SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

the only signal that makes any sense in c_sync_2, which is correct for what the horz. sync pulse should look like, but this is entering the monitor, does it still look ok when it gets to ic010 pin 7
so you replaced 60 elect. caps on the D board?
It is hard to tell from those first 2 images, what it looks like is part of one scan line.
I would have thought a similar signal would have been in picture 3 and would have an amplitude of about 3v peak to peak, but it is dc coupled so its hard to know what the ac video signal is at that point. I also see the scope measures 12vdc on pin 37? that seems to be close to the 12v vcc. I would expect the voltage to be more like 4vdc on pin37

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

Yeah, I recapped 60-some caps. One by one, double checking each one against the part list and the PCB layout I printed out as I replaced each one. About 4.5 hours of work. My desoldering gun got a work out that day. There was no change in the issue after the recap. Looked no better or worse from what I could tell.

It was dificult due to the bundle of wires in the way and the location of IC070, but I managed to capture a screen print from pin 7.

You are right about the pin 37 on IC402 being 12v. I was using the pin number markings on the board to count the pins, and the marking for 35 must have been off slightly. I recounted manually from pin 27 and found I was off by one pin. Which happens to be the 12v rail on pin 36. I've attached a corrected pin 37 from IC402.

I also noticed that for the A board, the manual shows several wave forms from the pins on IC403 in the corner of page 30 in the PDF. I figured I would take a look since it's very easy to get to those pins on the solder side facing out. None of them seem to look even remotely like the drawing in the manual. But that could be if I didn't capture them correctly. Sorry if these aren't useful. I just figure more info the better, because these don't seem to look anything like the samples in the manual.

IC010 - pin 7 HSYNC


IC402 - pin 37 R_OUT (corrected)


IC403 - pin 1 B_IN (waveform 1, page 30)


IC403 - pin 3 B_OUT (waveform 2, page 30)


IC403 - pin 5 G_OUT (waveform 4, page 30)


IC403 - pin 7 G_IN (waveform 3, page 30)


IC403 - pin 11 R_IN (waveform 5, page 30)


IC403 - pin 13 R_OUT (waveform 6, page 30)

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

These are ideal waveforms and are also displaying a test pattern, the video from the pc (windows desktop) is very complex and you are also using the wrong sweep rate to display them.
in picture 1 the sweep rate is 5µs and displays 3 positive going horizontal pulses. The distance between one pulse and the next is one scan line. between the horzontal pulse is where one line of video is displayed.
In picture 3 you are sweeping at 500ns, you have the horzontal (blanking) pulse stretched out and only the start of the video signal and no ending blanking pulse.

Change your sweep rate to around 5ms, this will display the video at the vertical scan rate, you only need to check one color, like green.

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

That makes sense. Thanks for the explanation. I'll try and recapture the in and out of one color from IC403 tonight after work. If there is a particular test pattern that I should have displayed at the time? Maybe I could find the image and display it full screen on the laptop?

How did the capture from IC010 look? Any other places of interest to look at?

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

the horizontal sync pulse going to ic010 looks ok, I would find a test pattern that is very simple, SMPTE color bars might be ok, You could even use Paint at full screen and just draw a box that covers half the screen and fill it with black so top half is white and bottom half is black.

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

I downloaded a GIF of the color bars, and ran it full screen on the laptop while scoping IC403 on the blue in and out pins. I made sure after using the auto function to change both to 5ms as suggested, but it didn't look stable. It was rolling horizontally on the scope, and just looked like some almost solid blocks. So just to be safe, I also took prints at 5us as well. Which look more like detailed waveforms. Hopefully at least one set of these is right.

IC403 - 1 B_IN - 5us


IC403 - 3 B_OUT - 5us


IC403 - 1 B_IN - 5ms


IC403 - 3 B_OUT - 5ms

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

At 5µs you are viewing one line of video (one line from left to right) at 5ms, you are viewing on field (all lines from top to bottom)
The video input to the ic has negative going blanking pulses, so the scope can trigger on them giving a more stable image, the ic's output is inverted and now the pulses are positive going and the scope is trying to lock on the video instead (and it can't) so the result is a image that looks like it does. It should look like the input but larger in amplitute and inverted.
If the image is scrolling horzontally you need to adjust the scope's trigger level and maybe trigger polarity. (looks like there is a VIDEO Trigger type, I would use that and then depending on the video polarity, switch the video Polarity as required to lock the display
How does the image look on the screen? I do see some thin lines in the 5ms images but I don't know if they are caused by the scope itself or are there in the waveform. have a look at the video right at the CN402 (at 5ms) and see what it looks like and if those lines are there.

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

I took a snapshot of the blue pin on CN402. Despite changing the trigger to video, and trying to find the trigger, the best I could do is get it to be "twitchy". Meaning it was kind of stable, but looked like it would occasionally jerk left or right. Trying both polarity options didn't seem to make a difference.

CN402 - 6 B_IN


CRT Image

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

The actual video does seem to be ok, but the vertical sweep is being modulated by something. I think you need to be able to access the D board, Could there be a elect. cap you did not change on the D board?
Go into the monitors menu and adjust the geometry controls like pincushion, and see how it effects that ripple effect in the picture.
If you can get at the signal VSHAPE, that should be checked, IC703, pin5. It should be a nice clean ramp waveform. at around 5ms

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

It's highly unlikely I missed a cap on the D board. I only ordered what was on the part list, and had nothing left over. Plus I printed out the part list and the PCB layout and checked them as I worked. As I removed the cap, I checked it off the list. Once I put the new one in, I X'd out the location on the PCB layout. Plus there was a layer of dust on everything. I just visually inspected the board, and all the caps are all dust free now.

IC703 is a problem. It's an SMD chip located on the back side of the board. It is in an area where there are ventilation holes in the metal board holding D. So I can see it. But the holes aren't big enough to get the probe through. I hoped maybe pin 5 on IC702 would be viable, since it is a through hole part on the top side of the board. But it is mounted against a very large heat sink on the board. And has several components right in front. So I don't have a way to get to the legs.

If this is the only thing we can look at next, maybe I could use a tin snip to cut a larger opening on the metal tray under D. Unless there is some trick I don't know about to make the tip of the probe longer to get through the existing holes.

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

Why can't you just remove the metal shield, you may need to lay the board on something so it does'nt short to the chassis. I would'nt start cutting holes in the shielding

Go into the monitors menu and adjust the geometry controls like pincushion, and see how it effects that ripple effect in the picture.

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

The deflection board hangs vertically, and the wiring harnesses are not really long enough to be able to lay it down. Trying to prop it up would be virtually impossible. Not to mention the fact it integrates the tray and another big heat sink in the back of the tray into the heat sink on the top side of the board.

I'll play around with the geometry settings tomorrow morning before work and report back.

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

watch that ripple along the side of the image and see if any of the geometry adjustments change its shape, also try the horz width and see if when the size is varied the ripple wave changes shape, this might give a clue as to where the fault is.

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

I noticed when doing pincushion and horizontal size adjustments, the peaks and valleys of the ripple tended to go up and down. I noticed on the horizontal adjustment there were two larger valleys on the bottom half of the right side that never changed, even though almost every other part of the ripple did. Not sure if that points to anything. But it seemed odd given the behavior of the rest of the ripple.



Also, if IC703 is still of interest, this is where it is. I was thinking I could just make two snips as marked and bend back the shield to get the probe in. I can then bend them back, and either try soldering the gap to seal it up, or use some copper tape to bridge the cut. If it's important to keep continuity over the entire shield.

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

It does look like the problem is on the D board so you will either need to remove the shield completely to check various components or replace the board. IC703 was just a point to check the waveform, you can also check the waveform at the yoke plug but you need to make sure you check the vertical pin 2&3. But at this point I am sure the fault is with the correction signals that get applied to the vertical ramp. The problem is not with the vertical output but likely in the pincushion correction circuit.

Without vertical pincushion correction the screen would be curved in on the sides, so a correction signal is applied to correct the scan to make it straight, that correction wave is not correct.

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

I can try and see if I can make up some kind of mount to position the deflection board so I can get to the back, if needed. But it's not going to be easy. And I'm assuming straight up replacing the board is out of the question, since it's probably made out of unobtanium these days. So repair is probably my only option.

A friend of mine offered a suggestion if I need to probe IC703. I could just take a piece of solid strand wire and use it to probe through the hole in the shield, and connect the other end to my probe.

But since you said I should get similar readings off the yoke, I did that instead. That connector is very easily accessible on the deflection board. I had the color bar pattern up, used auto, and set the horizontal to 5ms, like the other images you had me capture. Hopefully these help either narrow down where the issue is, or isn't.

Otherwise what would be the next thing to try and look at?

CN508 - pin 2 - V DY C


CN508 - pin 3 V DY H

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

Any other ideas of what to look at? I know film/mylar caps are generally very reliable. But would there be any chance one of them could be starting to fail and causing this? Looks like there are around 30 or so on the deflection board.

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

It could be a weak film cap or maybe a resistor changed value, As you can see the problem is not that bad, the little bit of ripple is modulating the the vertical scan causing the problem, normally pincushion correction and keystone signals are sent to the vertical scan to change the shape of the scan, but in this case the signal is that ripple, and being very small (you can see it in the last picture) it is hard to locate. the best way is checking with a scope.
One of the reasons I asked to try different resolutions was to see how that effected the problem. As an example if Q526 was causing the problem, changing the scan rate would switch to Q525 or Q524 circuit.

[nick3092]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

Sorry if I missed the suggestion about trying different resolutions. The laptop I'm using allows me to pick 3 of them when mirroring the laptop display: 1280x800, 1024x768, and 800x600. The issue exists in all 3. I don't think it looks any better or worse in any of them.

Switching to monitor only output, I can also select 1152x864, 1280x960, and 1280x1024. Issue also occurs in all 3 of those resolutions as well. Doesn't seem any better or worse than the other 3.

Does that help narrow down where to check next?

[R_J]

Re: SGI GDM-5011P CRT Monitor (Sony Trinitron) Help

If it did'nt change anything on the screen then it did'nt point to anything specific. all the correction signals will have a nice curve or angle to them, so if one looks all ripply, thats where the problem will be.