Phil, I'm also fighting a chroma-AFC problem in a friend's CT-100. It exhibits a similar off-color operation. In my case the oscillator is functional and the CW to the demodulators is there and at the proper amplitude. And, as well as can be seen on a scope, the CW is in quadrature at the demodulators.

As has been suggested here with your problem, I will next concentrate on insuring the Reactance inductor and circuit are working properly.

Closed loops are tough cookies to crack at times!

Pete

Thanks, I did that very check yesterday afternoon. It occurred to me that I already had a socket issue on the video IF board, so why not check this one? This socket seems solid, knock on wood.

I will try to follow up on the other suggestions. It would be nice if I could just substitute a 3.58-mc signal, as you could do with a video or horizontal drive signal, but this situation does not seem that simple.

Phil Nelson

Why not try a 3.58 MHZ signal from a signal generator? disconnect the
crystal, feed the signal from the generator to the no-longer
oscillator's grid and look on the scope at the two outputs of the
output transformer. See what happens if you vary the frequency around
3.58 MHZ (on the scope).

Another thing is to check the crystal ... connect the sign generator
to it with a 10K resistor and look on the scope. You should see the
sharp resonance as you tune. Remember that a previous thread had
what was thought to be a bad crystal; they tried two good crystals and
they didn't work. It was a bad part. After replacing the bad part
but leaving in a "good" crystal it didn't work ... but with the original crystal did.
So these circuits are picky about the crystal. If the crystal is bad, it may
be that only an exact replacement will work.

Hey, I just remembered that my Sencore VA62A video analyzer can supply a 3.58-mc drive signal. I hauled it off the shelf and fired it up. My scope has a hard time locking on the signal, but there seems to be a waveform of sorts.

http://antiqueradio.org/art/SencoreV...58mcSignal.jpg

I disconnected the crystal and connected the analyzer's 3.58-mc drive output to pin 6 (grid) of the oscillator tube. After adjusting the generator's output voltage, I began to see rolling colors moving upward on the screen, similar to what I've seen before when the color sync was messed up on my CT-100.

http://antiqueradio.org/art/RCACTC-4...rsAnimated.gif

I tried adjusting the reactance coil (L127 in the RCA manual) through its entire range, but this had no effect on the rolling, which never changed speed. And I still don't see normal colors, only rolling patterns against the same green-blue bar background. Turning the TV's Color control had the same effect as before: no change in the colors, and the image will just extinguish if you turn Color down far enough.

The symptoms are the same whether I send the color-bar signal from my pattern generator through the TV's tuner and (lousy) video IF sections, or inject video directly at the 1st video amp.

I'm going to go back and repeat the procedure outlined on page 22 of the RCA manual, disconnecting the bandpass amplifier, trying to adjust T115, and so on.

Phil Nelson

Its obviously best to use an accurate 3.58, but even the cheesiest signal generator will get close enough to test throughput ... just not phase.

Ignore the reactance tube for now, or at least keep the input grounded for the moment per the adjustment procedure so you can focus on getting the oscillator running. It's obvious you have no output from it, that's why you have zero grid volts on V124A and also sagging plate volts- that tube is probably turned on 100%. Make sure the .01uf caps to ground coming off pin 2 are not leaky (C2110/C211), then see if you can lift one of the primary winding wires of T115 (pin B or E) and measure resistance. It's possible your CW transformer has shorted or something, but tuned circuits can be deceptively hard to troubleshoot despite their apparent simplicity.

One last note: although the inductors in the CTC-4 chassis are not the dreaded white ones that always go bad like the CTC-2(B) chassis had, it's always possible you have a bad inductor still. Make sure L42 checks in at the proper resistance, and also ensure C209 isn't leaky.

Thanks, we hear and obey :)

Phil Nelson

There's also a couple of .001 caps (C219 and C220) across the secondaries of the CW xfmr. One of those could be wonky.

I believe C219 and C220 are inside the T115 can, but as chrisWA6TV mentioned, little micas inside xfmr cans have been known to go bad, especially the "naked" kind that are uninsulated mica sheets. I replaced some in the audio section of an Admiral set.

My back is just about played out for the day. Tomorrow I will grunt the chassis back onto the workbench and go on a cap-and-resistor hunting expedition.

miniman82 mentioned C209, which I already replaced, but that reminds me -- depending on what schematic I look at, that cap's value is given as 4, 4.7, or 6.8 pf. Here's the C209 that I removed:

http://antiqueradio.org/art/temp/RCACTC-4C209.jpg

I hate the codes on these little dogbone-shaped caps (and I have never found a bad one, but I replaced it anyway, while I was soldering on that terminal). What I had on hand was a 5 pf mica, so that's what I installed. I'm hoping that this code translates as 4.7 pf and that a difference of .3 pf doesn't matter here. If I'm wrong on one or both counts, someone tell me now, and I'll just put the old one back in.

Phil Nelson

P.S. The resistance across L42 is 10.6 ohms, which doesn't seem outlandish. The schematics don't give a resistance for it, only the notation 12 muh.

Check the CW trans, T115. It's a simple circuit, but the answer isn't always obvious.

Back in post# 149 you wrote, Resistance from pin 3 to ground is zero, suggesting that L42 has continuity."
Since L42 goes to pin 3 (cathode), pin 3 should be showing 10.6 ohms to ground.
Were you measuring the coil with one end disconnected? If so, could there be a short inside the socket from pin 3 to ground?

Sorry, I've been flipping back & forth between the Sams & RCA schematics and jumbled some numbers.

The 10.8 ohms reading was across L127, the reactance coil. That is with everything connected on that coil.

Resistance from pin 3 to ground is 2.1 ohms. That's across L42 with everything connected. The RCA manual has no resistance chart, but for what it's worth, Sams says resistance at pin 3 should be .4 ohms, where I now measure 2.1 ohms.

I did some checking of T115, the CW transformer. Resistance across the primary (from B to E) is 7.0 ohms. Resistance from F to A (F to ground) is .1 ohms. Resistance from D to C (D to ground) is .1 ohms. This is with everything connected to T115.

Someone has worked in this area before. C202, the 120-pf cap on pin 6, looked like a replacement, attached hook-and-loop style to a stub left on the terminal. R246, 22K, is another obvious hook-and-loop replacement. The crystal may also have been replaced; one of its leads was much too long, and with strange bends -- not neatly dressed as you'd expect from the factory.

Phil Nelson

The schematic indicates there should be three ferrite slugs inside the CW xfmr. Are all 3 present and intact? No doubt they've been diddled with if somebody's been in there before.

Also, you stated R246 is 22K, where the schematic shows it as 2.2K.
Is the replacement red-red-orange (22K) or red-red-red (2.2K)?
In any case, that would not stop the oscillator from running.

You're right, it's 2.2K (red-red-red). When I lose the ability to read and type, that's a sign to break for the night :)

T115 has one adjuster slug accessible from the bottom and a second accessible from the top. The RCA manual says that the bottom adjuster is used in color AFC alignment and the top one for B-Y phasing. Both slugs are present and can be moved (the settings may be completely wrong, of course).

The Sams schematic seems to show that the two secondaries are adjustable, labeling them A64 (top) and A60. No label is given for the primary. I don't know why the primary is shown with an "adjustable" symbol if no adjuster is accessible.

Phil Nelson

Could it be a slug is missing from the primary? Or is there a big enuff hole to allow its exit?

(Edit) On second thought, it does look like an untuned primary, since there's no cap in parallel with it.

Only the secondaries are tuned, not sure why the schematic shows it is. Never caught that before, but it doesn't matter. Keep looking for bad parts, I'm sure you'll see something obvious sooner or later.

My guess is that the problem is in the cathode and grid area of the 3.58 MHz oscillator. If possible measure the component values, especially the cathode inductor L42 , its capacitor C213, the 4 pf feedback cap C209, the crystal and its resistor. Or maybe some conductive goo on the tube socket pins? My 21CT55 oscillator cathode tuned coil had been mis-adjusted and caused the oscillator to stop often.

I will keep plugging along. On the positive side, I tried dtvmcdonald's trick to test the crystals, and they all look good, including the one that was installed in the set. Each of them shows a distinct peak on the scope at around 3.58mc.

I replaced a couple more components around that tube, without making any improvement. This updated schematic shows the 5 caps and resistors replaced there.

http://antiqueradio.org/art/temp/RCA...aticUpdate.jpg

None of those components was grievously out of spec -- well, R249 measured 54K instead of 47K -- but at least I can cross those off the list of possible suspects.

Of the remaining components connected immediately to that tube, we also can eliminate the crystal, which tested OK, and resistor R248, which measures a hair over 47K.

Edit: I just saw the reply from Zenith6S321. One thing I haven't tried yet is disconnecting L42 to measure it. I can try that later on, along with giving the socket another good hard cleaning.

Phil Nelson

OK, I tried this (substituting the 3.58-mc signal from the VA62A analyzer at the grid of the chroma reference oscillator tube). It looks as if T115, the CW transformer, can pass the signal and be peaked.

http://antiqueradio.org/art/temp/RCA...15Animated.gif

In that photo, the scope was connected to terminal F on T115. By turning the bottom adjuster of T115, I could increase the amplitude up to a point, beyond which it fell off again. When I moved the probe to terminal D, the waveform looked about the same, although I didn't try the top adjuster.

None of which helps me figure out why the oscillator doesn't oscillate, but I'm taking it as evidence that T115 is operational, anyhow.

I had to move the adjuster quite a long way from its original position to find the peak, but perhaps that isn't surprising, given that many of the components attached to T115 have been replaced.

Phil Nelson

By sheer elimination, it begins to look like maybe the resonance point of the L42/C213 tank circuit is too far off from 3.58 to allow oscillation.

Do you have a signal genny that's reasonably well calibrated, and can be manually swept thru 3.58mc? If so, you might try sweeping the tank and watch for the resonant point on the scope.
(But first disconnect the tank from the tube cathode and C209. Connect the genny and scope to the top of the tank.)

My signal generator is a rotten old EICO 324. You can sweep it manually, but calibration is a laugh.

Those two components are small and rather buried.

http://antiqueradio.org/art/temp/RCACTC-4L42C213.jpg

Seems simpler just to replace L42 unless it is unobtanium.

Phil Nelson

Try sweeping the combination of L42 and the 270 pF in parallel with it.
It should be close to 3.58 MHz. You could try disconnecting the 270 pF
capacitor and connecting in a standard 365 pF variable, and
varying that to see if you can get oscillation.

That's what I meant by 'tank' - both coil and cap in parallel.
Ha. I was going to suggest that very thing, assuming he's got an old AM radio tuning condenser knocking around somewhere.

Kinda wonder if the coil could have some shorted turns in it, considering the uber-fine gauge wire it's wound with.

Here's another idea that's sort of a 'Hail Mary' gambit..
See if the osc. can be made to free-run on just the crystal. Disconnect the LC tank completely and replace it with a resistor (leaving feedback cap C209 in place). Try several values of resistor, from maybe 47 ohms thru 560 ohms or more.
If it'll start and run, that would pretty well confirm the fault lies with the LC combo.

I simulated the circuit. The cathode tank needs to resonate
a bit below 3.58 MHz. Apparently its doing some phase shifting.
It seems non critical, 2.5 to 3.2 seems to work.

It's difficult to sneak a soldering iron into the place where those two components connect to the tube, much less other tools. In the event that I fry or mangle L42 in the course of experiments, what's a good replacement -- something like this?

http://www.mouser.com/ProductDetail/...tRRhUO5QN1Y%3d

Phil Nelson

I would say go for it. The extreme disparity in DC resistance is no doubt due to its having a ferrite core vs. the original's air core (hence fewer turns of wire).

I disconnected C213 and L42 from the oscillator tube, and when I lifted up L42, it looked scorchy:

http://antiqueradio.org/art/RCACTC-4L42Original.jpg

There was a layer of melted varnish sticking it to the green wire below. (I scraped that off before taking the photo.) The coil is missing varnish in the area where they touched, and some of its wires look burned black. It has continuity, measuring 2.6 ohms resistance. Not having another coil to compare, I have no idea what that figure means; if it has shorted coils, I suppose it could show continuity but have the wrong inductance.

I tried doing this, with inconclusive results. I can see a waveform, and I can see the waveform change in the X dimension when I sweep the generator past 3.58 mc, but I don't see anything like a peak.

It's possible I had things set up wrong. What I did was to connect C213 and L42 in parallel (but disconnected from the tube). Then I connected the sig gen at one end of this pair and the scope at the other end. I slowly changed the frequency of the sig gen and watched for a peak of some kind. Again, with my crude old generator, it's not possible to s-l-o-w-l-y vary the frequency around that exact point, or even hold it precisely, except at whatever frequency it flops to when you let go of the knob.

Phil Nelson

:eek:Dang. That swatch of carbonization looks like it involved a half dozen turns or more. That could sure play hobb with the 'Q' factor as well as alter the inductance.

Re. the sweep test, what you'd wanta do is leave coil and cap in parallel, with the lower end grounded. Then connect both genny and scope to the top (with their cable shields grounded, of course).

A parallel-resonant LC tank (which this is) exhibits highest impedance at resonance, grounding out all other frequencies.
But a series-resonant tank (L and C in series) exhibits lowest impedance at resonance, making it useful as a wave trap to ground out an unwanted signal.

OK. I visited the local surplus store and found a couple of 12muh inductors in a bin. Later today I'll try the sweep test on old & new parts and see what transpires.

Phil Nelson

To get the sharpest blip, you'd probably wanta use a small coupling cap from the gen to the tank. Maybe 10 pf or so.

Is this what you're talking about -- 270pf cap in parallel with the inductor, and 10pf on the end?

http://antiqueradio.org/art/RCACTC-4...ankCircuit.jpg

Seems to do nothing around 3.58 mc, but when I went lower, I could get a waveform showing up between about 2.3 to 2.7 mc, peaking in amplitude around 2.55 mc.

http://antiqueradio.org/art/RCACTC-4...rmAnimated.gif

Out of curiosity, I made another little circuit using the old L42 and two more caps (10 and 270 pf). The response was basically the same.

http://antiqueradio.org/art/RCACTC-4...42Original.jpg

So, either the original L42 was good after all, or I still don't have things set up correctly. (Or it could be some evil combination of antique signal generator, middle-aged scope, and antique brain :) )

Meanwhile, I installed a new inductor in place of the old L42 in the TV and reconnected things to the oscillator tube. No change or improvement. Still seeing 0 volts at pin 6 of the oscillator.

Phil Nelson

some obvious stuff here, is the tube socket good (use a test socket to make sure the tube pins are connected to the circuit. Different tube used? correct tube used? just trying to cover the bases.

Here is a crazy thought what if there is a tin whisker or some such shorting the grid to ground forcing it to be 0V?

Is that a low capacitance scope probe? If not, it may be pulling the tank freq down too much. To minimize loading, you could add a small cap in series with the probe. Just a 'gimmick' consisting of a few inches of wire twisted together should do it.

If you need to raise the freq, try a lower value tank cap. Maybe 100 pf to start, and see which direction you have to go from there.

It is a plain old direct scope probe. I tried adding a 2 pf cap in series with the probe and didn't notice a significant change.

I tried substituting a 100 pf and a 82 pf cap for the 270 pf cap in the tank circuit on the workbench. 100 pf raised the peaking point somewhat, and 82 pf raised it up to roughly 3.5 mc. I substituted an 82 pf cap in the TV, and observed no change or improvement.

I don't see any whiskers or solder crumbs anywhere on the socket, which I have cleaned and re-cleaned. I have tested the socket with an adapter in place, and tried running the TV with the tube in an adapter, with and without the shield.

When I tested the old coil (L42) in a tank circuit on the workbench, it responded the same as a new one. Substituting a new coil for the old one in the TV gives me the same result.

I guess I can't rule out a socket defect such as a hidden carbon track caused by a past shorting incident. That is a hard thing to disprove without stripping everything off the socket, at which point you may as well simply replace it.

Phil Nelson

try another Xtal?

geez this is going to be great when you find it (trying for a positive spin).

sorry a little late to the party, but was the 270pf the orig cap in the circuit?

I have tried substituting crystals (and testing those crystals on the workbench). Also swapping among my four available 6AZ8 tubes.

The 270 pf cap is new. If you go back a page or two, you'll see that I already replaced most of the components in the oscillator circuit. Nothing left to replace in the immediate vicinity except R248, the 47K resistor, and I measure 47K resistance at pin 6, as per the resistance chart.

Yes, I'm looking forward to making this set work . . . someday. It's pretty interesting, so far. The CTC-4 is one of those stops on the road between the CT-100 and later RCA color sets that I've worked on. When it gets frustrating, I just work on something easier for a while. The TV will still be there waiting, next day -- or next month!

Phil Nelson

what I was getting at was did you confirm the cap value against the orig? for value that is, just in case the schematic is wrong.

You could always ohm out the grid pin with respect to ground and surrounding pins to look for abnormalities....

one more, confirm the heater voltage.