nocash wrote on 2025-01-17, 09:48:

This is how workbench looks on a 1084S-P (and I assume 1084S-P1 should look similar), with, say, 8mm borders on left/right and t […]
Show full quote

This is how workbench looks on a 1084S-P (and I assume 1084S-P1 should look similar), with, say, 8mm borders on left/right and top/bottom,
https://www.amibay.com/threads/commodore-1084 … y-london.70989/

The folding issue will probably cut-off the leftmost 1mm and rightmost 12mm, but apart from that...
Can you adjust the width and height to match that? And adjuste the horizontal/vertical position to get it centered properly?

Are you sure the those horizontal borders are the same as the P1 monitor?
The issue is that I cannot tune the size any narrow then the pictures I posted. When I move the screen to the sides, it looks like the picture is coming out of the "dead zone" on the other side, so I assume this monitor can't display out of those borders, but I might be wrong.
On the vertical axis there are no issues and it seems like there are no really borders exist on this axis. The picture can be displayed until the very extreme vertical edges.

Deunan wrote on 2025-01-14, 19:56:

Well 10k is already 22 times better but if you can find a 220 ohm one it would be better yet 😀

Deunan wrote on 2025-01-14, 19:56:

So first I'd like to comment on the resistors from previous posts. Yes, if you can replace 3512 with 4.7 ohm fusible one, even at 0.5W instead of 0.33W, it would be better then your current replacement. As for the 1M resistor 3271, while it does create correct resistance when you connect 2x2M in parallel, it does nothing for the voltage rating. In fact it now offers two possible failure paths. It's far better, if you can't get a higher wattage / bigger 1M resistor, to create a replacement out of series of 2 or even 3 resistors. Then their voltage ratings add and make the whole circuit safer. The problem is you wan to have these 2/3 resistors more-or-less the same value for even voltage split. For two it would be a 510k + 470k, both selected to be a higher value, or 3x 330k, also selected. With 330k it might be easier as even 1% is already 3k so finding 3 resistors that are a bit above 330k would be easier. Still the easiest thing would be to get a higher wattage 1M in there.

Thanks for advising, I will search for those parts and order it.

Deunan wrote on 2025-01-14, 19:56:

I'm also wondering if the horizontal issue is not some sort of 50Hz vs 60Hz signal difference. Depends on the timings but the 50Hz will appear wider if it has the same amount of lines as 60Hz. One way to work around a different flyback would be to try adding a very small resistance (ohms at most) in series with the yoke horizontal coil. It wouldn't even be that difficult seeing how the compensation chokes are all over the PCB, connected with jumper wires. But before you start messing with that it would be good to rule out all other possible issues.

It's a good point to consider, I'll take it into account once I fix the issue below.

Deunan wrote on 2025-01-14, 19:56:

I recommend not changing any part values in the horizontal deflection, there's just too high risk of causing HOT/flyback damage. Not to mention I do not think changing the 3534 pot would actually affect anything. I'd like the -27V rail to be a bit lower, closer to that but that too is flyback-generated so... You could check the parts in the rectifier/filter (everything connected to flyback pin 10), especially the 2541 capacitor. ESR is not critical but it should be close to 22uF. So far the other caps tested OK, except that one in the yoke current path, so I doubt you'll find anything wrong here. Replacing 2541 with a 33uF or even 47uF might lower the voltage a bit but not by much.

So I took the 2541 cap off the board in order to replace it with a 33uF one. Then, I accidently turn on the monitor while this capacitor wasn't connected and I immediately observed a new issue.
The white background of the screen turned yellowish/brownish, and the blue colors seems very dark. When turning SK3 switch to "green" mode, the screen appears green but the top of the screen seems a lot lighter then the rest, something is off and it's hard to tell what.

Also, the color tone tuning POT 3654 doesn't respond when turning to either side. It worked before.
Replacing 2541 cap with a new 33uF one and later with the original one didn't solved the issue.

When I connect it to a CGA signal, the image seems to be better, but even there something is off.
The green mode issue is not present while monitor fed with CGA signal. Also, as I remember from previous tests, on CGA mode the 3654 POT didn't respond at the first place, so it might explain why CGA mode looks better? It might not be affected by the color tuning circuit due to differences between the Amiga and CGA RGB signals?

Anyway, I don't know how it happened. It might be the lack of 2541 cap when I accidently turned on the monitor, or maybe some trace/component/wire damaged while I took off the board.
Any direction of what to focus around this issue will be appreciated.

I'm attaching pics from Amiga and CGA PC in standard and green mode.
I took dave pics on the same scene as the previous pics I posted so you can see the difference. Honestly, I'm not sure how different it is but it seems more yellowish to me. The problem is highlighted in the Amiga pics.

x86_guy wrote on 2025-01-19, 20:17:

So I took the 2541 cap off the board in order to replace it with a 33uF one. Then, I accidently turn on the monitor while this capacitor wasn't connected and I immediately observed a new issue. (...)
Replacing 2541 cap with a new 33uF one and later with the original one didn't solved the issue.

The -27V rail also goes to the CRT - to bias the first grid. And it's used as a bias point for all three RGB amps (the driving transistors). In all these cases there is a resistor that will somewhat protect these circuits, and there is a 10nF cap on the neck PCB on this line. That being said with the main filtering cap out of the circuit there could be spikes here that killed something.

So first check if you still get -27V and if there is any excessive ripple. Check the 6674 and 6675 diodes in RGB amp for short/open. The color pot should only affect composite signal I think, not direct RGB drive, so CGA should not be affected (unless you are using the TV output on the CGA card?).

Deunan wrote on 2025-01-20, 13:59:

The -27V rail also goes to the CRT - to bias the first grid. And it's used as a bias point for all three RGB amps (the driving transistors). In all these cases there is a resistor that will somewhat protect these circuits, and there is a 10nF cap on the neck PCB on this line. That being said with the main filtering cap out of the circuit there could be spikes here that killed something.

So first check if you still get -27V and if there is any excessive ripple. Check the 6674 and 6675 diodes in RGB amp for short/open. The color pot should only affect composite signal I think, not direct RGB drive, so CGA should not be affected (unless you are using the TV output on the CGA card?).

The -27V reads -25V voltage, no noticeable ripple. Diodes 6674 and 6675 are OK.
The color pot as if I remember correctly worked with RGB signal from the Amiga, and not while viewing CGA.

I measured test points 4, 5, 6 and got the exact same waveforms for all of them on scope, which is weird because the schematic is showing something else. Maybe it's because I'm not feeding the monitor with a test signal (RGB pattern generator) as mentioned on page 20? Also, it might be an issue with my oscilloscope, I'll try to get another one for this test.
I'm attaching screenshots of the scope: Test1 & Test2 are in scale of 100us/div, I just stopped it twice, and Test3 in scale of 2ms/div just to show the whole picture as much as possible..

Do you have any idea where to start? Is it possible that 7671, 7680 and/or 7687 took the hit?

Edit:
BTW, before this issue happened I measured the components connected between to pin 10 of the flyback and 2541 cap.
Everything measured good except for one component in question - 2540 ceramic disk capacitor. It reads 233pF with ESR of 2M Ohms! Is it normal?

x86_guy wrote on 2025-01-20, 21:47:

The color pot as if I remember correctly worked with RGB signal from the Amiga, and not while viewing CGA.

The internal diagram of the 7640 video control chip (TDA3505) on the schematic is a bit confusing. Pin 11 should control the RGB switches, the chip does Y/C to RGB conversion but can accept external RGB input. It doesn't look like the color pot could affect RGB path so I guess your Amiga is actually feeding it Y/C instead? The pot is also clamped to GND by selecting green ouput with the switch.

Then there is the linear and digital RGB modes but that shouldn't matter either, all that switch does is pass the RGB directly or through 74LS-series receiver/amp to match the digital levels. There's 4 channels there not 3 because the fourth one controls the transistor that feeds +5V - this way the analog inputs can be connected to digital ones without another switch. BTW you need to switch that to digital RGB for CGA.

x86_guy wrote on 2025-01-20, 21:47:

I measured test points 4, 5, 6 and got the exact same waveforms for all of them on scope, which is weird because the schematic is showing something else. Maybe it's because I'm not feeding the monitor with a test signal (RGB pattern generator) as mentioned on page 20?

These pattern will depend on the screen contents. The imporatnt part is the voltage levels, schematic asks for 3.5V base and 2.5V swing - you have almost 4V base and more like 3V swing but I think that's fine. There isn't anything weird here.

x86_guy wrote on 2025-01-20, 21:47:

Do you have any idea where to start? Is it possible that 7671, 7680 and/or 7687 took the hit?

It is possible, but luckily these are pretty common (in Europe anyway) BC548B. All 3 should be matched - it's not required to get them super close together but at the very least the beta should be in the same range. Spec for B is 200 to 450, you can't have one being 200 and the other 450 is what I'm saying. All 3 being between +/- 50 to each other should be good enough, there are pots to set the white balance properly. If you have one of these parts testers based on ATmega you can probably desolder and check all 3 transistors. These will only tolerate about 30V between collector and other pins so a very negative spike could've damaged them. The 6k8 resistors would protect them from being outright blown but even short operation outside absolute max ratings will change their parameters permanently. And it might have happened.

x86_guy wrote on 2025-01-20, 21:47:

BTW, before this issue happened I measured the components connected between to pin 10 of the flyback and 2541 cap.
Everything measured good except for one component in question - 2540 ceramic disk capacitor. It reads 233pF with ESR of 2M Ohms! Is it normal?

It's OK-ish. You see most ESR tester can't really measure ESR as this requires knowing the capacitance, reading which is also ESR dependent. There are some ways around it but nothing is 100% perfect. So in the end what these meters usually measure is Xc+ESR, that is impendance (reactance in this case) and not pure ESR. For a capacitance that small the Xc part will be high so the "ESR" reading will also be high, even though it's not. And we could always argue what is ESR anyway. Point is if the cap measures close to what it should capacitance-wise I'd not worry about its ESR at all. It's probably a ceramic disc type? These either go short (which this one is not) or open, or partially destroyed due to a short self-clearing when part of the cap is blown away. Which is both obvious just by looking at it and with capacitance measurement. So this one is good. The 5541 choke could be partially shorted but that should show up as ripple at the horizontal frequency so also not the case.

If the colors are not quite right and you are sure the switches are in the right position for that input source then I'd say one or more of the BC548 took a hit. Check all the 6k8 resistors near these transistors, it should be mostly possible in-circuit. In fact you are mostly interested in seeing all of these agree across 3 color channels. Desolder only if you see a resistance that is considerably off, to make sure.

x86_guy wrote on 2025-01-19, 20:17:

Are you sure the those horizontal borders are the same as the P1 monitor?

Yes, I am sure of that.
If you are unsure how the picture should look like, please enter something like Commodore 1084S-P1 Amiga in your farorite search engine, and then select image-search.

x86_guy wrote on 2025-01-19, 20:17:

The issue is that I cannot tune the size any narrow then the pictures I posted.

Sorry, don't understand, what do you mean by any narrow, and how can your size differ from the picture you posted?

x86_guy wrote on 2025-01-19, 20:17:

When I move the screen to the sides, it looks like the picture is coming out of the "dead zone" on the other side

Don't understand that, too. It could mean you have black pixels on tne right side, and then followed by non-black pixels further right?

x86_guy wrote on 2025-01-19, 20:17:

On the vertical axis there are no issues and it seems like there are no really borders exist on this axis. The picture can be displayed until the very extreme vertical edges.

That sounds fine, but it isn't how it looks on you latest pictures, which do still have those weird black scanlines at the top and bottom.

When discussing the horizontal and vertical size, the term "picture" is terribly inaccurate, it could refer to the bitmap area, perhaps with or without the screen borders, or with only the visible part of those borders.
Generally, the horizontal (and same for vertical) signal should look as shown below. That is, in workbench you should have a blue border (which should extend beyond the visible portion of the screen).

1
2<--sync--><--black--><--border--><---------bitmap area---------><--border--><--black-->
3                           <-----------visible screen area----------->
4

x86_guy wrote on 2025-01-19, 20:17:

Deunan wrote on 2025-01-14, 19:56:

I'm also wondering if the horizontal issue is not some sort of 50Hz vs 60Hz signal difference. Depends on the timings but the 50Hz will appear wider if it has the same amount of lines as 60Hz.

It's a good point to consider, I'll take it into account once I fix the issue below.

I am not sure what Deunan means there... maybe that the monitor might misdetect the framerate?
Anyways, it would be good time to let us know what region you are living in, do you have a PAL or NTSC amiga, did you import the computer or monitor from a different region?

Deunan wrote on 2025-01-20, 23:33:

The internal diagram of the 7640 video control chip (TDA3505) on the schematic is a bit confusing. Pin 11 should control the RGB switches, the chip does Y/C to RGB conversion but can accept external RGB input. It doesn't look like the color pot could affect RGB path so I guess your Amiga is actually feeding it Y/C instead? The pot is also clamped to GND by selecting green ouput with the switch.

Then there is the linear and digital RGB modes but that shouldn't matter either, all that switch does is pass the RGB directly or through 74LS-series receiver/amp to match the digital levels. There's 4 channels there not 3 because the fourth one controls the transistor that feeds +5V - this way the analog inputs can be connected to digital ones without another switch. BTW you need to switch that to digital RGB for CGA.

So I may be wrong.

Deunan wrote on 2025-01-20, 23:33:

These pattern will depend on the screen contents. The imporatnt part is the voltage levels, schematic asks for 3.5V base and 2.5V swing - you have almost 4V base and more like 3V swing but I think that's fine. There isn't anything weird here.

Got it.

Deunan wrote on 2025-01-20, 23:33:

It is possible, but luckily these are pretty common (in Europe anyway) BC548B. All 3 should be matched - it's not required to get them super close together but at the very least the beta should be in the same range. Spec for B is 200 to 450, you can't have one being 200 and the other 450 is what I'm saying. All 3 being between +/- 50 to each other should be good enough, there are pots to set the white balance properly. If you have one of these parts testers based on ATmega you can probably desolder and check all 3 transistors. These will only tolerate about 30V between collector and other pins so a very negative spike could've damaged them. The 6k8 resistors would protect them from being outright blown but even short operation outside absolute max ratings will change their parameters permanently. And it might have happened.

I'm trying to get a hold on a BJT tester, will test it once I get it.

Deunan wrote on 2025-01-20, 23:33:

It's OK-ish. You see most ESR tester can't really measure ESR as this requires knowing the capacitance, reading which is also ESR dependent. There are some ways around it but nothing is 100% perfect. So in the end what these meters usually measure is Xc+ESR, that is impendance (reactance in this case) and not pure ESR. For a capacitance that small the Xc part will be high so the "ESR" reading will also be high, even though it's not. And we could always argue what is ESR anyway. Point is if the cap measures close to what it should capacitance-wise I'd not worry about its ESR at all. It's probably a ceramic disc type? These either go short (which this one is not) or open, or partially destroyed due to a short self-clearing when part of the cap is blown away. Which is both obvious just by looking at it and with capacitance measurement. So this one is good. The 5541 choke could be partially shorted but that should show up as ripple at the horizontal frequency so also not the case.

Got it, thanks for the explanation.

Deunan wrote on 2025-01-20, 23:33:

If the colors are not quite right and you are sure the switches are in the right position for that input source then I'd say one or more of the BC548 took a hit. Check all the 6k8 resistors near these transistors, it should be mostly possible in-circuit. In fact you are mostly interested in seeing all of these agree across 3 color channels. De-solder only if you see a resistance that is considerably off, to make sure.

The problem is the colors are horrible when the Amiga is connected to this screen via RGB interface. The colors are way off, the blue color is almost invisible, white seems like brown and the top of the screen is lighter then the rest of it, in normal and green mode.
In CGA the image is looking way more better somehow.
I'll check the parts you mentioned.

nocash wrote on 2025-01-21, 06:16:

x86_guy wrote on 2025-01-19, 22:17: […]
Show full quote

x86_guy wrote on 2025-01-19, 22:17:

Are you sure the those horizontal borders are the same as the P1 monitor?

Yes, I am sure of that.
If you are unsure how the picture should look like, please enter something like Commodore 1084S-P1 Amiga in your farorite search engine, and then select image-search.

x86_guy wrote on 2025-01-19, 22:17:

The issue is that I cannot tune the size any narrow then the pictures I posted.

Sorry, don't understand, what do you mean by any narrow, and how can your size differ from the picture you posted?

x86_guy wrote on 2025-01-19, 22:17:

When I move the screen to the sides, it looks like the picture is coming out of the "dead zone" on the other side

Don't understand that, too. It could mean you have black pixels on tne right side, and then followed by non-black pixels further right?

Take a look at this page:
https://www.reddit.com/r/crt/comments/1cyobzx … ht_green_blues/
In the second picture you can see that the image is pretty wide and folding on the edges, same as my monitor. On the third picture, the image is narrower and you can clearly see that that folding issue doesn't appear. On my monitor, the issue is that I can't get the image any narrower so the image stays outside the viewable area of the screen on the left and right edges. Anyway, that's what I was thinking and what I was understanding from the feedbacks on this thread, am I wrong?
If you think that the issue is not what I mentioned above, please share your thoughts.

nocash wrote on 2025-01-21, 06:16:

When discussing the horizontal and vertical size, the term "picture" is terribly inaccurate, it could refer to the bitmap area, […]
Show full quote

When discussing the horizontal and vertical size, the term "picture" is terribly inaccurate, it could refer to the bitmap area, perhaps with or without the screen borders, or with only the visible part of those borders.
Generally, the horizontal (and same for vertical) signal should look as shown below. That is, in workbench you should have a blue border (which should extend beyond the visible portion of the screen).

<--sync--><--black--><--border--><---------bitmap area---------><--border--><--black-->
<-----------visible screen area----------->

When saying picture I actually mean the displayed image on the screen. If it's viewable on all the screen and there are no borders that restrict it horizontally, then it's not clear what is the issue I'm facing with this monitor.

nocash wrote on 2025-01-21, 06:16:

x86_guy wrote on 2025-01-19, 22:17: […]
Show full quote

x86_guy wrote on 2025-01-19, 22:17:

Deunan wrote on 2025-01-14, 21:56:

I'm also wondering if the horizontal issue is not some sort of 50Hz vs 60Hz signal difference. Depends on the timings but the 50Hz will appear wider if it has the same amount of lines as 60Hz.

It's a good point to consider, I'll take it into account once I fix the issue below.

I am not sure what Deunan means there... maybe that the monitor might misdetect the framerate?
Anyways, it would be good time to let us know what region you are living in, do you have a PAL or NTSC amiga, did you import the computer or monitor from a different region?

I'm not sure if it's PAL or NTSC, I'll try to check it. The Amiga PC and monitor are imported each from another country, I'll check where it came from and will let you know later. Is it something that can lead to the folding issues I mentioned?

x86_guy wrote on 2025-01-21, 17:11:

The problem is the colors are horrible when the Amiga is connected to this screen via RGB interface. The colors are way off, the blue color is almost invisible, white seems like brown and the top of the screen is lighter then the rest of it, in normal and green mode.
In CGA the image is looking way more better somehow.

Are you still using the cable from this post: Re: Commodore 1084S-P1 horizontal fading issue
Or is it something different? Because that is RGB only cable so it should be bypassing the Y/C chip and the color issues you have, vs CGA, should simply not be there. Unless the Amiga output is now broken somehow?

Also I just remembered that these Commodore monitors had issues with switches. This was mostly the power switch but I suppose the TTL/analog switch could also be faulty. I can't remember the details but the problem is mechanical, the switch stops operating properly, and/or makes very poor connection. That could just explain the issue, if some of the signals are just not passing through properly. TTL would be less affected as the signal is being sanitized by the 74LS37 AND array.

Test the SK4 RGB switch, each contact should be less than 5 ohms (in fact less than half ohm would be best, but we'll take 5 too). And it should not radomly spike up when switch is operated or react to vibrations. If in doubt desolder the switch and install 4 jumper wires to set the monitor to analog RGB and try again.

nocash wrote on 2025-01-21, 06:16:

I am not sure what Deunan means there... maybe that the monitor might misdetect the framerate?
Anyways, it would be good time to let us know what region you are living in, do you have a PAL or NTSC amiga, did you import the computer or monitor from a different region?

I'm not sure if it's PAL or NTSC, is there any way to check it?
The monitor came from Germany and the Amiga PC from the UK.
I'm running both on 220V 50Hz network without any power converter.

Deunan wrote on 2025-01-21, 19:31:

Are you still using the cable from this post: Re: Commodore 1084S-P1 horizontal fading issue
Or is it something different? Because that is RGB only cable so it should be bypassing the Y/C chip and the color issues you have, vs CGA, should simply not be there. Unless the Amiga output is now broken somehow?

Still using the exact same cable with no changes. I also tested for continuity between the relevant pins, and opened both connectors to make sure everything is intact.

Deunan wrote on 2025-01-21, 19:31:

Also I just remembered that these Commodore monitors had issues with switches. This was mostly the power switch but I suppose the TTL/analog switch could also be faulty. I can't remember the details but the problem is mechanical, the switch stops operating properly, and/or makes very poor connection. That could just explain the issue, if some of the signals are just not passing through properly. TTL would be less affected as the signal is being sanitized by the 74LS37 AND array.

Test the SK4 RGB switch, each contact should be less than 5 ohms (in fact less than half ohm would be best, but we'll take 5 too). And it should not radomly spike up when switch is operated or react to vibrations. If in doubt desolder the switch and install 4 jumper wires to set the monitor to analog RGB and try again.

I replaced all the switches with new ones: SK2 (CVBS/RGB), SK3 (GREEN mode), SK4 (Analog/TTL), SK5 (CVBS/LCA) - All of them tested and functioning well. I just realized that SK5 doesn't exist on the schematics.
The main power button bypassed.

The thing that concerns me the most is the green mode when viewing the Amiga RGB signals. The brightness is fading from top to bottom. Is it also something that can come from the RGB section?

I'm about to test those BJT RGB power transistors, will post results later.

x86_guy wrote on 2025-01-22, 19:42:

I'm not sure if it's PAL or NTSC, is there any way to check it?
The monitor came from Germany and the Amiga PC from the UK.

Europe and UK were using PAL. The monitor itself will probably accept only PAL via composite input but in RGB mode it should work with both 50 and 60 Hz inputs. I'm not all that familar with Amigas though, I don't know how they output RGB. Quick online search says there is a switch? Not sure if that's the case for all Amiga models. If there is a switch then try both modes.

All this assumes the computer and monitor are European versions and not imports from USA or Japan for example.

x86_guy wrote on 2025-01-22, 19:42:

The thing that concerns me the most is the green mode when viewing the Amiga RGB signals. The brightness is fading from top to bottom. Is it also something that can come from the RGB section?

Not impossible but unlikely. Usually things like that are indication that there is some power supply issue, wrong voltage or dried out smoothing caps. But you replaced it. Just in case, inspect the -27V line, all the way from the cap, through the 4-pin connector that goes to the neck PCB, to the 6k8 resistors. These single sided PCBs can develop cracks in the copper due to soldering and flexing, so depending on where you measured the voltage it might have shown as good but is not actually reaching where it should.

If it seems OK then go ahead and try replacing the BC548 transistors. Even if you don't have a matched set, just to see if it changes the behaviour (note, replacing any of these 3 will change the white balance, that is normal). You can start with just the green one and use the monitor set to green only mode. If that doesn't fix the issue then perhaps the current source (7695+7696) is somehow faulty - but I really don't see why that would happen.

In RGB mode, especially in analog but digital is analog with just 2 possible states, there really isn't much to it. The signal is amplified by the IC (which can also do blanking on retrace), then the transistors - the pre-drivers and the final cathode drivers on the neck PCB. Now the cathode drivers are pretty robust transistors, not easily killed or damaged. And these are not amplifiers, just followers with sufficiently high CE breakdown voltage to drive the tube. Cathodes need about 100V or so on them make the guns go fully dark, and this monitor uses main B+ (128V rail) for that directly. Unless this has somehow gotten very noisy all of a sudden there is nothing on the neck PCB that would be causing color issues. So that leaves the pre-amps and the IC. And their voltage rails, esp. the -27V since that is where is all started.

EDIT: But before all that, as I've said the RGB input in digital mode is just sanitizing the input levels and then it's "analog" with two possible voltage states. If all you have is CGA card then try displaying uniform green or white screen, both will show the same in green-only mode. If you get uniform shade top to bottom, but not from Amiga in analog mode then the only culprits can be the cable or the Amiga itself. Somehow. Maybe it got capacitor issues of its own.

Deunan wrote on 2025-01-23, 00:57:

Europe and UK were using PAL. The monitor itself will probably accept only PAL via composite input but in RGB mode it should work with both 50 and 60 Hz inputs. I'm not all that familar with Amigas though, I don't know how they output RGB. Quick online search says there is a switch? Not sure if that's the case for all Amiga models. If there is a switch then try both modes.

All this assumes the computer and monitor are European versions and not imports from USA or Japan for example.

I can't find such switch on my Amiga. It might be a good idea to try the Amiga with another monitor, but currently I don't have it.

Deunan wrote on 2025-01-23, 00:57:

Not impossible but unlikely. Usually things like that are indication that there is some power supply issue, wrong voltage or dried out smoothing caps. But you replaced it. Just in case, inspect the -27V line, all the way from the cap, through the 4-pin connector that goes to the neck PCB, to the 6k8 resistors. These single sided PCBs can develop cracks in the copper due to soldering and flexing, so depending on where you measured the voltage it might have shown as good but is not actually reaching where it should.

I measured the -27V voltage on all 6k8 resistors and on the CRT neck, it's fine.

Deunan wrote on 2025-01-23, 00:57:

If it seems OK then go ahead and try replacing the BC548 transistors. Even if you don't have a matched set, just to see if it changes the behaviour (note, replacing any of these 3 will change the white balance, that is normal). You can start with just the green one and use the monitor set to green only mode. If that doesn't fix the issue then perhaps the current source (7695+7696) is somehow faulty - but I really don't see why that would happen.

All of those transistors tested with a multimeter with built in npn/pnp transistor tester. Below the results:
7671 BJT NPN - h21E ~ 480
7680 BJT NPN - h21E ~ 550
7687 BJT NPN - h21E ~ 510
7695 BJT PNP - h21E ~ 380
7696 BJT NPN - h21E ~ 620

Comparing it to datasheet, it looks fine.

Deunan wrote on 2025-01-23, 00:57:

In RGB mode, especially in analog but digital is analog with just 2 possible states, there really isn't much to it. The signal is amplified by the IC (which can also do blanking on retrace), then the transistors - the pre-drivers and the final cathode drivers on the neck PCB. Now the cathode drivers are pretty robust transistors, not easily killed or damaged. And these are not amplifiers, just followers with sufficiently high CE breakdown voltage to drive the tube. Cathodes need about 100V or so on them make the guns go fully dark, and this monitor uses main B+ (128V rail) for that directly. Unless this has somehow gotten very noisy all of a sudden there is nothing on the neck PCB that would be causing color issues. So that leaves the pre-amps and the IC. And their voltage rails, esp. the -27V since that is where is all started.

EDIT: But before all that, as I've said the RGB input in digital mode is just sanitizing the input levels and then it's "analog" with two possible voltage states. If all you have is CGA card then try displaying uniform green or white screen, both will show the same in green-only mode. If you get uniform shade top to bottom, but not from Amiga in analog mode then the only culprits can be the cable or the Amiga itself. Somehow. Maybe it got capacitor issues of its own.

128V rail also tested and looks fine. Maybe there is an issue with the Amiga output, I don't know and I'll need to check it.
But I see that the colors are off with CGA signal as I told earlier, I can't really point what is the issue because in fact all the RGB colors appears on the screen, but somewhat dull.
So I connected my Sega Master System to this monitor using it's analog composite interface and noticed something interesting.
On the main menu of a some cartridge, there are two grey lines on top and bottom that doesn't appear on the screen. I got pictures of the same menu that I took before the issue started, and you can clearly see that it was there before.
Also, just to make sure that nothing messed up with the Sega, I connected it to a different TV, and those lines exist there.
Is it making the issue more clear now? It might lead us to the problem?

I'm attaching pics of this menu displayed on this monitor before and after the colors issue, and on a TV.

Interesting. Did you try adjusting the contrast pot? On digital input the contrast control can kill the darker shades completly like that, while still showing the bright ones (although darker). It's also affected by the brightness control to some extent - what each pot does, excatly, depends on the monitor. On some it affects the cathode drive, on others it's the the other grids after the guns that have their voltages changed. This particular monitor seems to control everything on cathodes side, through that video IC.

Deunan wrote on 2025-01-24, 14:35:

Interesting. Did you try adjusting the contrast pot? On digital input the contrast control can kill the darker shades completly like that, while still showing the bright ones (although darker). It's also affected by the brightness control to some extent - what each pot does, excatly, depends on the monitor. On some it affects the cathode drive, on others it's the the other grids after the guns that have their voltages changed. This particular monitor seems to control everything on cathodes side, through that video IC.

Contrast and brightness are both tuned to maximum.
Do you have any idea which area on the circuit should I focus? Any possibility the the CRT damaged?

Edit:
Inspired by the info you mentioned above, I tried to tune the "screen" pot on the flyback and it fixed the problem for all inputs!!
The dull colors turned bright and clear , the brownish whites turned into definite bright white.
On green mode there's still slight fading in brightness from top to bottom but it's really insignificant.

So we are returning to the old good problem, screen is to wide. I'll try to put a series resistor to the horizontal deflection coil and see if it helps.

This is max? Seems a bit darker then it was, though it might just be the shutter effect and/or ambient light. Odd. Usually with both pots maxed you should get the screen so bright you'd see retrace lines, and the black level is too bright.

CRTs are not that easily damaged and there is RC filter on the neck PCB. I doubt that. Okay lets establish some baseline here. Do you have any programs for CGA that display all 16 colors on the screen? I'd like to confirm that is working properly and the intensity bit is also working. If you don't have such software I can write some BASIC code that you can run through QBASIC in DOS for this purpose.

Just in case also measure the 3138 (and 3139 while at it) on the neck PCB, both should be 1k5. You can also measure the 2138 and 2139 capacitors. Nothing needs to be desoldered or the PCB removed from the neck - just unplug the M1 connector from the main PCB.

One last thing, and I should've perhaps mentioned this earlier - the capacitor you've changed on -27V line (2541), it's position is very close to the flyback pots that control HV and G2. G2 is basically so-called sub-brightness control on color CRTs. It establishes the base brightness that can be fine tuned with the user accessible pot on the front/side of the monitor. On some flybacks/monitors the G2 can be a very touchy control. And since this is a replacement it might not be fixed in place by paint/glue. If you bumped it while (de)soldering the cap it might just result in your CRT looking like it's weaker and darker now.
If that is the case there are procedures to set the G2 properly - it's in the manual, page 396 point 1.4. It's a bit involved and frankly should be done after a flyback change but if you need a quick fix here's how I do it:
- contrast at about 25%-50% (the stronger the tube the lower you set it)
- brightness in the middle
- display a pattern of shades of gray, or 3 RGB colors in separate bars, preferably with analog input of 32 steps or more
- tweak G2 pot to get perfect black level (no light from black surfaces) and the pattern bars starting to be visible at step 3-5 of 32. The lower the better as long as black is still black. This needs to be done with as little ambient light as possible, just don't get yourself shocked in the darkness.
- make sure the brightness control can turn the screen pretty much off on one end, and bloom it on the other. This is so you can adjust it to normal ambient light. Re-adjust G2 if the pot doesn't have enough control.

On a weak CRT you might not get patterns to start being visible at such low values. In that case raise the contrast more and redo, you might even have to accept the black blooming a little.

The problem with this is you need a pattern generator because this is not a VGA monitor can be be easily driven from a PC to display all the required colors and shade levels. So you might have to do it roughly with CGA output, the dark grey (color 8 ) and black should be your reference points. I would advise against touching any of the color / white balance pots on the RGB path (unless you intend to replace the transistors). I mean it's not black magic but it can be rather annoying to get it in order, especially without proper signal source.

Deunan wrote on 2025-01-24, 16:38:

This is max? Seems a bit darker then it was, though it might just be the shutter effect and/or ambient light. Odd. Usually with […]
Show full quote

This is max? Seems a bit darker then it was, though it might just be the shutter effect and/or ambient light. Odd. Usually with both pots maxed you should get the screen so bright you'd see retrace lines, and the black level is too bright.

CRTs are not that easily damaged and there is RC filter on the neck PCB. I doubt that. Okay lets establish some baseline here. Do you have any programs for CGA that display all 16 colors on the screen? I'd like to confirm that is working properly and the intensity bit is also working. If you don't have such software I can write some BASIC code that you can run through QBASIC in DOS for this purpose.

Just in case also measure the 3138 (and 3139 while at it) on the neck PCB, both should be 1k5. You can also measure the 2138 and 2139 capacitors. Nothing needs to be desoldered or the PCB removed from the neck - just unplug the M1 connector from the main PCB.

One last thing, and I should've perhaps mentioned this earlier - the capacitor you've changed on -27V line (2541), it's position is very close to the flyback pots that control HV and G2. G2 is basically so-called sub-brightness control on color CRTs. It establishes the base brightness that can be fine tuned with the user accessible pot on the front/side of the monitor. On some flybacks/monitors the G2 can be a very touchy control. And since this is a replacement it might not be fixed in place by paint/glue. If you bumped it while (de)soldering the cap it might just result in your CRT looking like it's weaker and darker now.
If that is the case there are procedures to set the G2 properly - it's in the manual, page 396 point 1.4. It's a bit involved and frankly should be done after a flyback change but if you need a quick fix here's how I do it:
- contrast at about 25%-50% (the stronger the tube the lower you set it)
- brightness in the middle
- display a pattern of shades of gray, or 3 RGB colors in separate bars, preferably with analog input of 32 steps or more
- tweak G2 pot to get perfect black level (no light from black surfaces) and the pattern bars starting to be visible at step 3-5 of 32. The lower the better as long as black is still black. This needs to be done with as little ambient light as possible, just don't get yourself shocked in the darkness.
- make sure the brightness control can turn the screen pretty much off on one end, and bloom it on the other. This is so you can adjust it to normal ambient light. Re-adjust G2 if the pot doesn't have enough control.

On a weak CRT you might not get patterns to start being visible at such low values. In that case raise the contrast more and redo, you might even have to accept the black blooming a little.

The problem with this is you need a pattern generator because this is not a VGA monitor can be be easily driven from a PC to display all the required colors and shade levels. So you might have to do it roughly with CGA output, the dark grey (color 8 ) and black should be your reference points. I would advise against touching any of the color / white balance pots on the RGB path (unless you intend to replace the transistors). I mean it's not black magic but it can be rather annoying to get it in order, especially without proper signal source.

I edited the post probably few moments before you answered. The problem was exactly what you described, I had to adjust the screen POT and issue solved.
Anyway, I'll tune the brightness POT/scree POT per your instructions because it sounds much more professional.
It will be nice to have this RGB program you've mentioned, if it's not too much work for you. Thanks for that.

BTW, I'm starting to doubt that the screen folding on the right and left sides is because the picture is too wide, because it's actually fits the whole screen with a bit of clearance between the visible image and plastic borders, and when I tune it wider, it actually gets wider. So this might tell us that the image is probably need to be displayed the whole with no "imaginary" borders at the sides. So it leads me to think, maybe the signal gets too wide to the deflection section and therefore cuts at the sides?

Correct procedures for setting up the sub-brightness, and white balance, are often written for new or strong CRTs. I found them hard to follow (assuming I have most of the required equipment) if the tube is tired, usually you have to accept some things will not be perfect.

As for the horizontal size, I was under the impression this is as narrow as it can be and still spills out on both sides when centered. If that is not the case, as it now seems if you can get some clearance on both sides (even if very small), then yes the issue is not flyback or controls. Well it might be a bit different (original flyback could, perhaps, drive the picture even more narrow) but certainly acceptable if that is the worst case scenario. So the folds on the sides could be because the signal source is not blanking properly. Again this is me not knowing much about Amigas, some systems have a separate blanking signal and the monitor should be taking care of it. I don't think the Amiga does it this way but perhaps the way it implements blanking is causing the folds to be (partially) visible? You'd need to ask someone with similar setup (signal source and monitor type).

Note the folds will only be visible if the so-called border color is not black/off, and the screen edges are still visible. Perhaps the easiest solution to this problem is just widen the picture a bit so that the sides do move beyond CRT face. Maybe Amiga does it like this? Note there will not be any content here, except solid color, so you are not loosing any details. This is just the border color, meant to fill the screen sides. Just don't widen the picture too much or you will see the folds again, rolled over the actual picture - but that only happens if the picture is really way to wide vs horizontal sweep time, usually when the monitor can't sync properly to faster timings.

Deunan wrote on 2025-01-29, 17:21:

Correct procedures for setting up the sub-brightness, and white balance, are often written for new or strong CRTs. I found them hard to follow (assuming I have most of the required equipment) if the tube is tired, usually you have to accept some things will not be perfect.

The brightness is just fine so no worries here.

Deunan wrote on 2025-01-29, 17:21:

As for the horizontal size, I was under the impression this is as narrow as it can be and still spills out on both sides when centered. If that is not the case, as it now seems if you can get some clearance on both sides (even if very small), then yes the issue is not flyback or controls. Well it might be a bit different (original flyback could, perhaps, drive the picture even more narrow) but certainly acceptable if that is the worst case scenario. So the folds on the sides could be because the signal source is not blanking properly. Again this is me not knowing much about Amigas, some systems have a separate blanking signal and the monitor should be taking care of it. I don't think the Amiga does it this way but perhaps the way it implements blanking is causing the folds to be (partially) visible? You'd need to ask someone with similar setup (signal source and monitor type).

The thing is, it's not happening only with the Amiga. Even when I connect my Sega Master System using composite connection, the same issue occurs, the image cuts on the left and right side when the image is centered.

Deunan wrote on 2025-01-29, 17:21:

Note the folds will only be visible if the so-called border color is not black/off, and the screen edges are still visible. Perhaps the easiest solution to this problem is just widen the picture a bit so that the sides do move beyond CRT face. Maybe Amiga does it like this? Note there will not be any content here, except solid color, so you are not loosing any details. This is just the border color, meant to fill the screen sides. Just don't widen the picture too much or you will see the folds again, rolled over the actual picture - but that only happens if the picture is really way to wide vs horizontal sweep time, usually when the monitor can't sync properly to faster timings.

If I understand you correctly, you state that there are "filling" borders on each side that needs to be adjusted out of the visible area?
I'm not sure that this is the case. When adjusting the image left / right, a visible image comes up, no borders are seen on the edges.

Any idea how to tune the signal narrower so it fits the whole on the screen?
I thought about adjusting the 15.625kHz clock but as it looks from the schematic, all it does is to tune the image diagonally (resistor 3257). I'm attaching a screenshot from the schematic.

I'm a bit confused now as to what you consider an image "cut on the side". Can you make a short video, zoomed at the issue? The photos alone don't really explain it well. If possible also present what the width control is doing and what you'd consider a corrent and incorrect side of the image. That way we can be on the same page and not wonder if we are talking about the same issue.