This is an HE chassis that's exhibiting behavior I have never seen before and I'm not even sure where to go about starting to look. Video (https://photos.google.com/share/AF1QipMsqzzquJC2md1bRyx_cwJ7NJPpvnMkDHE7WJVc5iOrz3 Om6tqdIg4btvBl2pd_ng/photo/AF1QipO-LhBX3JBvM27F5JCI88JolFjVsmlGPA7eQpUc?key=TUhQTE82e nl2cnEyRDNXQ3BHSDBZMFM2UlpvbWNn)

The lower-left corner of the screen is completely stable. As you move further to the right or up toward the top, it starts getting jittery, kinda, in that it's the height and width of the picture that vary, not the position. In the upper right-hand corner it's very pronounced.

So far I've determined that B+ voltages are correct and there's no variation of them. If it were only happening on either H or V I might at least start looking at voltages in those sections but since it's affecting both, and by the same amount, I think there is a risk of going down a rabbit hole and chasing the wrong things.

Another bit of information that's probably important: This is a US-market set, NTSC, designed for 120V 60Hz. I'm feeding it a standard NTSC-M signal from a Fluke PM5515 signal generator, and have tried with the output from a HLLY transmitter set for NTSC-M and being fed an NTSC signal from a DVD player, and the behavior is identical. So I don't think this is signal-related. The set is running on 120V, 50Hz AC. I've tried both a variac and a normal autoformer to step down my domestic 230V 650Hz current.

One thought I had was that there is some interaction between the 60Hz vertical scan frequency and the 50Hz power, but the jiggle is around 2-3Hz, so it wouldn't be an artefact of the two beating together or the difference, and it's affecting the horizontal too, so 🤷. I also haven't found any evidence of other 60Hz sets having problems like this on 50Hz current, looking around the various fora.

Can anyone shed some light on what's going on here?

If your mains supply is 60 Hz, why are you trying to give the set 50 Hz? Is that all you have available in terms of your power conversion equipment?

You should use 120VAC, 60 Hz and see how the set behaves.

That was a typo, it's been corrected. I have 50Hz current, no option for anything else.

More clues please?

Does this set have a transformer? Is it sitting near a 50 Hz appliance? With a color picture, is the purity similarly disturbed? Does moving/turning the set make a difference?

Check that you're not overloading the poor thing with signal (turn down the signal gen or the AGC on the set.)

Also, being a GE, the filter on the power supply is probably marginal...

50Hz linear power supplies need larger filter and doubler caps than 60Hz supplies do.

There are options for getting 60Hz in 50Hz countries...Perhaps the most straight forward is to buy a US market automotive power inverter (12VDC to 120V 60Hz) off the internet and locally buy a beefy 12V supply....Another option if you know people that work in industrial electronics is to get an old used VFD (variable frequency motor drive) and set its output to 60Hz.
Small gasoline generators and dynamotors are also options but probably harder to obtain.

GE sets of this era used to get bad doubler caps. It usually gets extreme
but yours may be just starting. The main filter cans were not a
problem on them ( 40 yrs ago !). Since 'lytic cans rarely short try
bridging apx 80 mfd across the doubler cap & see if it gets better.
Running at 50 cps in theory requires bigger caps but there is some
room there.

BTW bridging electrolytics is an old trouble shooting technique.
It can save you time & $$$. Also give you a look at the pix to see
if you want to go further with an old set.

BTW2 If you change a doubler cap be SURE you use one rated at
the same or higher temp.

73 Zeno:smoke:
LFOD !

Thanks for the suggestions everyone, I dug back in today and here's what I've tried and the results.

Subbed filter and doubler capacitors: N.b., the B+ was fine before, in fact maybe 20V high. Swapping in the new filters and doubler made zero difference at all: B+ didn't vary more than a few mV.

old_tv_nut gave me an idea, I noticed the degauss thermistor had been replaced by some weird-looking thing I've never seen before. It measures the same cold as the original piece, and anyway I removed it from the circuit to be sure I'm not seeing magnetic effects from the degauss circuit. No difference at all.

I decided to tackle the vertical circuit next. Voltages on the tube are correct or a small amount high, output cathode is at 19.86, osc plate is at 210, output grid is at -18.2. None of the voltages vary even a tiny bit. The waveform on the grid of the VO looks exactly like the picture in the Sams, measures 211V P-P, and is completely stable, no change at all in amplitude or frequency. I tried to get a photo but at that slow sweep rate the camera wasn't cooperating.

Tell me if my thought process from here makes sense: it looks like the vertical circuit is operating as it should. What we're seeing on screen is pretty significant; if it's affecting the vertical it should be immediately evident in any of the waveforms or voltage measurements. Since this issue is affecting both the H and V equally, and in sync, and the B+ is both correct and stable, I'm thinking it could be a problem with high voltage. If the HV were fluctuating rapidly, it could cause exactly this symptom, no? The Boost voltage seems stable. The set was hissing a bit when I first brought it up but I blew out the HV cage and it stopped. I probed the HV and got 17kV. It didn't seem to be varying, but I don't know how accurate these meters are or how much variation would cause our symptom.

So that's where we are. This is my first color set. I'm suspicious of the HV but not sure where to go from here. Any other ideas?

The effect on both vertical and horizontal, plus it being stronger toward one corner, still indicates magnetic field interference to me. That's why I asked whether the set has a transformer or if there is other equipment near, or if moving and turning the set makes a difference. Is there a sound tone on your signal? If so, what happens if you use a signal with different audio?

No transformer, it's a portacolor series string set. There is a variac about a meter away, moving the set around doesn't seem to make a difference, also no effect on any other similar vintage sets on the same bench. As a test I plugged the set into an autoformer instead of the variac and got the same result.

There is a 1kHz audio signal, I have switched on and off the carrier but no change there either.

Are you sure removing the thermistor has completely disconnected the degaussing coil?

This is surely a puzzle!

Any progress on the wiggles?

Progress, no...but I have been working on it!

Thermistor completely removed, I also disconnected the degauss coil. Still doing it.

I did a test where I held up a piece of paper horizontally to the screen with a crosshatch pattern - vertically is really isn't moving hardly at all, it just looks like it is in the context of the whole picture. But with the paper there as a guide I can see it's pretty stable - it's maybe moving a little bit, but not nearly as much as horizontally. .

Horizontal, well, that's a different story. Voltages are weird on the oscillator, -4v on the grid vs -2, and plate is around 190v vs 165, though the waveforms look just like the pictures and frequencies are right. The curious thing is that if I put my scope probe on the grid of the H output, it immediately shuts down and the oscillator stops running. I'm not seeing any sign of the jitter in the waveform on the grid of the H osc, which is where my limited logic and experience tells me it would show itself. The frequency is stable and the waveform itself is completely solid.

I'm guessing (and it is a guess at this point) that the H osc shutting down when I probe it might be a clue.

ETA: High Voltage is correct and doesn't vary at all, from what I can see on the probe's tiny meter.

If your scope probe can shut down the oscillator, it might indicate that some coupling component connected to that point has gone open. Can you post a pic of the schematic in that area?

Here you go! https://i.imgur.com/nbdCpC6.png

I briefly considered C265 but given the -40V on the grid of the HO I dismissed it.

As long as my scope is in AC mode I should be able to see the same waveform on the plate of the oscillator, no?

So, let me understand - you put your scope probe on the grid of the horizontal OUTPUT, and the horizontal OSCILLATOR quits?

Yes. It's a recent vintage CRT analog Tektronix.

As long as my scope is in AC mode I should be able to see the same waveform on the plate of the oscillator, no?

I think C265 is large enough to make that correct.

I'll give it a try and see where that leads, thanks.

Does the oscillator also die when you probe the plate?

The -4 volts on the oscillator grid could be suspicious. It corrresponds to the high plate voltage. It says that the phase detector output voltage is low but the oscillator/hold coil is tuned to offset it.

Since this is closed loop operation, it is difficult to say if there's a foult or where it is.
Start by measuring the oscillator grid voltage while turning the hold control from unlock on one end to unlock on the other, and see if -2 volts grid occurs somewhere within the locked range. If the grid voltage is always low throughout the locked range, then I would check the phase detector circuit. Maybe the first thing is to replace the tube, then look for bad components.

Wild thought: is the 270V supply clean? Possible that vertical is causing a disturbance in the supply to the H oscillator?

Well, I turned the set up on its side again to take some measurements and noticed that it was back to the Horiz-Vert bouncing again. I turned it back over to try to get video, and...it was only moving vertically. Horiz was solid but vert was bouncing like crazy at the top of the screen but hardly at all at the bottom. This HAS to be some kind of intermittent bad connection, I thought, and spent the entire afternoon poking, prodding, twisting, wiggling tubes and components, tapping all over the circuit board, moving wires all over the place to see if wire dress could be the root of the problem, and...nothing. It sat there with only the vertical problem for two hours. I turned it over a dozen times. I ran it upside down. Nothing!

I got an idea that maybe this was related to convergence or magnetism or who knows what else, and grounded the grid of the VO tube to take that out of the equation. I got a straight line, no large movement, but it's still jumping a tiny bit. Feeding in a crosshatch, I was able to clearly see that the pattern of horizontal movement is not present. In the attached video you can see the line moving slightly, which I suppose is a clue. https://photos.app.goo.gl/WhGtJnhDhqmjUFdw9


So here's where I am:

I don't know why the problem has manifested in two different axes at different times. I went back this afternoon and messed with it for another half hour and it continues being only vertical. I am going to try to solve the vertical problem right now and worry about the other one if I see it again.

The fact that the vertical line exhibits a bit of movement at a similar 2-5Hz frequency, even though the vertical oscillator is disabled makes me think that...I'm not sure. That line pretty much can't move unless there is current flowing through the primary of the VO transformer.

Schematic https://i.imgur.com/MVZOy8J.png

The only things that could cause that line to move without the vertical oscillator running are:

1. Maybe the convergence coils but looking at the schematic they are in series with the yoke so, I guess not.
2. Some stray magnetic field but I'm confident there is none.
3. Maybe the vertical output tube is still conducting current even though its grid is grounded. (Is this possible? I need to measure the voltage on the cathode). IF it is still conducting a bit and we still see our problem on the screen, then we can assume that the problem lies in the circuit between the 270V B+, which I believe is clean, the VO transformer, output tube, and that bypass capacitor. It can't be much else. Please point out any holes in my logic, I'm trying to learn.

My plan of attack for tomorrow is:

1. Watch the B+ line with an analog meter and/or a scope to see how clean it really is.
2. Check the voltage on the K of the VO tube; it should be 0 if not conducting and have some voltage if it is (lin control is set to middle of range). This would be a good place to check with scope as well.
3. Determine the defective component based on the above tests. The earlier problem in the H axis makes me suspect that 270V supply but I watched it for a little while today with a DMM and it only bounced around 20-30mV.

Barring that, well, I guess ghosts.

At this point I would look for a H-K short in one of the
tubes.
Without a checker you can pull most tubes one at a time
& the raster will stay a few seconds to see if it goes away.
With a checker you may find it or you may not. You just cant
trust a checker.

73 Zeno:smoke:
LFOD !

I know on the US sets GE sometimes used the mounting lug of the terminal strips for ground connections to the chassis, but they were only screwed onto the chassis, not soldered. Loose screws or oxidized metal would cause strange symptoms.

Well, back to this...with some resolution, I guess. Today I decided to hook up my DVB tuner to my HLLY low-power transmitter and try to remove the pattern generator from the equation after another several hours of beating my head against this wall and blowing 30€ on sweep tubes. The HLLY is set for NTSC UHF channel 25, with standard 4.5MHz audio.

The DVB box has a unique feature in that you can switch its baseband output between PAL and NTSC, regardless of what the input format is. It plays MP4s from USB as well. So I set the box for NTSC, tuned in the PortaColor to UHF channel 25 and...exactly the same thing, jiggling picture. What a letdown. So it's not a problem with my signal source, this TV really is haunted.


OR IS IT?


I wanted to try changing some settings on the DVB box and the PortaColor was jiggling so much I couldn't even see the menu screen so I hauled out an old GEE (that is not a mistake (http://marcaporhombro.com/wp-content/uploads/2014/03/general-electrica-espanola.png)) B&W CCIR-B set that has a good picture on it and tuned the UHF tuner to the frequency of my transmitter. Jiggling like a MF. What? I could at least see enough to go through the menus and switch the output to PAL and the picture stabilized instantly. I didn't even have to adjust the holds.

So I hooked up my pattern generator (A Philips PM-5515TX) to the GEE set and jiggle city. I set the Philips back to PAL B, rebooted, and nice stable picture on the GEE. AARGH.

So, it is a problem with my signal source. Both of them in fact. Both of them are European products operating in a PAL world that let you set them to NTSC for one reason or another, and both of them exhibit exactly the same weird artifacts when used that way.

I'm at a loss for where to go from here. This is incredibly frustrating. Much of the frustration is of my own doing, I realize, wanting to working on these American sets instead of being content with the tons of Philips (and GEE!) that are here and were designed to work on the current and signals I have easy access to. But the heart wants what the heart wants.

I'm going to take a couple of days off and then see if I can get the PortaColor to produce a stable picture when fed a PAL signal. Color is another issue of course, but one thing at a time. I've also got a mint Predicta Holiday (http://videokarma.org/showthread.php?t=270848) I restored last summer sitting on a shelf because I thought I wired the power supply wrong and how much you wanna bet it's the same problem?

Does the Portacolor HE use an integrated circuit in the sync section? If so, this could be a problem with it not liking an NTSC signal that doesn't perfectly match all the details of FCC standard sync. This was a problem with some US sets (particularly Zenith) when video games with non-standard sync first came out.

Oh, no...it's all tube. Even the horiz AFC.

Who's got two thumbs, is in way over his head, and has a nice stable picture on his PortaColor?

��THIS GUY.��

Switched the pattern generator to PAL B and instantly got a nice stable picture on the GE. No color of course, and the V and H hold are both near the end of their range, but there you go.

Congratulations!

Time to fix the generator?

May also be time to buy some American market media players if your program sources do the same thing as your signal generator.

Yeah that's what I'm thinking. The DVB box also has an NTSC output option and does exactly the same thing, so I wonder if it's more complicated than a broken pattern generator.

I'll be back in the US in a couple of months; now just need to figure out what kind of hardware to buy.

I you don't collect vintage home video formats, a thrift store Blu-Ray player(hard to find new ones with composite video anymore), an HDMI TO AV (or HDMI to RF) and maybe a USB media player (if the Blu-Ray player doesn't have that feature) would be my picks... I'd also try to get a US market NTSC agile modulator like a blonder tongue.

Vintage game consoles, VCRs and analog video disc formats are fun if they strike your interest but not as practical if don't.

Funny how things sync up. Shango has a video today working on a Sony set that exhibits the exact same symptom (in a completely different situation). I haven't made it to the end yet but he's already indicated it's filter-related, and the way he explained it, it makes sense.