Today's Mystery Guest (RCA CTC-4)

[old_coot88]

If still wonky after adjusting T115 cores for best display, see what effect adjusting T114 core has.
Also check voltage on V125 plates (pins 1 and 6). Should be about 36V.
If not correct, check resistors
-
R256 (110K),
R258 (180K)
R262 (150K)
R283 (47K),
R282 (3.3K)
R286 (3.3K)


.

[Phil Nelson]

I spent some time adjusting T115 with two diddle sticks while watching the color bars, but didn't reach Nirvana. Lots of colors, but no clear path to the right ones. I think there is more going on here than mere misalignment. The lock doesn't seem very strong, either. The bars were prone to start pulsing, and tweaking the reactance coil did not always cure that in an obvious way.

I checked some voltages on V123, the 6AG7 demodulator driver:

pin 4: 0 (0 expected)
pin 5: 7.1 (6.2 expected)
pin 6: 175 (230 expected)
pin 8: 248 (280 expected)

The "expected" voltages are from the RCA schematic; the Sams manual gives lower voltages for pins 5 (5.5v), 6 (200v) and 8 (250v).

T114's secondary has continuity, measuring about 4 ohms from terminals A-B.

I checked the plate voltages of V125, the 12BH7 R-Y/G-Y demodulator:

pin 1: 33v
pin 6: 16v

I tried scoping the signal coming from T113, the bandpass transformer, and following it through the demodulator driver transformer (T114). Here is the waveform at the grid (pin 4) of the 6AG7 demodulator driver:



Here's the waveform at the plate (pin 8) of the demod driver:



Here's the waveform at terminal A of T114, the demodulator driver transformer:



And here's the waveform at terminal D of T114:



Out of curiosity, I put the frequency counter at these points. For what it's worth, it measured 3.80 MHz at the plate of the demod driver tube, 3.69 MHz at terminal A of T114, and 3.74 MHz at terminal D of T114.

Next stop: checking resistors.

Phil Nelson

[Phil Nelson]

Originally Posted by old_coot88

check resistors
R256 (110K),
R258 (180K)
R262 (150K)
R283 (47K),
R282 (3.3K)
R286 (3.3K)

I checked those (in place) without seeing anything alarming.

Maybe I will break down and install the replacement quadrature transformer. But first, I want to mess around with the breadboard circuit a little more, to see how close I can get to the ideal frequency. If there's any component tweaking to be done, I'd rather do it on the breadboard.

Phil Nelson

[old_coot88]

Originally Posted by Phil Nelson

I checked those (in place) without seeing anything alarming.

Did you clip one end loose to check them? ('Specially the high value ones. The 3.3K ones not so much.)

I checked the plate voltages of V125, the 12BH7 R-Y/G-Y demodulator:

pin 1: 33v
pin 6: 16v

The 16V is 'waay low, less than half the expected. What about the grid voltages (pins 2 and 7)? Both expect the same, about -20V.

[Phil Nelson]

I lifted those resistors to re-check, and didn't find anything more than 10% off spec, but I replaced them anyway. Now I have replaced most of the resistors on the video board hanging off the side of the chassis.

None of the voltages on the 12BH7 (including plate voltages) are very close to spec. These circuits have a bunch of 2-watt resistors connected to the +380V line. I don't keep many 2-watt resistors on hand, so I'm off to the local store to restock.

Phil Nelson

[old_coot88]

Holy !@#$%^&. I didn't see those. Are those the three paralleled 82Ks feeding the pin 1 plate? Your voltage reading there was near spec.

And I didn't see R259 (120K) feeding pin 6 plate, which is 'waay low.

[Phil Nelson]

And don't forget the two paralleled 82Ks on pin 7 of V124B (B-Y amp). Kind of ugly construction in that neighborhood. Maybe they got a good deal on 82K/2W resistors.

Phil Nelson

[old_coot88]

Yeah, and considering that 2W carbons are know to char and drift downward, it'll be interesting to see what those actually read.

[Phil Nelson]

I continued the long resistor hunt, and did find some bad ones. This pair of 82K/2W resistors should measure 41K, not 122K:



Most of the other 2W resistors look like a different brand -- the usual dark colors -- and they looked pretty good, but many got replaced anyway.

After those replacements, the voltages on the 12BH7 demodulator looked closer to normal. The voltages on the grids reflect the amplitude of the signals coming from T115, the 3.58-MHz transformer, so when you adjust T115 in the course of alignment, they also change.

At that point, I had basically the same issue -- colors in the color bar pattern were incorrect, and no amount of adjustment made them right.

The replacement 3.58-HMz transformer had produced reasonable-looking waveforms in the breadboard circuit, so I decided to pull out the old transformer and try it.

Comparing the transformers side by side renewed my doubts. The new transformer has a lot more windings, closer together. Remember, it also has 680-ohm resistors in parallel with each capacitor, and the capacitors are 1500 pf, not 1200 pf as in the original.



The new transformer fits physically, but I had to reroute a couple of connections, since its output to the phase detector is terminal D rather than F, etc.



(After installing it, I did reattach that hanging blue wire, seen in a previous photo running across the installation site.)

The bad news is that it doesn't work -- not even close. I ran through the color AFC alignment procedure and tried other cowboy adjustments while watching the screen, but one of the waveforms coming out of the transformer is very irregular and the two signals are badly mismatched in amplitude. Perhaps it is expecting an input signal level very different than what this TV provides. What I see on the screen is a very washed-out image of "colorless bars," with very faint indications of rolling bars (or whatever).

After double- and triple-checking everything on and around that transformer, I can't find anything wrong apart from this simply being an unsuitable part. I think I will pull it out and go back to the original. If the windings on this transformer looked more like the original ones, I might try removing the 680-ohm resistors and substituting 1200-pf caps for its 1500-pf ones. But that might just be time wasted.

Phil Nelson

[old_coot88]

Just a shot in the dark, but did you inspect lead dress carefully for any wires shorting? Looks like possibly a melt-thru of the insulating sleeve on that resistor going to lug #2 of the tube socket.

[Phil Nelson]

It's not beautiful, but I have checked it all carefully and there are no shorts or wiring errors.

Phil Nelson

[Username1]

I think your most telling post came around #134. In that post you put up
screen shots of the O'scope of the input of the video amplifier. This was also
smartly compared to a different set, a ctc 11, not that it matters.
What it does say, is that your scope, scope probe, are not distorting, or
un-necessarily loading the circuits.

The scope image also says to me that there is serious overshoots on vertical
rise items in the circuit. This can cause ringing, which could also be causing
harmonics, and other distortions around the burst signal as well.....
You should fix this first, I think it's giving you trouble down the line...

I think your problem is not in the color osc circuit, but before it..... You need
to figure out why your video detector circuit is allowing much higher
frequency response (is what I believe is going on) This could be an open
coil, or open cap in and around the Video IF and video detector area...

PS You already showed that your color osc works on yer breadboard
thing. Feed that osc a buncha noise and I bet it quits, or looks like
it quits. Your freq. counter acts like it does when noise is present
too, locking on different frequencies.....
It would be neat if'n you had a spectrum analyzer to see what's
getting to the vid. amp and ocs. grid.

just my 2 nuts worth....
(not those nuts)

.

[Phil Nelson]

Thanks, I am open to all suggestions at this point.

After I saw the signal coming out of the video IF section, I disconnected that output and injected a video signal from my pattern generator. I'm injecting it at the grid (pin 2-9) of the 1st video amplifier, V110. The idea is to get the chroma circuits working first, and worry about the video IF later.

I don't have a spectrum analyzer, unfortunately.

At this point I'm thinking you're correct, that the oscillator and color lock are working reasonably well. As I understand the lock system, the local oscillator puts out its 3.58-MHz signal and the phase detector compares that frequency to the incoming 3.58-MHz signal. If they are different, the reactance tube pulls the oscillator up or down to match the received frequency.

Here's a little test that I tried after getting my frequency counter. The counter showed different frequencies (at the oscillator) when I used different sources for the injected video input. I tried this with two pattern generators and a DVD player. I assumed this was because the 3.58-MHz component from each of those sources was a little different, and the CTC-4 was able to lock on those slightly different frequencies.

Then I connected those different sources to my CTC-11 and compared what I had measured at the CTC-4. For instance, here's my Leader pattern generator feeding a signal to the CTC-4, which appears to lock onto 3.579673:



And here's the same generator feeding a signal to the CTC-11, which locks onto 3.579679:



That is the same frequency out to five decimal places. The results were similar when I compared other sources with both TVs.

This suggests that my CTC-4's oscillator is running and that it can lock on the same incoming frequencies as my CTC-11 (which has great color).

Which leads me to think that its problems lie elsewhere -- maybe in the demodulator circuits downstream, or, as you say, somewhere upstream, where the incoming signal is getting attenuated or mutilated. The chroma circuits seem to work best with the strongest injected signal that I can provide. The schematic says that the CTC-4 expects +2.7V at the grid of the 1st video amp, and none of my equipment can provide such a strong video signal. When people want to inject video on a CT-100 or CTC-2B, it's necessary to provide a little video preamp.

Anyway, I'm open to fresh ideas. Yesterday I reinstalled the original quadrature transformer, which works just as it did before.

Regards,

Phil Nelson

[Username1]

Yah, from what I read, I don't think it was in that transformer.....
Bad looking signal at post 134 means the problem is not after it,
it's before it..... Trying to work on down stream stuff is really pointless
if the upstream circuits are feeding you trash....
That's just how I see it.......

.

[old_coot88]

Another shot in the dark, Phil. Have you tried subbing the vid detector diode itself?