   significantly reduces the  physical size of the  convergence 
yoke.
One   object of the  present invention is the  provision of an improved 
  convergence method  and apparatus for  statically converging electron 
  beams at the edges  of a scan line and  for dynamically converging the
   beams at the center  of scan.
Another  object of the present   invention is the  provision of a 
convergence  exciter assembly which   includes a single  coil on a 
single core leg of a  U-shaped core  assembly,  which coil is  coupled 
to matrixing circuitry  which combines  horizontal  correction  current 
and vertical correction  current.
A  further  object of  the present invention is the  provision of  
independent  convergence  controls for separately correcting  the 
lateral  position of a  vertical  line for initial and final  horizontal
 scanning  periods,  allowing for  example a CRT and deflection  yoke 
having a wide  blue field  and a  narrow blue field on opposite  halves 
of a raster to  be corrected.
Yet  another object of this  invention is the  provision of  improved  
convergence methods and  apparatus which are  applicable to both  active
  and passive circuits, and  which reduce the  drive requirements and   
the number of waveform  generators which must be  provided. The input   
waveform can be taken from  across the S shaping  capacitor in series  
with  the yoke coil.
Further  features and  advantages of the   invention will be apparent 
from the  following  description and from the   drawings. While 
illustrative  embodiments of  the invention are shown  in  the drawings 
and will be  described in  detail herein, the invention  is  susceptible
 of embodiment  in many  different forms and it should  be  understood 
that the present   disclosure is to be considered as an   
exemplification of the principles   of the invention and is not intended
   to limit the invention to the   embodiments illustrated.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG.     1 is a block diagram of one embodiment of the applicant's  
convergence    apparatus as incorporated in a conventional color  
television  receiver;
FIG.   2 illustrates the face of a picture  tube after  vertical and 
horizontal   static convergence has been  completed, and  depicts the 
dynamic  vertical  correction provided by  the present  invention;
FIG. 3  shows the  face of the picture  tube of FIG. 2  and depicts the 
dynamic  horizontal  correction provided  by the present  invention;
FIG. 4   illustrates the face of a  picture tube after  vertical and 
horizontal   static convergence has  been completed, and  depicts the 
dynamic vertical   correction provided  heretofore by the  prior art;
FIG. 5 shows   the face of the  picture tube of FIG. 4  and depicts the 
dynamic   horizontal correction  provided heretofore by  the prior art;
FIG. 6 is a schematic diagram showing the circuit of FIG. 1 in detail;
FIG.     7 illustrates the face of a picture tube during dynamic 
vertical     correction for a wide blue field occurring in the initial 
horizontal     scan period and a narrow blue field occurring in the 
final horizontal     scan period;
FIG. 8 is a schematic diagram of a modified   portion   of the circuit 
of FIG. 6, and illustrates a simplified   convergence   circuit for the 
blue color beam;
FIG. 9 is a schematic diagram of a passive convergence system in 
accordance with the present invention;
FIG. 10 is a schematic diagram of another embodiment of a passive 
convergence system in accordance with the present invention;
FIG.     11 is a schematic diagram of a horizontal convergence section 
and     showing a novel source of waveform, consisting of the S shaping 
    capacitor, for supplying current to the convergence section;
FIG.     12 is a schematic diagram of a horizontal convergence section  
   incorporating a series-parallel connection for the convergence coils;
FIG.     13 is a schematic diagram of a horizontal convergence section  
   incorporating a series connection for the convergence coils; and
FIG.     14 is a schematic diagram of a simplified horizontal 
convergence     section made possible by optimum selection of yoke, 
picture tube and     neck component parameters.
GENERAL OPERATION
FIG
.
    1  in conjunction with FIGS. 2 and 3 shows in an otherwise   
conventional   color television receiver, the applicant's static and   
dynamic   convergence method and apparatus for converging the plural   
color beams   of a trigun color cathode ray tube (CRT) 20. The CRT has  
 three identical   gun structures for generating three electron beams   
which impinge blue   (B), green (G) and red (R) phosphors on the face of
   the CRT. For   convenience, similar structures will be identified by 
  the same reference   numeral, followed by the designation B, G or R to
   indicate whether the   same is associated with the blue, green or red
   electron beam guns,   respectively.
Within the glass tube   envelope, a magnetic  shield  divides the neck 
of the CRT into three   compartments associated  with the  B, G and R 
electron guns. In the   absence of a convergence  magnetic  flux field, 
the beam travels   undeflected along an axis  perpendicular to  the 
plane of FIG. 1 and   equidistance from spaced  internal pole pieces  
23, 24 which each are of   generally L-shaped  cross-section. Associated
  with each pair of   internal pole pieces 23, 24  is a converg
ence
 exciter assembly 26 mounted external to the CRT.
To     adjust for static convergence, a permanent magnet 28 spans an air
  gap    between a pair of abutting L-shaped legs 30 and 31 which form a
    U-shaped  ferrite core. The permanent magnet 28 may be rotated, or  
  otherwise  suitably moved with respect to the orientation of its    
magnetic North and  South fields, in order to alter the direction and   
 magnitude of the  steady magnetic flux field which circulates through  
  the legs 30, 31.
To  effect dynamic convergence, each of the    exciter assemblies 26 
includes a  coil which is coupled to the    convergence circuitry, to be
 explained.  Dynamic convergence of the B    electron beam is effected 
by a coil formed  of two identical coil    sections 35 and 36 wound on 
the legs 30B and 31B,  respectively. Dynamic    vertical and horizontal 
convergence of the R beam  is effected by a    single coil 38, wound on 
the leg 31R of the red  assembly 26R. Finally,    dynamic vertical and 
horizontal G beam is  provided by a single coil  39   wound on the leg 
30G of the green assembly  26G.
The static  and   dynamic convergence procedure, using the  circuit of 
FIG. 1, may  be   understood with reference to FIGS. 2 and 3  which show
 a face 42 of  the   CRT 20. The corresponding convergence  procedure of
 prior  apparatus is   demonstrated by FIGS. 4 and 5. In the  
applicant's  convergence method,   the static convergence magnets 28B, 
28R  and 28G  are pre-set such that   the R, B and G scanning lines 
converge at  the  edges of the face of the   picture tube, namely, at 
the beginning and   end of vertical scan (FIG.   2) which occurs at the 
center of the   horizontal period, and the   beginning and end of 
horizontal scan (FIG.  3)  which occurs at the   center of the vertical 
period. Following  static  convergence at the   edges of the picture 
tube, the currents in  coil 38 of  the red assembly   26R and coil 39 of
 the green assembly 26G  are adjusted  to vary the R   and G vertical 
lines until they coincide.  The maximum  extent of   convergence 
deflection occurs in the center of  the picture  tube, as   represented 
by arrow 44 in FIG. 2. Next, the  currents in coils  35, 36   of the 
blue assembly 26B are adjusted such  that the B horizontal  line   is 
caused to be coincident with the G and R  horizontal lines. The    
maximum extent of deflection of the B  horizontal line occurs along the 
   center of the face of the CRT, as  represented by arrow 46 of FIG. 3.
By    statically converging  the edges of the scanning lines, and  
dynamically   converging the  center of the scanning lines, the driving 
 currents in  the  convergence  exciter assemblies 26 are never forced 
to  make a  radical  change or  shift during the retrace period. With 
dynamic   convergence  applied in  the center of the picture tube, as  
illustrated,  there is a   considerable time period for the correction  
current to build  up in the   exciter coils, resulting in a lower energy
  level  requirement. Also,  the  convergence procedure is simpler due 
to  the  independence of the   convergence controls in the various 
quadrants  of  the picture tube. 
The
   above convergence procedure should be   contrasted with the prior   
convergence procedure, as illustrated in   FIGS. 4 and 5. First the   
permanent magnets associated with the   convergence exciter assembly are
   pre-set during a static convergence   phase to cause the R, B and G  
beams  to converge in the center of the   picture tube, as shown in  
FIGS. 4 and  5. Next, the currents in the   coils associated with the  
red and green  exciter assemblies are adjusted   to cause the R and G  
vertical lines to  coincide at the top and bottom   edges of the  
vertical scanning period,  which results in maximum   convergence  
deflection in the vicinity of  arrows 48 of FIG. 4. Finally,   the coils
  associated with the blue exciter  assembly are fed currents   which  
cause the B horizontal line to coincide  with the G and R   horizontal  
lines, creating a maximum convergence  deflection along the   arrows49  
of FIG. 5.
Returning to FIG. 1,  the structure for the   red  assembly 26R and 
green assembly 26G is not  identical with the    structure for the blue 
assembly 26B. The structure of  the blue assembly    26B causes the B 
electron beam to be deflected in a  radial direction    51. While radial
 deflection is desirable for the blue  beam, due to  the   fact that the
 blue gun structure in a conventional CRT  is  orientated  in  a 
vertical plane, radial deflection of the G and R   electron beams   
would produce vector components in both the vertical  and  horizontal   
planes. A vertical component of movement for the R and  G  beams is   
frequently unnecessary.
To overcome this problem,  the  green   dynamic convergence assembly 26G
 and the red dynamic  convergence    assembly 26R are modified so as to 
produce a  non-symmetrical or    unbalanced flux field between the pole 
pieces 23,  24 associated    therewith. This field skews the path of 
deflection of  the associated    electron beam, and produces essentially
 lateral G and R  beam movement,    as illustrated by the arrows through
 the G and R  beams. To accomplish    this result, a single convergence 
coil is wound  over only one leg of  the   U-shaped core 30, 31. The 
green convergence  coil 39 is wound about   core  leg 30G, while the red
 convergence coil  38 is wound on the core   leg  31R.
If
 desired or necessary, the  core legs carrying the    single coils may 
abut a pole shoe (not  illustrated) sandwiched between    the core leg 
and the glass envelope  of the CRT. The net result is to    change the 
magnetic flux pattern  produced in the vicinity of the    electron 
beams, because the magnetic  flux passing through the core legs    30G 
and 31R is considerably  greater or more concentrated than the flux    
passing through the  opposed core legs 31G and 30R. For a more  complete
   description of  such apparatus and the resulting operation,  
reference   should be made  to the applicant's before identified  
co-pending   application, Ser. No.  263,632 filed June 16, 1972.
OPERATION OF DYNAMIC CONVERGENCE CIRCUIT
 A     description will now be given of the dynamic convergence 
circuitry     utilized in conjunction with the applicant's convergence 
method, which     may be characterized as an "inverse" convergence 
approach. It should  be    noted that while active circuitry is 
disclosed, passive type   circuitry   can also be used, as explained 
later. In the inverse   convergence   approach, the main objective is to
 supply the convergence   exciter coils   35, 36, 38 and 39 with both 
horizontal and vertical   parabolic currents   synchronized with the 
television scan circuits.   Unlike conventional   convergence 
approaches, it is unnecessary to   utilize separate coils for   the 
vertical and horizontal convergence   functions, thereby allowing the   
length of the core legs 30, 31 to be   shortened.
Generally   speaking, the inverse conversion circuitry   requires four 
parabolic   waveform generators. One generator provides   the vertical 
convergence   current, and three generators provide the   horizontal 
convergence   current. The output from all four waveform   generators 
are combined by   appropriate matrixing circuitry fed from   output 
amplifiers.
Considering
     FIG. 1 in detail, the color television receiver includes a  
horizontal    deflection circuit 60 (shown twice in FIG. 1) and a  
vertical  deflection   circuit 62, both of which may be conventional. A 
 horizontal  sawtooth   waveform 64 from the horizontal deflection  
circuit 60 is  applied to the   input terminal of a parabolic waveform  
generator A1. An  inverted  version  65, the same sawtooth waveform is  
applied to the  input terminal  of a  parabolic waveform generator A2.  
Generator Al  provides at its  output an  initial half 67 of a parabolic
  waveform,  which is coupled to  coil 36 in  the left channel of the  
blue exciter  assembly 26B. The other  half of the  dynamic correction  
parabola is  supplied by parabolic  waveform generator  A2, which  
generates a  terminating half 68 of the  parabolic waveform.
The   output from  generator Al is supplied  through a variable resistor
 72  to  the input  of an amplifier A3 which  forms a part of the left  
driving  channel for  assembly 26B. The output  from generator A2 is  
supplied  through a  variable resistor 74 to the  input of an amplifier 
 A4 located  in the  right channel. The reason for  separate correction 
 currents  supplied  through separate channels to a  split blue  
convergence coil is  to  provide separate and independent  lateral  
correction over the  beginning  and terminating portions of a   
horizontal scan. This allows a  wide  blue field/narrow blue field on  
the  face of a CRT to be corrected,  as  will be explained in detail  
later.
A  vertical frequency   sawtooth waveform 76 is supplied  to a parabolic
  waveform generator A5.   An inverted version 77 of the  sawtooth 
waveform  is also provided to   generator A5. The resultant  parabolic 
correction  waveform appears at an   output terminal, and is  supplied 
to the blue  conversion exciter  assembly  through a variable  resistor 
80 in series  with the wiper 82 of  a  potentiometer 83 having  one end 
of its fixed  resistance coupled to  the  input of amplifier A3  and the
 other end  coupled to the input of   amplifier A4. The position  of 
wiper 82  determines the ratio of the   vertical deflection  correction 
which is  applied to the left and right   channels of the  blue exciter 
assembly.  Variable resistor 80 allows the   strength or  amplitude of 
the vertical  correction signal passing to the   blue  conversion 
exciter assembly to be  preset. Wiper 82 of  potentiometer   83 allows 
the amount of correction  applied to the coils  35 and 36 to  be  varied
 as desired.
  The  vertical correction  signal from  parabolic generator A5 is also 
 passed  through an  adjustable resistor 86,  a fixed resistor 87, and a
  wiper 89  of a  potentiometer 90 to the red  and green exciter  
assemblies. One end  of  the fixed resistance forming  potentiometer 90 
 is coupled through an   output amplifier A6 to coil 38,  whereas the  
opposite end of the fixed   resistance is coupled through an  output  
amplifier A7 to the coil 39.   The undriven ends of coils 38 and 39  are
  coupled to a source of   positive DC voltage, or B+. Variable  
resistor  86 thus controls the   amount or level of the vertical 
correction   current which is fed to the   red and green convergence 
exciter   assemblies, whereas the position of   wiper 89 controls the 
ratio of   vertical correction current which   divides between the red 
and green   assemblies.
For horizontal   convergence of the R and G electron   beams, a 
parabolic waveform   generator A8 has a pair of inputs coupled   to a 
horizontal sawtooth   waveform 94 and an inverted form 95  thereof,  
which may be similar to   the waveforms 64 and 65 associated  with the 
blue  exciter assembly. The   resulting parabolic correction  signal 
from  generator 88 is passed   through a variable resistor 97 and  the 
fixed  resistor 87 to the wiper   89 of potentiometer 90.
A detailed  description of the circuitr
y
    shown in block form in FIG. 1 will now be  given with reference to  
 FIG.  6. The vertical parabolic generator A5  includes amplifier   
transistors  110 and 111. Voltage from the B+ supply  is coupled to each
   transistor  through a common load resistor 113.  Transistor 110 is   
driven into  conduction during the initial portion of  the vertical   
sawtooth waveform  76, whereas transistor 111 is driven into  conduction
   during the second  half of the vertical scan period, due to  the   
positive going inverted  sawtooth 77 which rises above zero volts.
Negative    feedback is  provided from the collector of the transistors 
110, 111   to  their  respective bases through capacitors 115 and 117,  
 respectively.  This  results in the generation of a parabolic waveform 
  120 which is  coupled  to a buffer amplifier stage 122. The generator 
A5   is extremely   flexible, and allows the shape of the parabolic 
slope  to  be modified   separately for each half of the vertical scan 
period  by  appropriate   adjustment of the time constants within the 
generator  A5.  Without   feedback capacitors 115 and 117, a tiangular 
waveshape  would  appear in   place of the parabolic waveshape 120. 
Potentiometers  124 and  126 provide   separate adjustment for the top 
and bottom  sections of  the CRT screen.
Many
   variations are possible in the design of  the parabolic  generator 
A5.   The output shape and amplitude of the  waveform 120 is a  function
 of  the  input DC offset, and the amplitude  and linearity of the  
sawtooth.  By  driving the transistors 110 and 111  into saturation for 
 various   portions of the scan period, the resulting  output waveform 
can  be   varied between a large number of shaped  tailored for the 
convergence    correction which is necessary for a  particular CRT 20.
Buffer    amplifier 122 provides impedance  matching for the output 
amplifiers.  The   parabolic correction waveform  is fed from 
potentiometer 80 to the  blue   output amplifiers A3 and A4  via the 
balancing potentiometer 83.  The  blue  section is different in  
operation from the red and green  sections  since  two separate output  
amplifiers A3 and A4 are used to  drive  separate  sections 35 and 36 of
  the blue convergence coil. Since  the  blue beam  moves on a vertical 
 axis, due to the geometry of the  delta  gun  configuration of the CRT 
 20, it is generally desired to  provide  balanced  currents in the 
output  amplifiers A3 and A4, which  may be  accomplished  by 
appropriate  adjustment of potentiometer 83.  The  vertical parabolic  
correction  signal from buffer amplifier 122 is  also  applied to the 
red  and green  circuitry via the adjustable  potentiometer  90.
The  red/green  horizontal parabolic generator  A8 operates in  a 
similar  fashion to  generator A5. A pair of  transistors 130 and 132  
are driven  from two  anti phase horizontal  sawtooth waveforms 94 and 
95.  The two   transistors share a common load  resistor 134, connected 
with  B+. A pair   of negative feedback  capacitors 136 and 137 shunt 
the  collector to  base  electrodes of the  pair of transistors. Thus, 
the  output waveform  140 is  of parabolic  shape for the same reasons  
described with  reference to  generator A5,  but occurring at the  
horizontal frequency.  Furthermore,  the same  modifications can be made
  to the generator in  order to tailor  the  waveform 140 for a 
particular  CRT 20.
The  major difference   between generators A8 and A5 arises  from the 
fact  that in most   television receivers, the necessary  sawtooth 
waveforms  94, 95 cannot  be  obtained directly from the  horizontal 
scan circuit.  Therefore, it  is  derived from the flyback  pulse which 
is integrated by  an RC  network. In  particular, a pair of  anti phase 
horizontal flyback   pulses 144 and 145  are applied to input  terminals
 for the generator   A8. The waveform 144 is  integrated by a  resistor 
147 and a capacitor   148, the capacitor having a  potentiometer  149 
coupled in shunt   thereacross. Similarly, waveform 145  is supplied  to
 an integrator   consisting of a resistor 152 and a  capacitor 153 which
  has a   potentiometer 154 coupled in shunt  thereacross. Thus, the 
wipers  of   potentiometers 149 and 154 may be  adjusted in a manner 
similar to  the   wipers of potentiometers 124 and 126  in generator A5,
 and provide    separate and independent control over the  left and 
right hand portions    of the screen. 
 It is not essential to  produce a parabolic waveform   140,  in that 
current integration produced  by the convergence exciter   may be  
utilized to convert the triangular  waveshape into a desired   parabolic
  correction waveshape, if desired.
The  waveform 140 is   coupled to  a buffer amplifier 160 which performs
  impedance matching   functions and  has an output coupled to a summing
  junction 162 which   feeds wiper 89  of potentiometer 90. Thus, the 
buffer  amplifiers 122   and 160 form a  mixer circuit for combining the
 horizontal  parabolic   wave with the  vertical parabolic wave. The 
composite waveform  is   coupled to the  convergence exciter coils 
through the differential    control  potentiometer 90 and output 
transistors 164 and 165, which form     amplifiers A6 and A7, 
respectively. The output transistors 164 and   165   are connected in 
common emitter fashion and isolate both the   vertical   and horizontal 
parabolic generators from the exciter   assemblies 26R and   26G.
The blue horizontal parabolic generator   A1 and A2  function  in a 
manner somewhat similar to the red/green   horizontal  parabolic  
generator A8. That is, a pair of transistors 170   and 172 are  
connected  in common emitter fashion, and have sawtooth   waveforms 64 
and  65  coupled thereto via RC networks which include   adjustable  
potentiometers  174 and 175. The basic dissimilarity with   generator A8
 is  that  transistors 170 and 172 are split, by use of   separate load 
 resistors  178 and 179, to provide two independent   outputs. As a 
result,  the first  half of the parabolic correction   signal 67 is on 
one output  line, and  the terminating half of the   parabolic 
correction signal is  present at a  different output line.
The   pair of output lines  are fed to a  buffer amplifier 180 which  
includes  separate transistors  181 and 182  with the potentiometers 72 
 and 74  forming the emitter  resistors for the  respective transistors.
   Amplitude correction is  accomplished by  adjusting potentiometer 72 
 for  the left half of the CRT  face. Both  potentiometers generate the 
  maximum amplitude of convergence  correction  at the middle of the 
scan   period, corresponding to the center  of the  screen as 
illustrated in   FIG. 3.
The output of  potentiometer 72  is coupled to the wiper   190 of a 
potentiometer 192  having its end  terminals coupled to the   base of 
transistors forming blue  output  amplifiers A3 and A4.   Similarly, the
 output of potentiometer 74  is  coupled to the wiper 194   of a 
potentiometer 195 having its end   terminals in shunt with the end   
terminals of potentiometer 192, and   hence connected to the base   
electrodes of the transistors forming blue   output amplifiers A3 and   
A4.
Amplifier A3 is formed by a   transistor 200 having its   collector 
coupled directly to one end of   winding 35, with its opposite   end 
being coupled to B+. Similarly,   amplifier A4 is formed by a   
transistor 202 having its collector coupled   to one end of winding 36, 
  the opposite end of which is connected to B+.   The pair of windings 
35   and 36 are shunted by resistors 204 and 206,   respectively. When 
the   wipers 190 and 194 are set to their center   positions, 
transistors 200   and 202 supply equal currents to the windings   35 and
 36, and the   magnetic field generated for convergence is   balanced. 
The balanced   flux state results in a straight vertical   electron beam
 displacement   at the center of the CRT screen.
W
ipers
     190 and 194 allow correction of what is termed wide blue lateral   
field   and narrow blue lateral field, which conditions are illustrated 
  in  FIG.  7. As noted in FIG. 7, the red and green guns have been   
properly   converged, causing the vertical R and vertical G lines to lie
   on top of   one another. Red/green convergenced can be obtained  
through  the   convergence circuitry, since these beams move toward one 
 another  in the   convergence correction procedure. The blue beam,  
however,  emanates from  a  gun which is not adjusted simultaneously  
with the red  and green  guns.
Thus,  the blue gun may create a  blue vertical  line offset  from the 
red/green  vertical lines at the  edges of the  screen. As seen  in FIG.
 7, the blue  vertical line B is  outside of the  R/G lines for the  
left hand portion of  the screen. The  amount of  offset, designated by 
 the arrow 210, is known  in the art  as a wide  blue field. It is also 
 possible that the blue  vertical line  B may be  inside the R/G line at
  the opposite edge of the  screen, as  represented  by the offset 212. 
This  latter condition is known  as a  narrow blue  field. Heretofore, 
it has  been possible to correct for  a  wide blue  field occurring on 
both sides  of the screen, or a narrow   blue field  existing on both 
halves of the  screen, but not a mixed   condition as  shown in FIG. 7. 
The apparatus of  FIG. 6 allows  correction  for this  mixed condition.
Returning
 to  FIG. 6,  amplifier A3  and coil 35  provide left channel 
correction, while   amplifier A4 and coil  36  provide right channel 
correction. The right   channel blue signal   applied to coil 36 rotates
 the magnetic vector   across the pair of   internal pole pieces in the 
direction 214, causing   beam displacement   along a transverse 
direction 216. The left channel   produces the reverse   operation, in 
that flux rotation occurs along  the  direction 216 and   hence beam 
displacement is along the axis 214.  The  horizontal  components  of 
vectors 214 and 216 provide lateral  correction  for  wide/narrow blue  
vertical lines.
Proper  adjustment of the   potentiometers 192  and 195 results in 
correction of  either a wide or a   narrow blue field,  which correction
 is  independent for opposite halves   of the screen. For  example, the 
left  side can be adjusted for a wide   blue field by moving  the wiper 
190  toward the upper position, while  the  right side can be  corrected
 for  wide blue field by adjusting the  wiper  194 towards the  lower 
end of  its range. A narrow blue field can  be  corrected by  reversing 
the  above procedure. Or, the two controls  can be  separately  adjusted
 to  correct for a wide blue field on one  half, and a  narrow blue  
field on  the other half of the screen. If  there is no  lateral blue  
error  which requires correction, then the  wipers 190 and  194 should 
be  set  to their center positions. The above  described  network also 
provides   the proper waveform necessary to  correct for blue  droop in 
the   deflection yoke and CRT.
For  completeness, the  steps taken to   converge the three beams are 
set  forth below. It should  be noted that   independent correction 
exists in  each quadrant of the CRT  screen, due  to  the minimum 
interraction  between the various magnetic  fields. A  typical  
convergence procedure  would involve the following  steps:
1.   Adjust the static  convergence magnets to apply static  correction 
at  the  edges or  periphery of the CRT face. Completion of  this step  
results in  vertical  and horizontal scanning lines as shown in  FIGS. 2
  and 3.
2.   Apply dynamic correction to adjust the red  and  green vertical 
lines   such that they become parallel. At the right  and  left side, 
the red and   green lines will also be coincident, but may   be 
separated at the  center  of the screen.
3. Apply dynamic correction to the red and green horizontal lines, which
 lines will now coincide from top to bottom.
4.     Apply dynamic correction to the red and green vertical lines at 
the     right side of the CRT face. The red and green vertical lines 
will  then    be coincident from approximately the center of the screen 
to the  right    side.
5. Apply dynamic correction to the red and green    vertical  lines at 
the left side of the CRT face. The red and green   lines  will  now be 
coincident at all points on the CRT face.
6.    Apply  dynamic correction to the blue horizontal lines in the 
central    portion  of the CRT face. The blue lines will now be 
coincident with  the   red and  green lines in the central portion of 
the CRT, but  elsewhere  the  blue  lines will tend to fall below the 
coincident red  and green   horizontal  lines, depending on the yoke 
field.
7. Apply dynamic correction to the blue horizontal lines at the top of 
the CRT face.
8.     Apply dynamic correction to the blue horizontal lines at the 
bottom   of   the CRT face. This completes the convergence procedure, 
and all   three   beams will converge properly for all points on the CRT
 face.
While     the inverse convergence method has been described with 
reference to   an   active convergence system, as illustrated in FIGS. 1
 and 6,  various    modifications and simplifications can be made as 
shown in the   remaining   figures.
MODIFIED EMBODIMENTS
The blue   correction   section shown in FIG. 6 may be simplified, if 
desired, to   eliminate the   separate lateral controls for the left and
 right hand   portions of the   screen. 
As
    shown in FIG. 8, the parabolic generator formed by  differential    
transistors 170 and 172 is integrated into a single unit by  use of a   
 common load resistor 230 coupled between the collectors of the     
differential transistors and B+. A single parabolic output, which     
corresponds to combined waveforms 67 and 68 of FIG. 6, is coupled to a  
   single stage buffer amplifier formed by a transistor 232 connected as
   an   emitter follower. Potentiometer 234 forms the emitter 
resistance,   and   suplies a signal to a summing junction 236 which is 
coupled to  the    output from potentiometer 80 in the vertical buffer 
amplifier 122  of    FIG. 6. The combined output drives a single output 
transistor 240   having   a collector coupled to B+ through the series 
connection of  blue   winding  sections 35 and 36. A load resistor 242 
shunts the  series   connected  coils 35 and 36.
Another alternate design  (not   illustrated)  for the circuit of FIG. 6
 is to provide a separate  pair of   output  transistors, corresponding 
to transistors 164 and 165,  for the   vertical  channel. Since there is
 a ratio of 262.5:1 between  the   horizontal and  vertical scan 
frequencies, the output power can be    reduced by providing  separate 
load impedances. That is, a set of    amplifiers, similar to A6  and A7,
 would be fed from the wiper of    potentiometer 86 of the vertical  
parabolic generator. The red coil 38    and green coil 39 each would be 
 split into separate sections by means    of a tap on the winding. The  
output amplifiers for the vertical stage    would be coupled between one
  side of the windings 38 and 39 and the  B+   connection taps thereto, 
 whereas the output amplifiers A6 and A7  for   the horizontal generator
  would be coupled to the taps and the B+  side   of the coils 38 and 
39.
The  inverse convergence method is    equally applicable to passive 
systems of  the type illustrated in  FIGS. 9   and 10. The inverse 
convergence method  can be implemented in  general   by a conventional 
blue convergence circuit  since maximum  correction   currents occur in 
such a circuit at the middle  of the  horizontal sc
an.
    The green and red convergence circuits can also  be implemented with
    the same basic design, because in the inverse  method, all three    
convergence fields require maximum correction at the  middle of the scan
    period. A passive system incorporating these concepts  is 
illustrated    in FIG. 9. In this circuit, separate horizontal  
convergence coils  250,   each associated with a different beam, are  
provided as is   conventional  in the art. A conventional blue 
horizontal  convergence   circuit 252  supplies horizontal convergence 
correction  current to   horizontal  convergence coil 250B, and is 
powered from the  horizontal   flyback pulse  144. A coil 254 provides a
 shaping function.
A    novel red/green  horizontal convergence section is coupled directly
    across the  horizontal convergence coil 250B. One end of the coil 
250B   is  coupled  to a variable inductor 256 in series with a 
capacitor 257    terminating  at a junction of the red 250R and green 
250G coils. The    opposite sides  of coils 250R and 250G are coupled to
 end terminals of a    potentiometer  260 having a wiper 262 connected 
in series with a   variable  resistor 264  connected to a source of 
reference potential, or   ground  265. The  junction between capacitor 
257 and the coils 250R  and  250G is  shunted  to ground 265 through a 
resistor 268 in series  with a  diode 270.
The  red/green section thus derives its  parabolic  current from  the 
same  source which feeds the blue coil  250B. Inductor  256 and  
capacitor 257  provide right side amplitude and  current slope  shaping.
  The amplitudes  of the left side are adjusted  by the setting  of 
variable  resistor 264.  The left and right R/G  horizontal lines are  
adjusted  simultaneously by  movement of wiper 262  of potentiometer 
260.  Since the  red and green  exciter coils are  driven from a common 
 source, a single  clamping  circuit, formed by  resistor 268 and diode 
 270, is sufficient. A   capacitor 272 provides  the proper current  
profile for the convergence   exciter coils.
The  above described  circuit can be simplified  if  the inductances of 
all  the horizontal  exciter coils 250 are reduced   sufficiently to 
total  the original  inductance of the blue horizontal   coil 250B. 
Various  parameters can  also be modified, both with respect to   
component  values and exciter  coil inductances, to produce the   
simplified design  shown in FIG. 10.  The function of variable resistor 
  264 is provided  by a resistor 280 in  series with a variable resistor
 282   coupled in  shunt between the  junction of all three horizontal 
coils  250  and  ground 265. The  conventional horizontal blue circuit 
252 can  drive   all three  horizontal coils 250 provided that the 
equivalent   inductances  of the  exciter coils equal the total 
inductance of the   original blue  coil.
The  necessity for converting the horizontal   flyback  pulse into a  
parabolic waveform for driving the inverse   horizontal  convergence  
circuit can be eliminated entirely in many  solid  state  television  
receivers. In FIG. 11, a portion of the  horizontal  deflection  circuit
  60 for a solid state television  receiver is  illustrated. The   
horizontal output amplifier 290 drives a  flyback  winding 292 connected
   between the output amplifier and B+.  The  horizontal deflection yoke
   windings 294 are connected in series  with an  "S" shaping capacitor 
296,   which improves linearity of the  scanning  current as is well 
known. A   diode 300 in series with a  centering  variable resistor 302 
is  connected  between B+ and the "S"  shaping  capacitor 296. The above
  described  horizontal deflection  circuit is  conventional, and 
provides   approximately 30 to 40 volts of  AC signal  which rides on a 
110 volt DC   level. This situation should  be contrasted  with many 
vacuum tube  designs  in which the "S" shaping  circuit  impedance is 
too high to be  used for  convergence  applications, and the  boost 
voltage is too high  for economic  coupling  of the waveform across  the
 "S" capacitor to a  convergence  circuit.
With  solid state  television receivers of  the type  described above, 
this  situation is  changed. The voltage  across the "S"  shaping 
capacitor  can be coupled  through a capacitor  306 to a simplified  
horizontal  convergence control.  The single source  for driving the  
horizontal  convergence control is a  parabolic current  already 
available  in the  television receiver, namely,  the signal  across the 
"S" shaping   capacitor 296. This eliminates the  flyback  pulse 
converting techniques   previously used with passive  circuits.
The  passive inverse   circuit, which is simplified by  using the 
parabola  developed across  the  "S" correction capacitor 296 of  the 
horizontal  output stage,  operates as  follows. The horizontal  output 
transistor  controls the  current through  the flyback winding 292  and 
deflection  yoke windings  294. The horizontal  centering circuit  
consists of the  variable  resistor 302 in series with  the diode 300 to
  provide a  differential  voltage between flyback and yoke  winding.
The   resulting  horizontal parabolic waveform 317 is  developed across 
the   "S"  capacitor 296 by the sawtooth yoke current. The  "S" parabola
 is    coupled to the convergence circuit by capacitor 306.  The 
capacitor 306    performs a dual function by blocking the DC potential  
from the scan    circuit, and providing a reactive component for the  
convergence    circuit. The output of the capacitor 306 is connected to 
the  R-G-B    color circuits.
The blue circuit amplitude, phase, and  waveform    profile are adjusted
 by the combination of the inductor 316 and     resistance control 314. 
Capacitors 315 and 318 are selected to provide     the proper phase and 
current waveform parameters for the blue     convergence exciter coil 
250B.
The red and green circuit is     matrixed with a differential 
potentiometer 310 to provide adjustment of     the horizontal lines over
 the center of scan. As previously  described,    static convergence 
occurs at the ends of the scanning  lines. The    vertical lines are 
adjusted by a resistance control 309  and a variable    inductance 307. 
Capacitors 308, 311 and 312 are  selected to provide the    proper 
current waveform for the convergence  exciter coils 250R and   250G.  
The ampere turns of the convergence  exciter coils are also   selected 
to  provide the optimum magnetic  characteristics and circuit   
inductances for  the system.
The  active inverse design can also   be simplified  by using the 
parabolic  waveform derived from the   horizontal "S"  correction 
capacitor. The  parabolic generators A1, A2,   A5 and A6 of FIG.  1 can 
be eliminated by  the application of the "S"   parabola.
The  inverse concept  permits additional simplification   of the R-G-B 
circuit  since the  maximum dynamic correction is required   at the CRT 
center for  all  three color beams. A simplified approach is   shown in 
FIG. 12. The   blue coil 325B is in series with both the   horizontal 
parabola source,   which desirably is the "S" shaping   capacitor 
although other sources  can  also be used, and a parallel   combination 
of the red 325R and  green 325G  coils. The series-parallel   exciter 
circuit design shown in  FIG. 12  requires fewer components than   FIG. 
11.
Another  design  simplification occurs by using all   series connections
 for the  exciter  coils. This configuration is shown   in FIG. 13. The 
matrixed  exciter  design permits great simplicity in the   convergence 
system. In  both FIGS.  12 and 13, a horizontal parabola  320  is 
derived from the  "S" capacitor,  or circuitry that will convert  the  
flyback pulse to a  parabolic  waveform. The coil 322 is adjusted to   
correct for the blue  field. The  blue section is adjusted first since  
it  requires the  maximum drive. The  red and green section is adjusted 
 with  two  resistance controls 326 and  329. The inductances of the red
  and  green  exciter coils will be  different in the series-parallel   
configuration  of FIG. 12, than for the  series design of FIG. 13.
The    deflection yoke and picture tube  design can also be simplified 
by use    of my invention. The present  combination has been developed 
over the    years to be compatible with the  present passive nonlinear 
convergence    system. However, the disclosed  inverse concept makes it 
possible to    modify present design and improve  the optical system 
performance. For    example, the picture tube  fabrication can be made 
easier. Presently,    the R-G-B electron gun mount  is tilted to 
converge the beams at the  CRT   center. A tilted gun is more  difficult
 to align at a precise  angle.   However, my inverse convergence  system
 permits all three guns  to be   constructed on a parallel axis. This  
allows the free fall beam  landing   geometry to be arranged in an  
underconverged pattern.
Since    parallel gun design permits  greater accuracy and better 
uniformity  of   the beam landing pattern, it  is possible to eliminate 
or greatly  reduce   the static magnet flux  required from the exciter 
assembly.  This  static  flux reduction reduces  the misconvergence of 
the op ical   system.  Improved optical uniformity is  possible due to 
the reduction   of the  random leakage flux surrounding  the static 
permanent magnets.
By    selecting the optimum design  parameters of the deflection yoke,  
  picture tube, and neck components, it  is possible to reduce the    
convergence circuit to the design shown in  FIG. 14. The simplified    
circuit of FIG. 14 requires a horizontal  parabola 320 which supplies   
 the convergence exciter current. A capacitor  321 isolates the DC    
potential from the scan circuit, and provides  current shaping for the  
  remaining circuit. The exciter coils 325B, 325G  and 325R have the    
required ampere turns needed to compensate for  symmetrical or    
asymmetrical R-G-B fields. Capacitors 323, 327 and 328,  in conjunction 
   with the exciter coil inductances, are used for current  waveform    
shaping. The current amplitude and phase is adjusted by a  variable    
inductance 322. While certain modifications and simplified  circuits    
have been described, it will be appreciated that many other  variations 
   are equally possible.   



















No comments:
Post a Comment
The most important thing to remember about the Comment Rules is this:
The determination of whether any comment is in compliance is at the sole discretion of this blog’s owner.
Comments on this blog may be blocked or deleted at any time.
Fair people are getting fair reply. Spam and useless crap and filthy comments / scrapers / observations goes all directly to My Private HELL without even appearing in public !!!
The fact that a comment is permitted in no way constitutes an endorsement of any view expressed, fact alleged, or link provided in that comment by the administrator of this site.
This means that there may be a delay between the submission and the eventual appearance of your comment.
Requiring blog comments to obey well-defined rules does not infringe on the free speech of commenters.
Resisting the tide of post-modernity may be difficult, but I will attempt it anyway.
Your choice.........Live or DIE.
That indeed is where your liberty lies.
Note: Only a member of this blog may post a comment.