vertical lines in a Toshiba TV

[vol.2]

Originally Posted by Username1

.
Hi!

Hello! Happy Thursday morning!

An Observation - At the edges of the screen on both the right & left it seems
that where the scan lines end it looks a bit brighter - Is it just the picture
artifact, or is it real?

The "bright" part is not real, but there is a real dark line in that spot on the edges of the screen on both sides. The Sharpness is creating the light artifact on the edge of those dark lines.

In your scope picture it looks like your scan is 60hz, frame rate, If you set it
to sync to horiz sync and scoped the video amp or any of the color drivers
you are looking at, I think you would see the vert lines in each line scan.
Can you count the ripple on what should be a flat line?

Not really. I tried to do that, but it looks like the ripple left over from the flyback derived power supply occurs out of frame, and I don't particularly see a distinct waveform that reflects the lines I see on the raster. If I display a dark grey (that shows the lines on the raster) and look at the cathodes (any of the three), the scope looks pretty flat (during scan time) with just some random noise. If it's in the video signal, it's very subtle because I can't see it on a scope. It may be that it's masked by my probe. Not sure.

That's why I initially thought this was not in the video. Am I correct in thinking that it has to be in the video signal if I can effect it with the Sharpness control?

Sharpness knob is a single knob attached to the NTSC IC:



And here in the schematic, it's pin 51 (in the lower lefthand corner)

You may want to find similar bright or dark lines on the tv screen across the entire
face of the screen, count them up, and multiply times 15,734 to see what the
frequency might be, and then use some math and see what inductor & cap.

So, I think I did approximately this by using the cursors on my scope to measure the ripples present on the power supply in various places, and it seems like it's somewhere between 250kHz-600kHz. The lines are more obvious on the left side of the screen, and basically disappear (on the raster) by the time they become about 350kHz.

I tried to dampen them by putting a 0.22uf ceramic cap near the 1ohm resistor that feeds the 12V supply on the video amplifier circuit. I succeeded in damping the flyback ripples from about 300mVpkpk to about 230mVpkpk, but it made zero difference to the raster that I could perceive.

pair makes a filter close to that frequency and those components may
be the trouble spot..... You know that tv has a LOT of coils in it.

This TV does NOT have a lot of coils in it, and almost all of the coils are in the RF section, which I am not at all using in any of these tests. There is a relay switch that isolates the RF side of the signal and switches the set to composite mode.

The non-RF coils are the choke in the main AC rectifier and three small transformers used in the power supplies, the audio and the deflection.

There are zero coils on the video input board and the RGB board which handles the OSD mux and color outputs to the neck board.

There is exactly one inductor on the neck board, L521, and it is between the 178VDC supply and the three collectors of the output transistors heading to the cathodes.

And that voltage source that you have growing on the neck board is another
possible noise injection point. There is probably a focus voltage showing
up on that board to do it's business, and is it possible it's drooling a
few stray electrons around....?

What does the fact that the Sharpness effects the signal say about that hypothesis? The Sharpness happens in the NTSC IC all the way back in the signal chain. Could it effect the lines on the screen if they were sneaking into the signal at the cathodes?

You may also want to do a little unorthodox troubleshooting for this problem....
Take a 1/2" size nut (steel not a squirrel kind of nut) mount it tightly on a
wooden dowel and carefully move it around near all the goodies that can
be a problem area, like the coils, horiz. everything, output, flyback, etc.
Keep a good eye on the tv screen while doing it..... You know that
white screen above with the sharpness turned up. You never know
you may find a spot that changes something.... You might also
try a small magnet on a stick...... (Weak magnet)

Sure I can try that. I have a couple of those small convergence strips with tiny weak magnets in the end. I can tape one to a wooden dowel and wave it around inside. They are already plastic covered anyway, and this is a small set so easy to do by myself.

Another thought I has was that maybe this is due to the nature of the set being a hot chassis that has an optically isolated composite video input. I had a thought that maybe this has to do with either the opto IC getting weak, or possibly a side-effect of the floating ground used for the opto-input section. Maybe this is all weird grounding issues caused by that arrangement?

You can see here that is has a separate little ground loop for the input:

[Username1]

.

Good Evening....

I would hope you could see the line pattern on the scope while syncing off the horiz.
pulse from wherever you wanted to pick it off.... And I can see your thoughts that
if you don't see it there, then you may think it's not in the video...
If that's the case, you should be able to still sync off horiz, and see the noise
on any of the power supply lines you want to look at.... Syncing off the
horiz would just make it stable.... I also agree 100% with your
estimate at the harmonic frequency..... You know what I mean - most
scopes have an external sync input, good in this case to be sure there
is a pulse large enough to trigger on.

If it don't have ext. sync, you could always put channel 2 on horiz, and
sync of channel 2, use 1 to find the noise.


As for the opto-isolator you should be able to take that out, pick another source
select and do whatever to get a blank white screen.... Or just remove it -

So if you feed this tv a signal from an rf modulator with no signal, or if you feed it a
signal from a dot generator with no or minimum dots, no lines, or color bars just
as blank as it can be, do you still see the vertical lines? same , or effected in
any way...? - I mean something blank, but with a sync pulse in it.... no color.

The one thing I disagree with it that for us to see that lines on the screen they
have to come from somewhere.... That scope picture you posted right under the
line " On the left side of L521, it looks fine, but on the other side of L521 it has
this 40Vpkpk 60Hz waveform. I guess it's the raster. It's probably normal, but
I haven't looked at that many different cathode voltages in my life. "

Where your cursor identifies a frequency of 60hz, those vertical tight lines - that
must be what is appearing on the screen - isn't it? Yah, each of those tight
lines is the individual scan lines and you can see they are of irregular height,
should they not be all equal for a white blank screen?

I guess you also have to be careful which ground you pick....? So if your scope
is completely floating.... Like battery power, what do you see if you look at
the ground.... Like hook your ground up to one ground, & scope the other,
and then switch it around....? Anything interesting?

Does it look like anyone might have removed shielding from anything?

And you see what Zeno said about that IC - it's got a few isolated circuits
within it..... Not all on a single Pringles chip.... Several GND's
Several VCC's.....


Good Night....


.

.

[vol.2]

Originally Posted by Username1

.

Good Evening....

I would hope you could see the line pattern on the scope while syncing off the horiz.
pulse from wherever you wanted to pick it off.... And I can see your thoughts that
if you don't see it there, then you may think it's not in the video...
If that's the case, you should be able to still sync off horiz, and see the noise
on any of the power supply lines you want to look at.... Syncing off the
horiz would just make it stable.... I also agree 100% with your
estimate at the harmonic frequency..... You know what I mean - most
scopes have an external sync input, good in this case to be sure there
is a pulse large enough to trigger on.

If it don't have ext. sync, you could always put channel 2 on horiz, and
sync of channel 2, use 1 to find the noise.


As for the opto-isolator you should be able to take that out, pick another source
select and do whatever to get a blank white screen.... Or just remove it -

So if you feed this tv a signal from an rf modulator with no signal, or if you feed it a
signal from a dot generator with no or minimum dots, no lines, or color bars just
as blank as it can be, do you still see the vertical lines? same , or effected in
any way...? - I mean something blank, but with a sync pulse in it.... no color.

The one thing I disagree with it that for us to see that lines on the screen they
have to come from somewhere.... That scope picture you posted right under the
line " On the left side of L521, it looks fine, but on the other side of L521 it has
this 40Vpkpk 60Hz waveform. I guess it's the raster. It's probably normal, but
I haven't looked at that many different cathode voltages in my life. "

Where your cursor identifies a frequency of 60hz, those vertical tight lines - that
must be what is appearing on the screen - isn't it? Yah, each of those tight
lines is the individual scan lines and you can see they are of irregular height,
should they not be all equal for a white blank screen?

I guess you also have to be careful which ground you pick....? So if your scope
is completely floating.... Like battery power, what do you see if you look at
the ground.... Like hook your ground up to one ground, & scope the other,
and then switch it around....? Anything interesting?

Does it look like anyone might have removed shielding from anything?

And you see what Zeno said about that IC - it's got a few isolated circuits
within it..... Not all on a single Pringles chip.... Several GND's
Several VCC's.....


Good Night....


.

.

Thanks for the thoughts here. I did once capture the collectors of the cathodes while syncing off the HOR pulse and I didn't see anything obvious there.

As an experiment, I also once put .082uf caps between the collectors of the video output and ground. I was given this experiment by someone to check for the banding in the video as the theory being that placing these caps across the video output should kill all AC components and the banding with it.

The banding in that case became lighter, but it was still there.

The conclusion of this experiment was that the banding must not be in the video, or is not fully in the video, and that it must be coupled some other way.

Currently a proposed idea is that it may be magnetic interference from the flyback itself and that I should pull out the flyback and extend the leads in order to move it around and see if that changes anything.

Do you agree that the caps across the video output collectors should have killed the banding if it was wholly present in the video?

[Username1]

.

"Do you agree that the caps across the video output collectors should have killed the
banding if it was wholly present in the video?" - No, because a cap in that spot, or
any spot will only partly reduce, or enhance any interference. You could only
completely remove it if you set a bias in the signal drivers, and removed
any possible signal from effecting the drivers, tube or transistor. Not
to ground the base, but bias it and cut the input signal So you don't
drive it into saturation, or cutoff.

I'm pretty sure the interference is from the Horiz circuit as the lines are stable
with respect to the horiz sweep. Is it magnetic? Not sure, that is why I thought
moving around a piece of steel might show something. Most likely it's
electronic, a harmonic that is not sufficiently terminated in this design,
or it is and a component is bad, a circuit trace is broken, ground broken,
something just allowing a harmonic large enough that you see it
on the screen.... Bad coil, cap, etc, around the horiz drive,
flyback, yoke, you name it, it's in that area, or something
is picking it up & sending it through the set.

The problem with the interference in the power supply is that it will be present
in all amplification circuits - even if they are biased on in steady state and
not doing anything else.... Think about a bad filter cap in an old 5 tube
radio, I'm pretty sure we have all heard it. Volume effects it a little,
if it has a tone control it will effect it a little, and being on a station
will effect it a little too. - But is it in the signal path?

I once had a white hum bar floating through a tv picture, Naturally I thought
it was a cap, so did everyone else... The set went around the shop till I got it
back and changed some tubes.... Turns out it was a heater cathode short
in one of the video tubes...


.