AC In My B-

[Jon1967us]

I have a model TV-37 I'm working on. This is my second one. My first one didn't have this same issue with what appears to be AC ripple disrupting the Horizontal and slowly makes the image do the "snake" The symptom is a Hula dancing effect with the Horiz. The pulling only occurs when a video signal is present. When there's only raster, like just at start up, the image is stable.

https://youtu.be/EPdc9n_1bpI

I have found a bit of AC current leaving the 35W4, which is the - rectifier.
On this set the B- feeds the Sync, Horiz Osc and cathodes of vert and horiz outputs.

I have found approx 2.3VAC on the plate of the 35W4 and the following AC values found on the cathodes of the following tubes.

35W4: 2.2 VAC
Hor Osc: .2V
Hor Out: .2V
Sync: .2V
HV Osc: 1.5V
Vert Osc: .28V
Vert Out: 1.3V

AC is not found on measured cathodes fed from the B+, which is the 25Z6 dual rect, even though I measured 9vAC on the cathode of that tube!

This set adheres more to the Rider version of the schematic than the Sams. Here is a link to the PDF, from Phil's site:

https://www.antiqueradio.org/art/Pil...7RidersTV2.pdf

Work I have done up to this point:

All original filter caps have been replaced with tested new caps twice, just in case there was an undetected defect that my capacitance and ESR tester didn't catch. Added an extra third 120uF parallel to the B- filter caps which lessened the pulling. Adding an extra parallel to the B+ does nothing.

All non polarized caps replaced, including all small value ceramics and coupling types.

All deflection caps replaced with WIMA HV film caps, except for the cap in the HV cage which was replaced with a 10KV ceramic. Note: this problem existed before and after the in-cage cap replacement.

All tubes have been swapped out with tested ones and this doesn't improve the symptom.

I know this is a lot of info, but I've kind of hit a wall here and am looking for ideas if anyone can enlighten me.

[Jeffhs]

Originally Posted by Jon1967us

I have a model TV-37 I'm working on. This is my second one. My first one didn't have this same issue with what appears to be AC ripple disrupting the Horizontal and slowly makes the image do the "snake" The symptom is a Hula dancing effect with the Horiz. The pulling only occurs when a video signal is present. When there's only raster, like just at start up, the image is stable.

https://youtu.be/EPdc9n_1bpI

I have found a bit of AC current leaving the 35W4, which is the B- rectifier.
On this set the B- feeds the Sync, Horiz Osc and cathodes of vert and horiz outputs.

I have found approx 2.3VAC on the plate of the 35W4 and the following AC values found on the cathodes of the following tubes.

35W4: 2.2 VAC
Hor Osc: .2V
Hor Out: .2V
Sync: .2V
HV Osc: 1.5V
Vert Osc: .28V
Vert Out: 1.3V

AC is not found on measured cathodes fed from the B+, which is the 25Z6 dual rect, even though I measured 9vAC on the cathode of that tube!

This set adheres more to the Rider version of the schematic than the Sams. Here is a link to the PDF, from Phil's site:

https://www.antiqueradio.org/art/Pil...7RidersTV2.pdf

Work I have done up to this point:

All original filter caps have been replaced with tested new caps twice, just in case there was an undetected defect that my capacitance and ESR tester didn't catch. Added an extra third 120uF parallel to the B- filter caps which lessened the pulling. Adding an extra parallel to the B+ does nothing.

All non polarized caps replaced, including all small value ceramics and coupling types.

All deflection caps replaced with WIMA HV film caps, except for the cap in the HV cage which was replaced with a 10KV ceramic. Note: this problem existed before and after the in-cage cap replacement.

All tubes have been swapped out with tested ones and this doesn't improve the symptom.

I know this is a lot of info, but I've kind of hit a wall here and am looking for ideas if anyone can enlighten me.

It seems strange you are getting ripple voltage on the B-minus side of the circuit, as B-minus is supposed to be ground. There must be some strange kind of short in that circuit which is somehow putting voltage there. I would think any voltage on the B-minus side of the circuit would immediately blow the TV's line fuse or trip the circuit breaker.

[Jon1967us]

I may be using the wrong nomenclature, what I mean is that the portion of the rectifier circuit in question is negative relative to ground.

[old_coot88]

Originally Posted by Jon1967us

I may be using the wrong nomenclature, what I mean is that the portion of the rectifier circuit in question is negative relative to ground.

Yes. It's an independent "B-minus" supply with the 35W4 working as a conventional half-wave rectifier, 'backwards' in polarity. So the filter caps are likewise in reverse polarity (positive ends to ground).

[Yamamaya42]

is it ripple from the 35W4 being loaded down too much? or perhaps a socket problem where AC from the nearby heater pins is bleeding in somehow, after all. this is in series, direct off the AC line, with no x-former, who knows what odd things it may do with socket problems.

[Jon1967us]

It's definitely the 35W4 (B-) that is the source of AC ripple. It's a single diode, whereas the 25L6 (B+) is a dual diode and does a better job in smoothing out the DC.

I'm measuring 2.4VAC at the 35W4's plate and it gets dropped to 1.6v at the junction of the 10Ω and 270Ω - only .2VAC is leaving the smoothed DC coming from the 25L6.

I'm wondering if I should try putting a 47uF or something between the plate of the 35W4 and the 10Ω?

[Yamamaya42]

Originally Posted by Jon1967us

It's definitely the 35W4 (B-) that is the source of AC ripple. It's a single diode, whereas the 25L6 (B+) is a dual diode and does a better job in smoothing out the DC.

I'm measuring 2.4VAC at the 35W4's plate and it gets dropped to 1.6v at the junction of the 10Ω and 270Ω - only .2VAC is leaving the smoothed DC coming from the 25L6.

I'm wondering if I should try putting a 47uF or something between the plate of the 35W4 and the 10Ω?

you could start, by sticking an ammeter in series with the 10 ohm resistor, and using these specs, see if the tube is being overloaded or not, https://frank.pocnet.net/sheets/093/3/35W4.pdf
if it IS being loaded beyond it's spec, you gotta track down what is loading it down, if not, the AC is bleeding in from something else, you should not have to mod anything to get it to work, within reason

[old_coot88]

Originally Posted by Jon1967us

It's definitely the 35W4 (B-) that is the source of AC ripple. It's a single diode, whereas the 25Z6 (B+) is a dual diode and does a better job in smoothing out the DC.

The 'Z6 is strapped as a simple half-wave diode like the 35W4, making it irrelevent that it's dual. The job of smoothing is done 'downstream' of the tube.

I'm measuring 2.4VAC at the 35W4's plate and it gets dropped to 1.6v at the junction of the 10Ω and 270Ω - only .2VAC is leaving the smoothed DC coming from the 25Z6.

From the voltages you cite, it's doubtful that supply ripple is causing the large bending in the raster. But if you still suspect ripple, there's an easy test for it: While observing the raster (use a mirror if necessary). Get another 47uf electrolytic and momentarily bridge and un-bridge each of the two filter caps (observing proper polarity of course). If there is little or no change in the bending, it's not a ripple problem.

I'm wondering if I should try putting a 47uF or something between the plate of the 35W4 and the 10Ω?

If you mean bridging the 10 ohm with the cap, NO. That would simply restore ripple that was being dropped across the 10 ohm.

[Yamamaya42]

It kind of makes sense that there will be SOME amount of ripple on the plate side of the 10 ohm resistor, how much is the question, and how much, if any should be seen on the far side at the filters.
Before you go tracking a possible red herring, you did say you had restored another one of these sets? Perhaps it would be productive to go and see what the negative power source on that looks, and how much ripple it has comparatively, who knows, it may be the same, or you may be on to something and it is too much in the set.

[Kevin Kuehn]

Originally Posted by Jon1967us

The pulling only occurs when a video signal is present. When there's only raster, like just at start up, the image is stable.

This sounds to me like the AC is coupling into the video signal as apposed to power supply ripple. Could be heater to cathode or grid leakage affecting the video signal.

[old_coot88]

The pulling only occurs when a video signal is present. When there's only raster, like just at start up, the image is stable.

That's a major clue right there (I shoulda caught it in your original post). It it were supply ripple, the raster edges would be hula'ing on a blank (or snowy) raster.

[Jon1967us]

Thanks for the helpful replies. As far as the video signal goes, I haven't detected any AC on the cathodes of the IF or Video tubes, nor have I detected any on the cathodes of the RF, Osc or Mixer. I haven't tested all the grids. Maybe it's possible some AC is leaking into the video grids and coming out of the anodes of the IF or video tubes.

One thing I'm thinking about is the TV37 had at least 3 production changes. This set however is labeled "TV37" not "TV37U". There was a little work done in the past on this set and one of the multi section cans was replaced. One production change, represented in Sams shows a 3 section ballast resistor, instead of discreet ones. It's possible the ballast was removed and those resistors weren't completely replaced (only the 470Ω is present in mine, like the Rider's schematic shows)

It could be that the B+/- aren't loaded down enough without the potentially "missing" (if we believe the above scenario possible) two 10Ω.

Later Sams Schematic showing low power supply and ballast resistors:



I'm not sure what my other options are, but I may need to bust open the working TV37 I have once again for a deeper look (I had done this before, scope and all and didn't find any significant difference in the spot checks I performed, but maybe I missed something)

[Jon1967us]

Originally Posted by old_coot88

That's a major clue right there (I shoulda caught it in your original post). It it were supply ripple, the raster edges would be hula'ing on a blank (or snowy) raster.

Well, at the very start the plain raster does sometimes hula, sometimes I'll start it up, turn up the brightness and the raster fills the entire screen and is stable and then the video comes in and hulas.

[Kevin Kuehn]

The video amp cathode is grounded so you wouldn't see it there. If you haven't already, try subbing new 6AU6's for both the video amp and DC restorer/sync tubes.

[Kevin Kuehn]

Originally Posted by Jon1967us

Well, at the very start the plain raster does sometimes hula, sometimes I'll start it up, turn up the brightness and the raster fills the entire screen and is stable and then the video comes in and hulas.

Oh, OK, you just moved the target on us. But if it's really intermittent as you say, I'd still be suspicious of heater leakage.