Got a new oscillator crystal courtesy of Phil Nelson (Thanks a bunch). I got it awhile ago in fact. I just haven't had time to get back to the set until this weekend. Hopefully replacing the bad crystal will clear up some problems. It's been almost a year working on this set and i've still got a ways to go. :thumbsdn:

My thoughts exactly as I looked at those screen shots. I’m by no means a color TV expert, but I do have a Zenith roundie that had the slightest arching in the HV cage. It was hardly audible, but that was exactly what the picture looked like. Turned out that the HV rectifier socket was fried. Replaced that and no problems. In addition, I think that set needs degaussing. Most of us aren’t lucky enough to have a real serviceman’s degaussing ring, so I just use my Weller soldering gun in the same manner that you would use the ring.

I've got a real service degaussing coil and I've made spares....Allz you have to do to make one is find a decent sized 20" or bigger BPC set at the curb, steal it's degaussing coil, loop that 3 times around so it becomes a smaller coil ~12-14" in diameter, wrap it in electrical tape so it holds that form, and attach a cord.
...It will get hotter faster than a servicing coil, but one can get about a minute run time (plenty for most sets) out of it before it needs a cool. I made a thread on the process a while back.

I think that this handy tip should have been made a sticky:

http://www.videokarma.org/showthread...hlight=degauss

jt

I've made it into one.:thmbsp:

I picked up a Radio Shack bulk tape eraser, that works great! It's a black plastic unit, that looks like a travel iron. It's a real powerful item, that can be used to demagnitize, just about any kind of magnetic media. It also works well as a deguasser, even though, I still have the first RCA DG coil, I bought over 50 years ago. :scratch2:

I too use a Radio Shack bulk eraser. It is very powerful and does the job.

Put in the new crystal and things have improved I believe. The oscillator isnt running at the correct frequency as indicated by the lack of color sync. I dont see nearly as much interference anymore. I did blow out the hv cage with compressed air and moved 1 lead that was somewhat close to a solder joint. I'm not sure if I should adjust L127. Fine tuning isnt able to lock the color and neither is the color control. I think I may have to adjust L127 (reactance coil). Still have no red. :thumbsdn:

14 turns counter-clockwise on L127 made the rolling bars a lot higher, but I still cant get it to lock. I reverted L127 back to its original position. I need to hook up the frequency counter and see what the oscillator is running at.

I connected the frequency counter to the oscillator. It measured out at about 3.58MHz which was good. The target frequency is 3.57545 so I went off of that. The lowest I could get the oscillator by turning L127 (reactance coil) was 3.57970 which was plenty off. I think I need to add a cap in parallel with the 270 pf cap on the oscillator tank circuit. Your thoughts?

Do you know if the capacitors are OK
C201 2p
C202 120p
C209 6.8p
C203 .01
C213 270

How about the reactance tube itself (6AN8)?

The reactance tube works by amplifying the effect of C201 by the variable gain of the tube.

You can check if the oscillator can be pulled by putting, I don't know for sure, but maybe 100pF, from the plate of the reactance tube to ground. If this can pull the oscillator to a lower frequency, and the reactance tube cannot, the problem is in the reactance tube circuit.

I will check the tube first. This is somewhat confusing though because the previous crystal was able to oscillate at the right frequency and maintain color lock until the crystal broke. And after doing some reading around it seems Phil Nelson had this same problem with his 4 and adding a cap in parallel with the tank circuit allowed the correct frequency to be reached.

I think you are right to try the tube first. If that's not the problem:

A crystal for this circuit is cut to parallel resonate with a certain net amount of capacitance. That is, the crystal is operated above its series resonant frequency, and behaves like a large inductor, and is tuned with a small parallel capacitance. If the tank coil is tuned right on frequency, the net capacitance is supplied by the reactance circuit through the 120 pf C202.
The tank being on the far side of the 6.8 pF, it takes a large change in the tank tuning to change the total parallel capacitance across the crystal. I don't doubt that it worked, but a more direct way might be to put a small cap (5-10 pF) directly across the crystal. This way, the tank doesn't have to be detuned so far to get the required net capacitance. Adding capacitance across the tank means you are detuning it from exact parallel resonance to be capacitive when the oscillator is locked, while I think it is really intended to be more or less exactly parallel resonant for oscillator operation.

Different TV manufacturers may have specified different parallel capacitances, for nominal parallel resonance of the crystal. (probably between 18 pF for Zenith [if i recall correctly] and possibly higher values [maybe up to 32 pf but probably in the 20s] for other manufacturers). So, it's possible the crystal you got is not cut for the net capacitance in this chassis, and some more capacitance is needed. This should come from the reactance circuit, not the tank circuit, which is why you tune the reactance coil to zero beat the oscillator.

I tacked on a 5 pf cap across the crystal and I can almost get the oscillator to run at the right frequency. It still wont lock even after 20 turns on the reactance coil. Here is what I could get (note the bars rapidly change color every second but I no longer have the barber pole rainbow.)

Ok, now you have a chance to find out what part of the loop is dead.

If you have a scope, connect to the top of C198 (4700). If you don't have a scope, connect a DC voltmeter there. Center the AFC balance control. Tune the oscillator so the color cycles through slowly, and you should see the voltage cycle slowly also, as the phase detector cycles through its full range. I don't know what the normal range is, but it should be at least a few volts variation, maybe +/- 10 from nominal zero at most? [Note that the schematic says the nominal control voltage on the grid of the reactance tube is zero volts, so the phase detector output should vary +/- around zero when it is out of lock.] If there is no variation, then either the burst keyer or the phase detector may be dead. If there is a variation, then your problem is still in the reactance tube or oscillator.

I checked voltages and waveforms at C198 which is a .01 in this case since I have the early production run of the chassis. I got about 10 volts on the top side of the .01 and 0 volts at the ground side. I didnt see any variation in voltage when adjusting the reactance coil (provided I adjusted the right coil). Here is the waveform I observed at the top side of this cap. Please note that all these measurements were taken with the chassis on the workbench. So no HV or connection to the CRT at all. Here is what I found.

Forget the HV, but is the horizontal running on the workbench? It has to be running in order to get a burst gate pulse. No burst gate, nothing gets to the phase detector, so impossible to tell what's wrong or not.

Also, is the scope set for DC coupling? The variation you are looking for is slow, same rate as the color changing a few times per second, so with your high speed scope setting and DC coupling, you would expect the trace to go up and down a few times a second, in sync with the color phase changing. If the scope is set to AC coupling, you won't see such a slow variation at all.

One more thing - if the color is changing rapidly, the voltmeter may not respond in time, that's why the scope is preferred.

Oh duh...horizontal was not running. I will need to make a shorting bar for the yoke socket to get it to turn on. Back to square one. Scope was set on ac. When set on DC the trace disappears (please bear with me im new to scopes).

Got an analog VOM with a d'arsonval meter? That ought to capture the variation.
Dirty secret to taking under-chassis measurements with the chassis in the cabinet: get some clip leads connect one end of a lead to a relevant test point and run the other out from under the chassis to your meter probe and slide the chassis back in the cabinet....That method has saved my bacon on sweep and boost troubleshooting a few times.

Also table model CTC-4s have enough harness length (except maybe HV, but that can be addressed) that you can set the chassis up on the floor behind the cabinet with all the under-chassis exposed...For a console like yours if you can find a sturdy box or some such as high off the ground as the chassis shelf then you should be able to pull it up to the back of the set place the chassis on it and make everything reach/accessible.

I have a voltohmyst, it needs to be calibrated pretty bad. Havent used it in awhile.

Like a picnic cooler:

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

Phil Nelson
Phil's Old Radios
http://antiqueradio.org/index.html

And the best part is that you can keep your beer cold in the cooler while you work!

But it will be tough to get at...

I do the same thing, but I use a sturdy stool instead of a cooler. Makes everything so much easier for sure!

I replaced the CRT with a 21FB that is a bit stronger. It displays a beautiful picture with all colors strong except for red which is weak. This has do to a problem with the circuitry driving the red cathode. I have been tracing back through the circuit and have been finding resistors that have gone off. I think I will tackle this problem before I continue on the color sync issue.

Im starting to really think the replacement crystal may be bad since it does have a big dent in the side of it (Must have occurred during shipping) and I cant seem to get the color to lock no matter what I do. With the last crystal I could get the color to lock I believe. I will swap the crystal once I can find one. Work has been slow this whole year and I admit I have been lazy with the restoration of this set and have not worked on it as much as I should have. Summer is coming up soon so I hope to have the set finished by August. Slowly but surely this set is shaping up!

Sounds like you're nearing the finish line!

If you have a signal generator and scope, you can do a bench test of the crystal. In my CTC-4 article, scroll to the section titled, "Bench Testing the 3.58-MHz Crystals":

http://antiqueradio.org/RCACTC-4ColorTelevision.htm

Thanks to dtvmcdonald for that tip.

Phil Nelson
Phil's Old Radios
http://antiqueradio.org/index.html

From what I am gathering, I think the crystal may be bad. I have it setup to be bench tested. I'm not seeing a real big change in amplitude when I get around 3.58. First picture is at 3.58 and second is way down around 1mhz. This is what I believe the change should look like https://antiqueradio.org/art/RCACTC-...lAnimation.gif

Any thoughts? I believe the crystal is bad, but I'd like some advice before I go out and buy one.

They're cheap - get one or a few
http://www.newark.com/raltron/a-3-57...49u/dp/96F2798

Cant go wrong at 58 cents. Ill pick one up soon and go from there. I now have a whole lot of time to work on the set now. I hope to get it wrapped up by august. :yes:

New crystal is in and I can see a bit of improvement. Still cant get color to lock. But I can faintly see color through the rolling bars. Black and white picture is perfect I just cant get this pesky color to lock. I have tried putting caps across the crystal but I cant seem to get the frequency below 3.57900 (with a cap across the crystal). Reading through Phil's article on his 4 reveals that a 100pf cap can be put across L142 I believe. This TV wants to produce color, it is so close! Should I add a cap across L142?

Did you mistype the frequency? 3.57900 is too low - should be 3.579545.

Edit: oops - I see you meant 3.57966 - still high.

More C across L142 is easy and worth a try, I think.

Yes right, my bad. I can get it down from 3.57966 to 3.57900 with a cap across the crystal.

What size cap across the crystal? It's probably too big if it pulls it 545 Hz too low. It needs to free run close to correct so the reactance tube and coil can pull it either way.

I don't have the alignment procedure, but I imagine at some point you are instructed to tune the reactance coil for zero beat. If you can't make free-running frequency correct, the reactance circuit has little chance to pull it in.

I seem to be mixing up my results here. What I get with the crystal across the cap is 3.5760. The target is 3.57545. I am not putting the right results here. Let me start over now. The closest I can get the oscillator to run at is 3.5766 without a cap across crystal. With a cap across the crystal I pull about 3.5762 or thereabouts. the capacitor across the crystal does not pull down the frequency much at all. I would venture to guess the cap was maybe 5pf.

Are you getting tired? You seem to be missing a 9 this time. I think this time the numbers you meant to type were something like:
crystal across cap: 3.57960
target: 3.579545
without a cap: 3.57966

If that's right, the cap is pulling it 60 Hz, which is a nice amount of pull, so it seems something else is wrong. Which capacitors in the reactance tube section and the 3.58 MHz oscillator have you NOT replaced? Do you know for sure the reactance tube is good? Is the plate voltage 5 volts lower than the B+ supply as the schematic says? Time to look for anything that is not right.

By the way, too much capacitance across the crystal will lessen the effect of the reactance tube, so, since you see the frequency being pulled a good amount by the added cap, it may be better to remove it while looking for other causes.

I'm going to bed now, will check in tomorrow to read the latest. Good luck!