zenith porthole question
 

I picked up this h2439r with chassis 24h20 right before we moved. Removed crt and chassis on site to transport. Just got it together this weekend. Only got half a picture. Where to start, I'm assuming vertical section? Got 1 vertical half frame. Thought maybe it was yoke or crt was twisted up or down, but both the yoke and crt are in a fixed position. Vertical size adjustment does not stretch it all the way. Horz. and vertical do lock. I have not messed with any tv's for over 2 years and am rusty. Could use some help....

Has this set been restored? If not, then you should be thinking full restoration, and not of repairing the set. If you insist on operating the set as-is, keep close
tabs on the four filter cans. If any start getting hot, you had better shut it down in a hurry. Also, there is a picture control switch in the pencil box. You
need to have it in full circle mode before you can make height + linearity adjustments. Looks more like the AGC is set too high. Do you need a channel knob for it?

Looks like it may be running at twice normal vertical rate (120 Hz instead of 60).

Someone prior to me owning it recapped all the paper caps, and only 2 single ecaps. I just recapped the remaining 4 cans, checked a few mica's, replaced a out of spec bumblebee cap on the width control, and found and replaced a bunch of out of spec resistors in the horz and vertical section.
Yes I have it set to full circle. I have the knobs for it in storage, just did not find them yet.
Also all the tubes checked out ok.

I read that in a book someplace. Only problem is the book did not say how to correct it...

The image is stretched big time vertically.

I would re-check values of all replaced parts in the vert section. A single misplaced decimal point can cause all kindza woes.

I was thinking half frequency, since double would result in the top half of the raster being scanned over the bottom half (which I've seen before).

The problem with vertical running off frequency usually is a result of RC timing components in the feed back/feed forward loop of the vertical osc/multivibrator circuit having drifted off speck or being replaced by the wrong value due to human error. Check all resistors in the V osc/output stage for out of tolerance duds. Replace any original capacitors in that area, and make sure your replacements are of the correct capacitance. I don't know if Zenith portholes use vertical osc transformers or not, but if it does check the transformer too.

Riders list c37 4 section can with 2 20uf sections. Now I did not have any 20's that high of voltage, so I used 33uf....is that going to make a difference?

Also on c37 riders lists the unmarked terminal 80uf but the actual can is marked 20uf. So which to go by?

checked the vertical output, vertical osc. and agc section. Did not find anything significant... But I did find a new problem. I decided to check voltages, and I started at the power supply. There are 3 tubes in the supply, a 5y3gt, 5u4g, and a 6bf5.
So here's what I found:

5y3gt pin 2 riders says 187 volts. I got 218 volts.
5u4g pin 2 riders - 382 volts. I got 415
6bf5 pin 6 riders - 97 volts. I got 137.

My line voltage is 116 volts. Anyone have any ideas?

Are you using a modern digital volt meter to take these measurements ?

If so , your readings look fine . If you have a good ol Simpson 260 or other such vintage meter around , I'll bet you find the readings will more closely resemble spec. Remember , our modern DVMs don't load the circuit they're measuring like the classic 260 would , and thus the higher readings .

PS , what did the schematic call out for Input voltage ?

If they listed 110 and your at 116 , that could also explain slightly higher readings ...

I got a simpson, but did not use it. I used a digital. Schematic calls for 117 volts. Maybe I should check with the simpson.

Sorry to be a fuddyduddy, but the 260, with sensitivity of 20,000 ohms/volt, will impose a loading of 5 megohms when on the 250V scale (250X20,000=5 meg.).
The only time loading would be a problem with the 260 is in very high impedance circuitry.

In the old days meters typically had sensitivity of 1000 ohms/V, 5000 ohms/V. etc. Those could cause signifigant loading errors.

Ok got the simpson 260 out and checked.

5u4g pin 2 420v should be 382v
5y3 pin 2 220v should be 187
6bf5 pin 6 105v should be 97

The simpson says my line voltage is 125v

Hooked up the variac and dialed in input voltage to 105 volts. At that voltage I get correct readings, but picture is about the same. The top half of the picture just fades away to nothing.

Finally got some progress on this set. I injected a IF signal on the 4th IF tube and I got a decent test pattern. Backed up to the 3rd IF tube and got the half picture. So the problem is before the 4th IF stage. I checked the resistors in the whole IF stage and all the 470 ohms were way out of tolerance, some 200 ohms or more. Replaced them, but still no change.
I also injected a video signal at the video amp and got a decent pic.
The IF section got lots of "dogbone" capacitors. Should I be looking at those, or do I have to search before the IF stage?
Also my voltages are running 30 volts high on the IF stages.

I'd try subbing the IF tubes with known good/NOS tubes. If that fails start checking for shorted or drifted capacitors, and open or shorted transformers/chokes.

Thanks Tom for the tips. Most of my tubes are still packed away, got to dig them out this weekend.
Most of those dogbone caps are in mmf. How many zeros do you put in the front of the number to convert to mf, for example 500 mmf would be .05mf?

Six zeros.
It would be .0005 mf
(The modern designation for that particular value would be 470 pf or .00047 mf.)

Nowadays, values below .001 mf (1000 mmf) are generally expressed in pf.

Well I finally found the problem. One of the 6au6 IF tubes was bad. Put another one in and bingo :banana: Some slight adjustments and now got a decent picture. I consider it done!:yes: