RCA 21-D-7177 Horz Drive Issues
 

I restored this set back in December of this year and have been watching and trying to learn proper setup with it ever since. Overall it's working well, but I've never been able to adjust the horizontal drive to my satisfaction. Regardless of the Horz drive cap adjustment, I end up with what looks like ringing, or a ripple in the left side of the picture. I mostly notice this on the bottom left screen while watching program material. Pictures are taken with test CRT, but the same issue persists with the sets CRT. I also see faint light and dark vertical streaks (drive lines?) in the picture. With the fine cross hatch pattern, I get the bright band as can be seen in the pictures. In general the picture looks best with the HORZ drive cap turned all the way closed. Adjusting the Horz drive cap moves the white band seen in the cross hatch across the screen, but I can't make it go away. All my horizontal waveforms look correct as shown in SAMS.

Would appreciate any ideas on where to look for the problem.

Sure the horizontal wave coil has a correct looking waveform on it? You should have equal peaks.

I think it's real close. This is what it currently looks like with the fine cross hatch pattern. This is looking at point I on the schematic.

To offer a WAG, possibly leakage between the V and H windings of the yoke, with the potential for developing into the classic trapezoidal-shaped raster. oc

Here's what the Horz drive waveform looks like, and a screen shot that shows the horz wave over on the far left (look where it says QVC price $13.98 in the black bar). I think I can actually see the small dip in the bottom of the drive waveform? Also some ringing way at the bottom. My problem is I don't know how much of this distortion is considered normal for this set.

I wondered about that. I guess I will need to find a new yoke in order to test that theory. I also wondered about the cap directly across the one yoke winding. I have not replaced that. I think it's rated at 2500 volts.

I've has the same issue with a few sets. It's usually something in the damper circuit. I'd check that yoke capacitor too.

These are my few thoughts.... I do not think there is a horizontal drive problem. Horizontal overdrive leads to non linear scanning where the beam slows during forward horizontal trace before the Horizontal output tube conducts. Instead I believe the line is displayed during the horizontal retrace.

When you adjust the horizontal hold, does the line move? If so, the problem is too slow a horizontal retrace. What you are seeing is the first vertical line in the cross hatch which should be on the left of te picture but is being displayed in the middle of the screen as the beam too slowly swings back to the left.

There are a few reasons why this may be so.

1) Is the test generator horizonal blanking wide enough. Examine the horizontal sync pulse with the scope. Horizintal blanking is supposed to be 10.9us. If the video starts too soon on the generator, you will see the white line. See the link below I found on the web:

http://www.maxim-ic.com/app-notes/index.mvp/id/734

2) The horizontal retrace is too slow due to a circuit fault. The horizontal output circuit is tuned to about 70kHz to facilitate the fast timed beam swing back t the left. Anything which may affect the horizontal "tuning" eg bad capacitor on the linearity coil, bad damper tube, shorted horizontal output transformer, shorted yoke etc may slow down the retrace.

Do you see the line on normal pictures? If not, I'd have a closer look at the cross hatch generator.

Thanks for the suggestions.

The line definitely moves while changing the Horz hold control.

Don't see a line on normal pictures, but something is bending the horizontal scan lines over on the left-hand side of the picture. It's not bent all the way top to bottom, just at certain points in the picture. Moving the hold all the way to the right seems to help minimize the horizontal bending. I'm not sure if the two symptoms are related?

With the RCA Synchroguide horizontal hold setting, I generally set the horizontal frequency (transformer core adjustment) so that with the full right setting of the front control it is on the verge of losing horizontal sync. At this point the horizontal oscillator is most firmly locked

The reason the white bar disappears when turning the knob and the video to the right is because the retrace is starting earlier hence the left portion of the video (vertical bar) moves to the left side of the raster where it belongs. This is further indication of too slow retrace.

Have you changed the capacitors in the horizontal sweep section? If so, recheck the values. If a capacitor is too large in value, it conceivably could reduce the resonant frequency of the flyback system, leading to too slow a retrace.

I shrunk the horizontal and vertical picture size down so I could see the edges of the picture, in an attempt to get a better idea of what the deflection circuits are doing. The horizontal linearity is off pretty bad and the horizontal lines still have a lot of wave over on the far left. For some reason the designers chose not to put a horizontal linearity control on this set. The only way to get decent horizontal linearity, is to adjust the horizontal width for a much wider than needed picture. Any idea's how I can adjust the horizontal linearity? I’ve tried several horizontal output and damper tubes with no change. :scratch2:

Hi Kevin,

Typically, with good tubes and good components (flyback transformer, yoke, resistors and capacitors), the linearity should be okay. From your earlier posted photos, the linearity did not look too bad to me. Can you outline or show by perhaps with a photo where the linearity is bad?

With the flyback horizontal sweep design, the damper provides the first half (left hand side) of the sweep and the horizontal output the second half (right handside) with the horizontal drive facilitating the best crossover in the middle of the screen. With this in mind, the pursuit of better linearity can be focussed on the particular part of the circuit.

Cheers,

Terry

Terry-

I'll post a few pictures using the sets 21AMP4B CRT. Will probably be an hour before I get them posted.

Ok...these pictures are all with the vertical and horizontal size brought down so we can just see the edges of the raster. The line patterns are from my B&K 1250 NTSC generator. The B&K target pattern is actually being played on DVD.

All the patterns seem shifted right from center when I have equal spacing on the outside edges of the raster. In the last target picture, you can see what happens when I center the image on the screen, the horizontal linearity issue becomes much more noticable as the target is move to the left.

Now if I adjust the width control for a wider horizontal scan, linearity does get better, but then the edges of the picture are also being cut off. Sams tells me to adjust the picture size until the image just fills the screen, but then the horizontal linearity issue is very noticeable.

I'm also still concerned about the vertical lines over on the left side of the screen. I'm guessing all these issues are related to the horizontal flyback system not operating at the correct frequency.

Kevin,

Sorry for the late reply. To be honest, the linearity does not look too bad to me. The horizontal sweep system in the early '50's is pretty crude by today's standards. However, the picture is pretty cramped on the left and I would suggest that the horizontal output tube is either a little weak or else some of the resistors around the horizintal output tube have drifted in value a bit.

Reducing the horizontal drive and increasing the width would probably provide the best compromise. The stretch on the left may also be due to too much horizontal drive.

I would be curious to see how the video fits properly within the video signal lines up on the raster eg. video starting 10us after horizontal sync. This would indicate if the retrace is timed correctly, that is the beam arriving at the left with a slight margin prior to scanning the active picture.

Terry,

You sort of lost me here. I'm not exactly sure what you're asking, or how to go about checking for it.

Thanks again for all the help, I'm learning stuff.

Kevin

Hi Kevin,

From your earlier photos, it appeared as if the horizontal retrace time was too long. That would mean that as the beam is swinging back to the left and if the active video in the signal begins to modulate the CRT beam before the beam reaches the left, then stretched ghost like images (including the white line) will appear in the picture. That is why when you adjust the horizontal hold, the white vertical line moves to the left when the picture moves to the right.

Note that about 10us is allowed to for retrace in the signal. The set has to to be able to retrace within the 10us.

A good test for retrace time is to see if you can see the edge of the active video on the left. When you shrink the scan or shift the raster so you can see the left of the image, up the brightness a bit and you should see a narrow black vertcal band prior to the vertical edge of the active picture. If you do not or if you see the image folded over, then something is wrong.

I only mention this because I had the impression something was wrong with the retrace time from your photgraphs. The same fault also may affect linearity.

Can you hift the raster to the right and photograph the left of the picture?

Thanks,

Terry

Hard to see in these pictures, but I don't see the black vertical bar on the left side. You can just make it out on the right side.

I've also noticed that the picture is off centered to the left while using a 8XP4 self focusing test crt. Even with the width turned all the way down(narrow), I can't ever see the left edge of the picture. So it does seem this set has a slow horiz retrace problem.

Hi Kevin,

That is interesting. One extra thing: your set uses the synchroguide horizontal oscillator control. The optimum sync position is with the horizontal hold turned to the right just before it loses sync. If the range of your horizontal hold is not sufficient, then you can turn the hold control fully right and slightly adjust the horizontal frequency slightly so that it just loses sync and back up so it is in sync.

If you have to adjust the coil, make sure you adjust slightly the frequency coil end and not the horizontal waveform adjustment on the other end.

Can you still not see the black stripe on the left? If not then we will have to investigate the retrace time.

Terry

Terry,

I'm not so sure this TV uses that exact horiz oscillator circuit. Take a look at the schematic below. I don't see the separate horiz frequency control you speak of. All I have are the horiz hold, and the horiz waveform controls. The horiz hold on this set is a coil (L40 on the schematic). RCA called this a deluxe model TV, but it sure seems to be missing a few key circuit adjusting controls. :scratch2:

Hi Kevin,

Thanks for the diagram. It is a synchroguide circuit. Only the horizontal hold control isn't a portentiometer in your set: the control is the coil itself. If I recall correctly, your set may have a limter to prevent adjusting it too far. Try adjusting the horizontal hold so that the image moves to the right at the point just before it loses sync. That is the optimum point in a synchroguide circuit and that should provide a good reference to examine the left hand of the picture.

Is it possible to post the damper circuit?

Thanks,

terry

Here's the damper section. Hope it's enough.

Ok... with the horiz hold adjusted so that the picture is farthest to the right(before loss of horiz hold), there is still no dark bar present on the left side of the picture.

I can see both sides of the picture as the set warms up and the picture slowly expands across the screen. There is clearly a bar on the right side.

I did recieve in the mail this week a set of Merit replacement yoke and flyback transformers. I've yet to try those. I'm not looking forward to replacing the flyback. I'll try the yoke after I wire up a new octal plug for it. I may also need to order a new balancing cap for half the horiz winding.

This is interesting: With the line voltage turned down to around 85-90 volts, we do seem to have a black bar on both sides. But I'm also seeing some waving of the scan lines over on the left. You can also clearly see the horizontal linearity is off.

BTW, this is now using the little 8XP4 crt.

Reducing the line voltage provides a clue. The reduced line voltage will lead to reduced horizontal drive. There might very well be a problem with the flyback transformer. But, I wonder if it is simply a matter of too much horizontal drive?

Have you tried reducing the drive and increase the width?

Terry

Terry,

I've wondered about the horiz drive being too strong. I normally have the drive cap turned all the way closed (min drive), but that could possibly still be too much. Any suggestions on how to further reduce the drive? Could I lower the value of the HOT grid resistor, or add more capacitance across the variable cap?

FWIW, I do have the correct voltages around the HOT.

Another thing: I have not replaced C105 (560mmf mica) because the horiz drive waveform looks fine. Plus I don't have a 560 @ 1kv.

In fact I have not replaced any mica's in this chassis, only the paper and electrolytics.

Kevin,

You seemed to have already done everything I can think of.

Not too long ago I had a similar problem on another RCA chassis. All the voltages were about correct and the horizontal drive was turned right down but the right hand side was compressed. I swapped the horizontal output tube which tested good otherwise. Same problem. I figured it had to be the tube.

I then bought a couple of NOS tubes from Ebay, swapped the tube again and "voila" the raster filled out on the right.

I know you said earlier you had swapped the tube. Was it a NOS tube and what is its condition?

Terry

Terry,

I have at least a half dozen nos 6DQ6's. Honestly I've probably only tried 2 or 3 besides the one original to the set. Never got anywhere, so I gave up on that direction, it's still a possibility.

But, if the right side "were" compressed, would that also be causing the left side picture to be cut off (no horizontal space bar?) I didn't think it would do that.

It sure would be interesting to have the actual RCA service literature for this chassis. Possibly there are notes about scan issues.

Does anyone know what was used for pincushion correction (if any) on this TV? Watching this 8" test CRT, the pincusion effect is really bad.

Remember that the horizontal output tube only begins to conduct from the middle of the scan to the right of the screen. If the tube is weak then the right of the screen would be compressed. The fact that the voltages are normal and you have a healthy scan otherwise points to the output tube.

Is the 6DQ6 screen voltage correct? Perhaps the screen resistor has drifted upward or the screen decoupling capacitor is faulty?

I do not think the RCA Notes will share this information. However I do recall some good articles surfing the web. I just can't seem to find them at the moment...

Small permanent magnets can be used on black and white set to correct pincushioning. They will be placed either side of the yoke near the CRT.

Terry,

I'm sorry for the confusion. I don't think I've ever had a problem with the right side of the screen being overly compressed. The reason the right side is currently compressed is because I have the width control turned all the way out. I only did this in an attempt to see the dark space bar on both sides of the picture. If you recall (before I sidetracked the issue) we were discussing that the picture information should start scanning after that dark space on the lefthand side. The width control does have an effect on horiz linearity, but my main concern is that I can not narrow the picture width enough to see the left most edge of the raster. I just thought it curious that the picture is not centered on the 8XP4 test CRT, but it's probably nothing unusual.

I still think that your original theory is correct: that the horizontal retrace time is too short. I think possibly this is also shifting the picture to the left of screen center. Does that combination make any sense, or am I still confused?

Hi Kevin,

I suppose I am now a little confused. I recall the original problem was that the center of the horizontal scan is to the right. This means that the left side is stretched and at the same time the right side is compressed. Is that correct?

By the width control being all the way out, does this mean the core of the width coil is out so you have minimum width? Then I suppose overall you have the necessary width adustment range but the left side is still stretched.

This would suggest to me you have too much horizontal drive, even with the trimmer capacitor tighted right down. Is this correct?

The Horizontal drive control is part of a capacitive divider. Perhaps there is something wrong with the trimmer? You should be able to adjust the trimmer so that as it is tightened, the left side of the raster slightly compresses and the CRT will dim slightly as the HV drops.

Maybe you should test the trimmer? Or as a test to reduce drive more, you could put a small capacitor (50 to 80 uufd) in parallel with the horizontal drive trimmer?

Terry

Correct. Horizontal center was originally off to the right using the sets 21AMP4 CRT. When I changed to the 8XP4 test CRT, then I discovered the opposite true. Also, with the 8XP4, I'm not able to see the left edge of the horizontal scan. I suppose I should forget using the 8XP4 for these centering tests and just go with the 21AMP4?

Correct, but really only became an issue when I switched to the 8XP4 test CRT. And this extreme reduction in width was only in an attempt to see the left scanning edge.

I'll check into reducing the horizontal drive even more and report back. I've never seen any noticeable compression on the left side, even with the trimmer completely closed. I'm pretty sure I've never seen the CRT dim with it fully closed. I know the trimmer checks ok, but I've yet to check the value of C105 (mica 560 pf). Using the sets 21AMP4 CRT, I generally get the best horiz linearity with the trimmer fully closed. This would seem to indicate there is still too much horiz drive.

I'm still trying to make sense of this horizontal scan issue.

I did try lowering the horizontal drive some more. That did improve the horiz linearity some, but then the picture collapses before it reaches optimum linearity. The poor linearity doesn't bother me so much as the wiggle in the horizontal scan lines over on the left screen.

I've also installed a NOS Merit flyback and yoke. Amazingly nothing changed with the picture, other than the Merit yoke causes significant barrel distortion in the picture. Not sure what causes that, so I'm currently using the original yoke and the new Merit flyback. I've even tried changing value of the yoke balancing cap across the top half of the vertical winding. Also tried replacing the damping resistors across the horizontal windings, even with lower values. At least a half dozen NOS damper tubes have be tried, and they show very similar results.

Not sure where to go from here. :scratch2:

Pictures below are with the horizontal width reduced. It does look like there is horizontal blanking on both the left and right sides of the picture. You can also see light and dark vertical lines, typical textbook examples of horizontal ringing on the far left. It's hard to see in the pictures, but I can also see wider light and dark vertical columnes all across the screen.

Kevin,

My guess is the wavy scan on the left is due to imbalanced distributed capacity on the horizontal deflection coils. Check the small capacitor: it should be between 50 to 100uufd and its value cancels the waviness.

The vertical stripes you call ringing I guess may be the unblanked burst on retrace. You can confirm this by adjusting the horizontal hold back and forth: if the stripes move in the opposite direction to the picture as you turn the control, then it is the burst.

All the best,

Terry

Hi Terry,

I've tried adjusting the horiz yoke balance capacitor from its stock 150 pf value, all the way down to nothing. Making it smaller does lessen the waviness, but it's still there. This TV has its keyed AGC pulse taken from the yoke windings. I'm starting to wonder if those associated capacitors aren't the cause of this whole issue. :scratch2:

Is there a practical way to completely blank the retrace burst you speak of?

Terry,

Those vertical stripes do not move with the horiz hold control. I suppose you could say they move opposite relative to the picture movement, but the lines themselves do not change location on the CRT.

Man, that is a brain teaser. From the FWIW department regarding the 'ringing', here's what I would try at the present stage:
Get a disc ceramic cap rated at at least 4KV, around 82pf (anywhere from 68 to 100pf will do). put it directly across the damper, plate to cathode. This will increase the width just a smidgin*. It may (or may not) have some noticable effect on the 'ringing'. Now connect the cap from the cathode (pin 3) to the other side of L39 and see if there's any effect (other than the increased width). If nada, try the cap directly across L39.

If any of those three arrangements had any effect on the ringing, either increasing or decreasing it, it will have at least localized the problem area.

*Besides increasing width, this will also reduce H.output tube cathode current slightly, and reduce HV slightly. Thus you can't go much higher than 100pf with the cap.
oc

OC,

Will try your suggestions.

Last night I unhooked the keyed AGC coupling components form the yoke socket, then hooked a battery to the AGC line. Ringing still persisted, but it did change the frequency slightly.