While you've got L42 disconnected, check socket lug#3 (cathode) to ground. It should show open circuit (infinite resistance on highest ohms scale).

The C213 that I removed was marked as 270 pf, a "postage stamp" type mica. I just checked the heater voltage; it's 6.32VAC at a line input of 117VAC. With everything disconnected from pin 3 of the oscillator tube, resistance from that pin to ground is infinite.

Since the schematics call for 270 pf, I removed the 82 pf cap (which didn't improve anything) and reinstalled a 270 pf for C213. While I was there, I also replaced R248, the 47K resistor. With everything reconnected, the TV behaves the same.

Phil Nelson

Well, to re-verify the following:
1). B+ is correct (at the jct. of R249/R250).
2). R250 is correct, 1K (brown-black-red).
3.) Plate voltage (pin 1) is considerably lower than spec, indicating the tube is overconducting from the no-bias condition.

The only remaining suspect preventing oscillation is the primary of the CW xfmr possibly having shorted turn(s). Can the xfmr be unmounted and opened to inspect the primary?

Based on similar previous threads: the crystal is always suspect unless
tried and working in an identical circuit. Known good crystals that were
not 100% identical replacements have been known to not work.

Here's another (way out:rolleyes:) idea, since Phil now has an extra 12uh coil, and assuming he has an old 10-365 pf tuning condenser. But first I'd like to run it by the collective wisdom of the group and see if it has any merit.

Cut the crystal and R248 out of circuit as shown, and replace them with a tank consisting of the 12uh coil and the tuning cap. Sweep the cap and see if it'll hit a resonant spot where the osc will start and run.
The xtal is primarily for freq stability, but the osc should free-run OK on the LC circuit (shouldn't it?):saywhat:

1. Correct.
2. Correct.
3. Not sure what qualifies as "considerably lower." I'm on the road for a few days and away from my notes, but I believe I have measured anywhere from +249V to +270V at pin 1, maybe a little more.

When I get back (Thursday), I can try some more things, including substituting different crystals and going on a treasure hunt for an old variable tuning condenser.

Phil Nelson

The circuit should not run without a crystal. It is the tank circuit.

Sorta what I figgered. Just wanted to toss the idea out there before nixing it.

Phil wrote,You indicated about 249V, which is down from the spec'd 280V, showing the tube is dropping about 30V across the 1K resistor (R250). That's .9 watt and the resistor would be running quite hot. If it is, you might want to sub a 2 watt resistor during the troubleshooting period.

Wow! You got an amazing deal on that. Welcome to the Director 21 owners club lol!

I just remeasured those voltages. Today I measure +267.5V at the junction of R249 and R250. I measure +255V at pin 1 of the oscillator tube. R250 seems to be holding up.

The Sams and RCA schematics give somewhat different voltages at various points. In Sams the voltage at junction of R249 and R250 is given as +270V. In RCA it is given at +285V. Both give a nominal line voltage of +117VAC and I set my variac at approximately that level, but it's not a laboratory grade instrument and I'm using a DMM.

The Sams and RCA schematics agree in showing a 5V drop across R250 (either 270-265 or 285-280). Today it looks like I am dropping 12.5V (267.5-255).

I remeasured voltage at pin 2. It is +107V (Sams calls for +85V and RCA calls for +110V).

While I was away, I ordered a couple of new 6AZ8 tubes and tried subbing them out of superstition. No change.

I also installed two very short leads, coming out from pin 6 of the oscillator tube and ground, that I can use to substitute crystals more easily. I swapped all four of my crystals with no change. Still measuring zero volts at pin 6.

I removed the can from T115 (the 3.58 mc CW transformer) to replace the two 1000pf caps, C219 and C220. Now that I look closely, it looks pretty toasty.

http://antiqueradio.org/art/RCACTC-4T115Exposed.jpg

The resistance on the primary should measure 6.8 ohms according to the Sams schematic. I measured it at 6.8 ohms. When the caps were off, I measured the other coils; both of them have continuity and they measure at .2 ohms. I just noticed that the Sams parts list has a note stating, "R-Y windings = .2 ohms."

I tried the TV again after those cap replacements and saw no change.

I'm not sure how much faith to put in those resistance measurements. I'd feel a lot better if these coils looked brand spanking new.

Phil Nelson

I have ordered a replacement for transformer T115. A while back (post #179), I did a test that seemed to show that T115 is at least partly functional, but I suspect that didn't tell the whole story, and those coils look they overheated in the past.

If the new transformer doesn't improve the situation, I'm not sure what else to try, except to replace the tube socket.

I wonder if there's a way to do a bench test of the old transformer after I take it out?

Phil Nelson

Does it work with the capacitor from the
previous stage and the two outputs of T115 disconnected?

I would suggest the next step after checking that is to use the new transformer plus a complete set of new parts, including tube and tube socket,
and build a breadboard of the whole thing leaving out the capacitor
that connects to the previous stage and leaving the two outputs of
T115 disconnected. Run it off any 250-300 v power supply.
There is no need for short leads in this circuit, leave them long.

You need to get that working and then transfer parts to the
main chassis. These oscillators can be picky.

Let's see if I follow what you're suggesting. Do you mean disconnecting C202 from pin 6 of the oscillator (V124A) and disconnecting the leads from terminals D and F of T115?

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

And then I put the scope on terminal D or F to look for a signal at those points?

Phil Nelson

The new part arrived (the extra can is from my TV).

http://antiqueradio.org/art/RCACTC-4...formerT115.jpg

The RCA part number is 100430 and the Meissner part number is 17-6014. I got this from Play Things of Past. I also checked with Moyer; they don't have the Meissner part but they do have a Miller 6025, which is equivalent according to their records.

I'm leaving tomorrow morning for another short trip, so I probably won't have time to play with this until next week.

Perhaps breadboarding this circuit is not a bad idea. In different schematics, the value of L42 is given as either 11 or 12 muh, and the value of C209 is given as 4, 4.7, or 6.8 pf. I believe I have all the needed parts on hand, except for a ready-made power supply. Maybe I can pull out the power/audio chassis from my Admiral 24C15 B/W set and tap that for B+ and filament voltage.

Phil Nelson

Before doing an external breadboarding, I would maybe try this first.. go ahead and mount the new can and hook up the primary (terms. B & E). Leave the secondaries completely unconnected to minimize loading.
Bring the scope probe close to either L42 or term. B to look for the osc signal.

It's amazing that part was still available.

Hmmm. I looked inside the can of the new transformer and discovered a resistor (looks like blue-gray-brown, 680 ohms) in parallel with each cap on the secondary coils:

http://antiqueradio.org/art/Meissner17-6014Internal.jpg

The data sheet also shows these resistors:

http://antiqueradio.org/art/Meissner...rSchematic.jpg

Their "typical" circuit is different than the CTC-4, with a 6AN8 quadrature amp between the reference oscillator and the transformer. It looks like the circuit from the RCA CTC-2B; this schematic specifies the resistors as 680 ohms and the caps as 1500 pf:

http://antiqueradio.org/art/RCACTC-2...fSchematic.jpg

The labeling of the secondary terminals is a little different and the resistances differ from the original. I measure 1.7 ohms across the primary (B-E) and .3 ohms across each secondary (F-D and A-C). The original measures 6.8 ohms on the primary and .2 ohms on each secondary.

The question is whether this is a good replacement for the CTC-4. I suppose I could just stick it in (noting the labeling differences on the secondaries) and try old_coot88's most recent suggestion . . . .

Phil Nelson

Ain't those caps plactic molded paper type?

They look like good old-fashioned "postage stamp" style micas to me. The little resistor is tucked under the cap.

Phil Nelson

I had the same concern as Tom when I saw the size of the cap. Are you sure it isn't a molded paper type?

http://i157.photobucket.com/albums/t...amoldCapAd.jpg

I hate these danged dot schemes, but according to this cheat sheet, the upper right (in that photo) dot would be silver if it's molded paper; a white dot indicates mica:

http://www.tpub.com/neets/book2/3g.htm

As for the rest, brown-green-red means 1500, silver means 10% tolerance, and the last brown means who knows what.

There's a little point on the green dot showing which direction to read.

The caps are stamped El-Menco on the back, which tells me nothing, since E-M made all sorts of caps, including mica and molded paper.

Phil Nelson

I could swear I've seen some domino cap charts that list white as paper...
The physical size, look and value (.0015uF) all scream paper cap to me.

I disconnected one lead of the cap for testing.

http://antiqueradio.org/art/Meissner...sconnected.jpg

It's a little hard to see in the glare, but the eye tube's shadow is fully open, showing no leakage under 450 volts, using a new 1500-pf mica as a comparator.

http://antiqueradio.org/art/Meissner...290TestsOK.jpg

I guess I will call this cap C290 after the CTC-2B schematic. Whatever it's made of, I'm calling it good enough for this application.

Now the trick will be to install it without torching everything else in the neighborhood, including parts I previously replaced.

http://antiqueradio.org/art/RCACTC-4T115Underside.jpg

Maybe I will install all new components and leads on the transformer first, slip the pre-wired assembly through the hole, and then connect up.

Phil Nelson

wow some of that lead dress looks really tight, with crossing paths.
I keep hoping to open this thread and see a big SOLVED beginning.

Definitely a mica - either silvered or mica foil. White is always Mica using the EIA code. I had to learn both the EIA and military (MIL-C-39) code.

Cheers,

Mica or Paper issue aside... Is that even the correct transformer... looks like it is configured the same as L-45, the quadrature transformer in the CT 100. Might be good to "breadboard" the circuit first to see if it will indeed oscillate without the buffer amp. :scratch2:

jr

Yes, that was my question -- should this xfmr work at all, given its component differences and the circuit differences between the earlier sets and the CTC-4? (The CT-100 has the same circuitry shown in the CTC-2B snippet I included a few posts ago.)

I sent a follow-up email to Moyer asking if they happen to have a data sheet for the Miller 6025 xfmr, which purported to be a match. They emailed me a scan of the data sheet, showing that the Miller xfmr is a match for the Meissner part -- not the CTC-4.

One way or another, I guess I'm destined to take the old transformer out. If breadboarding the new one doesn't work, then the last resort may be trying to rewind the original. But I could try the original in the breadboard circuit before attempting a rewind.

I have these parts on hand. Is this what we're talking about building?

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

I'm hoping I can tap the 6.3VAC filament supply and B+ supply from the power/audio chassis of my Admiral 24C15 B/W set.

Phil Nelson

Phil, there's something I was gonna suggest earlier to see if the osc can be coaxed to run, but thought the better of it until seeing the CTC-2 schematic. The CTC-2 has the osc plate clamped hard with a .01 cap.
Try disconnecting the CW xfmr's primary completely, and shunt directly across where the primary was. Bypass cap C211 is now clamped directly to the plate. This duplicates how the CTC-2 osc plate is wired.
Bring the scope probe next to L42 to see if the osc is running.
--
If not, maybe try a few higher values for R250 up to 39K to see if it makes any difference.
--
If still no go, maybe there's a conductive track inside the tube socket itself. Most likely lugs 3 and 6 (cathode and G1). If you disconnect those two lugs (no tube in socket), they should show infinite resistance to ground, verifying no conductive path.
Then (with lugs 3 and 6 still floating) power up and see that there's no positive voltage on them, verifying no leakage path from G2 or plate lugs.

If the socket shows good, I dunno what else would be revealed by breadboarding. But what the heck...

You mean, disconnect terminals B and E, and then connect junction of C211/R250 to pin 1, right?

Again, you mean to disconnect everything from pins 3 and 6, right?

Phil Nelson

maybe I am missing something here but is that just a amplifier that is tuned for the horz rate? maybe the issue is the tube before it? I see a cap coupling a signal to the control grid, does not look like a correction voltage but more like the signal to drive the tube.

Yes.
--
Correct.

Which print are you referring to here?

Okay, with that lash-up, I see a regular waveform at L42 (pin 3 of the oscillator). This is with no signal to the receiver and a high sweep rate.

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

If I supply a color bar signal at the video amp and slow down the sweep rate, I see a more complex waveform; the amplitude of this one goes down if I turn down the Color control.

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

Not sure if the second trace proves anything, but it was interesting to look at.

Incidentally, with nothing connected to the primary (terminals B-E), it still measures 6.8 ohms as given in the Sams manual.

Phil Nelson

Yowie zowie! :banana:What's the actual p-p amplitude of that?
Looks like the thing is actually running.
OK (assumimg you haven't unmounted the old CW xfmr yet), just take the new one and connect its primary in place of the old one, using a couple of short leads (after removing that jumper). See if the osc still runs.

Maybe this photo will give a better idea what I'm seeing. My scope has a lot of trouble locking onto the signal -- hope it's not a phantom.

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

Edit: I just connected the primary of the new transformer in place of the old one (i.e., B connected to pin 1 and E connected to junction of R250/CC211. I get the same waveform, whose amplitude collapses if I temporarily disconnect the primary.

Phil Nelson

OK. Now reconnect the old CW xfmr's primary. Does it kill the osc? If so, this proves the old xfmr is at fault.
-
But the question remains, with the new xfmr's differences, will it deliver correct phasing, amplitude etc downstream?
-
Also, is the osc running with the new L42 or the old one?

Do you have a frequency counter or digital SW receiver to check the frequency? looks like it is running close to 5MHz, if your 'scope is in the calibrated position.

jr

Let's try this again. The scope is on channel B with the vertical set at 2 volts/cm. The sweep time is set at .5 us/calibrate, and I pulled out the 5x mag knob.

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

I scrounged up my digital SW receiver and got a nice clear tone at 3586 kilohertz when I held the antenna near L42. The tone disappeared when I turned off the TV. I do have a frequency counter somewhere out in the garage, but I don't trust its accuracy more than the digital radio.

The oscillator is running with the new L42. It's running with new everything, except for the tube socket and the quadrature transformer.

If the oscillator is indeed running, that feels like progress. Time to call it a night; I can try some more things tomorrow.

Phil Nelson

I'd say the tube socket is fine, provided there's no intermittency when you wiggle the tube.
-
If you end up using the new CW xfmr, it'd be a good idea to clip out those two resistors before installing it. That way, if the resistors are actually needed, they can be placed externally.
-
If it's the primary bad in the old xfmr, it may be re-windable using roughly the same gauge wire and same number of turns.
The primary is untuned, serving only as coupler or 'tickler' to the secondaries, so its exact inductance is not critical.

When I reconnect the old quadrature transformer (still in place and connected to everything else), the 3.58-mc waveform is visible and I hear a tone WHEN there is no signal to the receiver.

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

When I turn on the color bar generator to inject a signal at the 1st video amp, the waveform is flatlined.

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

I guess the oscillator is capable of running, but something else is killing (or swamping out?) the oscillation when a video signal is present.

For what it's worth, I reviewed my notes of voltage readings in the color circuits. I found wrong voltages on the grids (pins 2 and 7) of V125, the 12BH7, R-Y/G-Y demodulator tube, measuring near zero when those grids should have around -20V (or around -15V according to Sams). At the time I attributed this to the oscillator not running, since the outputs of T115 connect to those grids, but who knows?

Another anomaly at V121B, the 6AZ8 bandpass amp, where the grid should have about -12V and I measured only about -3V or less.

I haven't forgotten the toasted appearance of the old transformer.

Phil Nelson

:saywhat:So the osc runs now with the old T115 in place but wouldn't before.:headscrat What happens if you disconnect C202 (thus severing the osc from everything 'upstream')? Does it still run? If so, is there any sign of color barberpoling when you inject signal?