Admiral 12x12

[rbc7snc]

Hi All,

I'm a fairly experienced radio restoration guy attempting his first TV restoration.

The set is an Admiral 12x12 that a friend of mine bought at a local antique store. It's the same model that he had as a kid, and supposedly produced some type of raster prior to purchase. I don't know exactly how it appeared prior to our restoration attempt as I prefer not to power up unrestored electronics from the late '40s.

We finished recapping the chassis, but are running into some oddball problems.

The audio section seems to be working, and the horizontal section seems to be working. The problem is with the vertical section.

When I first power on the set, I can display a test pattern without too much difficulty except that the pattern only takes up the top 2/3 of the screen (picture attached) Waveforms in the vertical section initially look fine. After about 2 minutes resistor R77, which is in the plate supply of the vertical output tube, overheats. When this happens, the picture scrambles and the waveforms at the grid on the vertical output tube (2nd section of a 12AU7) become very distorted.

I've checked my wiring several times, and haven't found any issues (at least not yet).

Someone suggested checking the domino mica caps, which I did (2 were probably actually wax/paper caps, and were replaced during the recap procedure). They all are within tolerance of their value and have leakage current in the microamp range at 300V.

The DC resistance on the vertical oscillator transformer is within spec, the DC resistance of the vertical output transformer is within spec, and the DC resistance of the yoke is within spec.

I'm starting to wonder if there is a problem with the vertical section of the yoke, but I'm open to other thoughts and suggestions. This set is complicated by the fact that there are damping resistors across the yoke sections, which makes using a yoke tester a little tricky.

This is my first foray into TV restoration, so please forgive me if I've missed something obvious here. I'm happy to post a copy of the schematic, but the images I have exceed the file size limit for the forum.

Any thoughts or suggestions would be greatly appreciated.

[Electronic M]

If it gradually drifts to that state after warmup, and you left original resistors in it one may be drifting high. I'd check resistors that set the grid/cathode bias on the vertical output...My guess the DC grid voltage on the vert out is becoming more positive relative to the cathode turning the tube on harder (increasing DC current and cooking that resistor) till the vert drive waveform gets distorted at the output plate.

[old_coot88]

Have you subbed the vert output tube with a known good one? Occasionally a tube may test 'good' on a tester but fail under operating load.

[jr_tech]

Is C2a (20 uf/450v) ok and installed correctly?

jr

[rbc7snc]

Hi All. I have some updates.

First, I checked the wiring yet again, and didn't find anything.

I re-checked my voltages, and the grid at pin 7 of the 12AU7 vertical output tube ends up with around +20V on it. It should be zero. The plate voltage started out high at around 420V, but as the grid voltage increased and the tube current went up, it seemed to draw down the power supply to the low 300V range. This is when the picture deteriorated and waveforms became irregular. If I wait a minute more, R77 starts smoking.

1. The grid bias is set by a single 2.2 meg resistor. I replace the resistor and got the same results.

2. I disconnected the input cap (I've never seen a new cap go bad, but thought I should rule this out). Same results.

3. I decided to make sure the yoke wasn't supplying too small of a load, so I replace the yoke in the circuit with an appropriately sized resistor. Same results. (the DC resistance on the vertical yoke is within spec)

4. I decided to make sure their wasn't something going on in the vertical output transformer (DC resistances were within spec), so I replaced it in the circuit with an appropriately sized resistor. Same results.

5. Just for kicks, I disconnected the input coupling cap and tied the control grid directly to ground. Nothing smoked, and the other stages in the vertical chain remained stable for a good while.

6. I had already tried 2 different 12AU7 tube, so I decided to sub in a third. This time it worked for a while. The plate voltage stayed high at around 410V, but the grid voltage stayed near zero. After a while it reverted back to the previous behavior. The 12AU7 also got REALLY hot.

I'm starting to wonder if this particular might be really sensitive to B+. The 20/40/80 uF electrolytics were replaced with 22/47/82 uF modern caps, and this often raises the B+ somewhat. If so, I may have to take some steps to drop the B+ back down to the usual range. That said, I'm still getting my head wrapped around boosted B+ power supplies, and I may be missing something in the theory.

Comments are greatly appreciated.

I put a copy of the schematic online at:

http://metro706.hostmetro.com/~archr.../schematic.pdf

(I maintain the website for our local antique radio club, and it dawned on me that I could simply put the file on our server)

[old_coot88]

What about the 100mf electrolytic on the cathode of the output stage? High leakage or reversed polarity of this cap could wreak havoc, 'specially with such high voltage on the plate (leakage of this cap would make the grid less negative relative to the cathode, thus raising the plate current).

[Notimetolooz]

Check the voltage on that cathode (looks like pin 8). Maybe the Linearity pot (R6) or R54 is changing. You might have cooked a couple of 12AU7s.
EDIT: Another possibility is that the tube socket has developed leakage so that it raises the grid voltage.

[old_coot88]

Originally Posted by Notimetolooz

EDIT: Another possibility is that the tube socket has developed leakage so that it raises the grid voltage.

Yeah, if you mean making the grid more positive, if there were leakage like a carbon track from pin 6 (plate) to adjacent pin 7 (grid) it sure could.

[MadMan]

Originally Posted by rbc7snc

The 20/40/80 uF electrolytics were replaced with 22/47/82 uF

I have the same set and recapped with similar modern values for everything. I never had any such problem, I doubt that's the case. More likely there's a bad resistor drifting high, or one of your solder joints didn't stick. Now you might scoff and say that you did a perfect job, which is how I felt. Imagine my shock when I found a joint that looked fine but was actually loose when wiggled. It can happen to anyone!

A problem I did have, however, was an internally-broken focus potentiometer. It didn't want to focus very well, and as it warmed up it would sporadically misbehave. And the actual pot felt like it didn't have enough travel. I cracked it open and found the wire winding burnt open and the plastic-y thing it's wound on had melted and malformed.

[rbc7snc]

MadMan-

That's some really good information.

I'll take yet another look at my solder work. I once restored a VFO for a vintage Viking transmitter where the only way I was able to fix an intermittent problem was reheating every solder joint in the box. As Murphy's law would have it, the problem was somewhere in a group of joints that were all but un-accessible.

Your pot comment may be prescient as well. It seems the problem is intermittent now, and I've run into pots that fail intermittently under load. The pot checked out on my DMM, but I've seen pots that look fine on a DMM and don't work in the circuit. A friend of mine recently worked on a radio with a bad volume control. It checked out just fine on my DMM, it was a little jumpy on a VTVM, and it failed miserably in the circuit. The vertical linearity pot is part of that circuit and very easy to change out.

I deeply appreciate all of the comments on this thread and I'll keep you guys posted with what I figure out. As I've said before, it's my first TV restoration, so I don't have the same level of experience as I do with radio work.

[Electronic M]

Some Intermittent pots have tin whiskers (which auto correct wants to change to read fun whiskey ) in them. If you don't have an equivalent you can often dismantle clean and reassemble the original.

I would second checking for a carbon track on the socket...I have seen it on other sets where it was so bad the socket started arcing and burning.

[old_coot88]

If it's a phenolic wafer socket, a carbon track could be inside and not even visible. Considering the proximity of the plate and grid pins (6 and 7 respectively), and the high level spike on the plate, it might be worth changing the socket as last resort, preferably with a ceramic type.

[rbc7snc]

Hi All,

Thanks for all of the helpful suggestions.

I replaced the vertical linearity pot, which seemed to fix the problem. The old pot tested fine on the bench, so I'm not really sure the exact issue. I wonder if it might have been shorting under load. In that case, the cathode voltage would approach ground, which would increase the tube current. As the grid and cathode approached the same voltage, the grid would also start to act like a cathode, which would raise it's voltage even further via current through the grid resistor. The end result is a run away tube. I've got a fried 12AU7 to suggest something along these lines probably happened.

The other minor detail I ran into is that I was trying to adjust the height and vertical linearity by looking at the picture. I got much better results when I looked at the vertical refresh waveform on my scope, adjusted for good linearity, then readjusted the vertical sync.

Here are a couple of screenshots of the Admiral on the bench.

http://metro706.hostmetro.com/~archr...r/working1.jpg
http://metro706.hostmetro.com/~archr...r/working2.jpg

Thanks again!

Roger