But the buss wire could have enough resistance that there could be a small voltage across it. It may not take much voltage on the ground to bother the sync. The chassis has a low enough resistance that there will be no voltage across it.

You could take your scope, connect it's ground to the chassis, set it to a very sensitive range and check all of the ground points.

I suppose I'm not very clear at times, but I have tried exactly that. With the TV powered through an isolation transformer, I would think the TV chassis would float at the scopes ground potential, anywhere it's connected to the chassis. Then I should be able to read any ground differential from that point to any other ground point on the buss. I've also tried a battery operated volt meter, just in case the scope AC ground is messing up the measurement. I really appreciate all the suggestions.

[Edit] Actually the TV has its own isolation transformer, but I still have it connected through my bench supply isolation transformer just as a safety backup.

What scale did you have your scope set to? Depending on how sensitive the circuits are, levels as low as 10 millivolts could cause trouble.

Sorry for taking so long to get back to this. To answer Tom S. , I'm using a 10X prob on my scope, so I'm reading on the 20mv/div. setting. My Fluke DVM will read down to 1mv on AC. I'm just not seeing much 60hz ripple on the ground buss, actually see more horizontal switching noise than 60hz.

I made a couple circuit changes as an experiment.

Firstly I changed the half wave voltage doubler to a full wave version, I'm not sure why that was not done in the first place, the component count stays the same. This proved interesting in that I still have about 1V of 60hz ripple ridding on the 235V B+. It seems this 60hz ripple is coming from the vertical output rather than 60hz AC line ripple.

The second thing I tried was to add 100 ohms series resistance and 220uf of decoupling to the 235V B+ on the horizontal module. That pretty much killed the bends in the picture, at least to the point that it's hardly noticeable to my eye.

I'm currently of the impression that the whole bends issue is caused because of poor power supply decoupling between the horizontal and vertical circuits. I'm a little perplexed as to how these sets originally passed the grade. The overall build quality of their sets seems above average for the period. Is it possible that modern electrolytics don't provide as much regulation as the large older units?

Maybe. All electrolytics have an equivalent series resistance (ESR) that acts like there is a resistor in series with the cap. The value of the ESR depends upon the brand and the grade of the capacitor. Electrolytics (and all caps for that matter) also have some series inductance.

In the DigiKey catalog both the Panasonic and the Nichicon brand caps come in several grades. There is the general purpose 85 degree centigrade cap, the 105 degree C cap, a long life cap, a low ESR cap and if I remember correctly a long life with low ESR cap. Each step up in grade has a longer specified lifetime.

I don't know if you want to bother, but you could order some low ESR caps.

On second thought I did originally have the set running on its original electrolytic caps. So that pretty much rules out ESR differences. Then again I can't remember now if I was aware of the bending then.

It's pretty much a mystery what could have changed since the set was new. I guess it's very likely that they relied on better vertical sync to the AC line back then. The fact that it works great with my BK 1077 without the additional power supply decoupling, that shows that its very AC line sync sensitive.

I had a very slow moving wave on a magnavox, took about 15 sec IIRC to cycle thru, the consensus was it was a beat frequency of the vert rate to the line rate. the heaters were connected to the chassis thru compression of the pcb to the chassis. I added a ground strap from the pcb ground to the chassis and all was well. I guess the slight voltage drop across the resistance of the pcb to ground was modulating the horz osc since it was sharing the same trace on the board as the ground side of the heater.

I found the problem by running the entire sets filament on pure DC (6v battery), then one section at a time returning the filaments to AC. when I got to the horz osc the problem returned.

How far back does one have to go to have a time when vertical was synced to the power line? I remember as a kid, there was a set at the barber shop that had this problem. That would have been around 1969. At that time, without a doubt, vertical was not synced to power line. The "wave" was very evident.

If this is something that changed with the introduction of color broadcasting, it already would have been unsynced when this set was manufactured.

I suspect something has indeed changed in the horizontal sync/oscillator, making it more susceptable to power line ripple than it originally was. It is not easy to track down the root cause of this kind of problem.

I'll probably keep picking away at this. I likely just put on a band-aid with the extra power supply decoupling to the horizontal tubes. It really bugs me not knowing the exact cause of the symptom.

Before color, the vertical rate was exactly 60 HZ (or as close to exactly as they could make it). When color broadcasting was introduced the vertical rate had to change slightly to allow the horizontal scan rate to change slightly. I don't remember the details, but the frequencies had to change slightly to minimize the interference between the color signal and the black & white signal. I understand that in the early days of color the broadcasts would revert to the original frequencies when not broadcasting in color. It took awhile for the stations to switch permanently to the color scan rates.

We had this set, or perhaps a slightly newer production run of it. Had the wooden cabinet, and the UHF tuner. Starting in the early 1960's, my mother opened what would now be considered "Day Care" for the kiddies. That Setchell would go on in the afternoon for kiddie shows, and keep on going until I went to bed when the test patterns would show up.

In fact, of all the sets the family eventually owned, that Setchell was probably the most reliable. We got rid of it ONLY because we went to color throughout the house. Once a year, I would clean the tuner, as we were downwind from New Jersey, with all it's manufacturing plants. I'd cut an index card to size, spray the cleaner on the edge of it, then rotate the tuner turret so the wet card would slide under the fingers. Rotate through all the channels, and it would be good for another year. Hell, in 16 years, we never even changed a single tube in it. Nor any of it's modules.

Sorry to hijack and reminisce, but this set never showed any bars or shading.

Hope you nail it.
John S.

Hello Kevin,
what cute set I think that would be a fun set to restore and watch .
I bet you will get it far as fixing the problem.
I'm like you things like that cause me to lose sleep
Rich

Hello Kevin,
what cute set I think that would be a fun set to restore and watch .
I bet you will get it far as fixing the problem.
I'm like you things like that cause me to lose sleep
Rich

I've been watching this set quite a bit. The extra power supply filtering I added to the horizontal module seems to have really cured the bends, to the point that I hate to mess with a good thing. I struggled a while with a broken IF slug, but was finally able to do a pretty decent alignment. The picture looks very sharp, but I'm now getting some picture induced buzz in the audio. My old Eico 360 FM/TV sweep with a separate marker generator makes it really tough to see the alignment marker positions, they're just sort of there. I'm currently working on finding a better post marker generator. I'm pretty slow at fixing TV's, but it's a fun and challenging learning experience. I’ve gained a great deal of respect for folks that earned their living doing this work.