Help me debug what is wrong with my Commodore 1084S-P CRT Monitor

[horttanainen]

Hello @Relayer ,

By the way, where are you getting these pin designations from???

There is a four wire connector on the PSU board. It has pins labeled from 1 to 4. On the schematics there are labeled lines (1M2, 2M2, and 3M2) coming from PSU from places like the cathode of D141, D142, and D143. These *M2 labels indicate the pin on the four wire connector socket on the PSU board.

Here's what you must do:
Make sure the PSU is attached to the monitor.
Disconnect one end of S494 (yes, you read correctly) and place the 240V bulb/s in as the dummy load for
the 126V rail on the cathode of D141. As I remember, you had two 240V bulbs. They should be in parallel with
each other when using them.

I did as you asked, but instead of desoldering S492 I cut the wire between S494 and D141.
The voltages measured:
Pin 1 measured 125.4V
Pin 2 measured 25.8V
Pin 3 measured 15.8V
Pin 4 is ground.

What I gather is that the loads on 26 volt and 16 volt rails pull the voltage down 0.2 volts which seems healthy.

At some point in your measurements you should see at least two voltages that should be correct. When that
hapends, then the fault will lie in the rail with the light bulb/s being used as a dummy load.

It seems that the fault lies in the 126 volt rail.

Thank you as always,
Horttanainen

 

[Relayer]

It seems that the fault lies in the 126 volt rail.

Spot on.
Next step is to remove the horizontal output transistor (TS467) which should be
connected to a largish heatsink.
Measure it to see if it's shorted. You may find an integral diode between the
collector and emitter. It should measure a diode drop one way only.
It's important that you measure it out of circuit, as there's very low ohms between
the base and emitter. Around 4 odd ohms due to R465, S465 and T401.
If the transistor measures OK, then leave it out of circuit, turn the monitor on, then
measure the cathode of D141 again.
If the voltage returns to normal, then TS467is unserviceable and needs replacement.
If the voltage is still low, we need to look elsewhere.

Please let us know how you get on.
Regards,
Relayer

 

[horttanainen]

Hello again @Relayer,

Next step is to remove the horizontal output transistor (TS467) which should be
connected to a largish heatsink.
Measure it to see if it's shorted. You may find an integral diode between the
collector and emitter. It should measure a diode drop one way only.
It's important that you measure it out of circuit, as there's very low ohms between
the base and emitter. Around 4 odd ohms due to R465, S465 and T401.
If the transistor measures OK, then leave it out of circuit, turn the monitor on, then
measure the cathode of D141 again.
If the voltage returns to normal, then TS467is unserviceable and needs replacement.
If the voltage is still low, we need to look elsewhere.

I have good news! TS467 is faulty. TS467 is a Philips BU508A NPN transistor (https://datasheetspdf.com/pdf-file/1144968/Philips/BU508A/1). I measured it out of circuit and got the following readings:

B -> C 275 ohms
C -> B 275 ohms

B -> E 100 ohms
E -> B 100 ohms

E -> C 178 ohms
C -> E 178 ohms

The internal circuitry seems to be shorted. I will order a replacement transistor. Do you want me to check any other components in the meantime?

Best Regards,
Horttanainen

 

[Relayer]

I have good news! TS467 is faulty.

I had a strong feeling it would be faulty.

TS467 is a Philips BU508A NPN transistor

The "A" in the part number tells me there isn't an integral diode built into it.
Therefore...

Do you want me to check any other components in the meantime?

Yes. I'll get you to check D467 and D468. D467 is used in place of an internal diode to the
horizontal output transistor. It's there to suppress high voltage transients that may occur.

Please let us know how you get on.
Regards,
Relayer

 

[horttanainen]

Hello again @Relayer.

Yes. I'll get you to check D467 and D468. D467 is used in place of an internal diode to the
horizontal output transistor. It's there to suppress high voltage transients that may occur.

D468 is a RGP10J diode which seems to be ok. The resistance from anode to cathode is 568 ohms with no reverse leakage.
D467 is a BY448 diode which seems to be broken. The resistance from anode to cathode is 586 ohms with reverse leakage around 285K ohms.

I will order a replacement for D467 as well. Are there any other components you would like me to check?

Thank you,
Horttanainen

 

[Relayer]

Are there any other components you would like me to check?

At this point there's no specific component that I need you to check, but I would
like you to test the 125 volt rail while the horizontal output transistor is out of circuit.
I'll get you to attach the 240V light globes to D141's cathode and measure the 125 volt
rail to see that it is correct. If it does read correctly, then we know with a very high
certainty that the horizontal output transistor is the only thing wrong with the unit.

Please let us know how you get n.
Regards,
Relayer

 

[horttanainen]

Hello @Relayer,
and again thank you.

I have a generic question on this debugging stuff: How did you know to ask me to check TS467? I am asking this because my guess would have been any component thats attached to ground and the 125 volt rail, like R494, C494, R469, C469, D468, D467 and so on. Did you ask me to check transistor TS467 because transistors are somewhat suspect, or is there some other reason? Furthermore, if transistor TS467 would have been fine what component would have you asked me to check next, and why?

BR,
Horttanainen

 

[Relayer]

How did you know to ask me to check TS467?

Since TS467's collector is connected to the 125 volt rail, that was the next logical step that
would load down that voltage rail. Loading it down would be a fault with that transistor.

I am asking this because my guess would have been any component thats attached to ground and the 125 volt rail, like R494, C494, R469, C469, D468, D467 and so on.

Those components you've listed would be much more reliable than TS467. Especially since semiconductors are more prone
to damage than passive devices, such as resistors and capacitors. But, you could always have an open resistor or a shorted
capacitor that could load down the 125 volt rail. The most obvious and easiest choice was to point the finger at the transistor
first.
If the 125 volt rail isn't normal after the removal of TS467, then a resistor or cap within the 125 volt rail could be the cause.

Furthermore, if transistor TS467 would have been fine what component would have you asked me to check next, and why?

I would get you to check C467, C468, C469, C470 and C471.
C467 to C470, are being used for bypassing, therefore a fair amount of ripple on the 125 volt rail would be introduced. Thus
loading the rail down. Though if any of those are shorted, the 125 volt rail wouldn't be able to be generated.
As for C471, it's used for coupling and if that was open circuit, no signal would be able to pass to the scan coils.
A short would allow the monitor to work, though you wouldn't see any symmetry with the picture.

I think I got the above correct.
I hope that answers your query.
I would still like you to check the 125 volt rail without TS467 in circuit.
Regards,
Relayer

 

[horttanainen]

Hello @Relayer,

and my apologies for the long wait. I have had practically zero spare time to work on this project in the past weeks.

At this point there's no specific component that I need you to check, but I would
like you to test the 125 volt rail while the horizontal output transistor is out of circuit.
I'll get you to attach the 240V light globes to D141's cathode and measure the 125 volt
rail to see that it is correct. If it does read correctly, then we know with a very high
certainty that the horizontal output transistor is the only thing wrong with the unit.

I attached the light bulbs and the monitor as instructed. I also had D467 out of circuit in addition to TS467. The voltage rail measured correctly as 125.5 volts.

Unfortunately, it seems that the post office lost the replacement for D467 in transit. This means that I have to order a new one and we will have to wait for another week or so.

I hope that answers your query.

Huge thank you for the thorough answer! This helps me understand what is going on and why.

As for C471, it's used for coupling and if that was open circuit, no signal would be able to pass to the scan coils.
A short would allow the monitor to work, though you wouldn't see any symmetry with the picture.

While we wait for the post office to try to deliver the replacement for the replacement of D467, I would like to ask you another question: where does the signal, that capacitor C471 couples, originate?

 

[Relayer]

and my apologies for the long wait. I have had practically zero spare time to work on this project in the past weeks.

No problems. I understand.

I attached the light bulbs and the monitor as instructed. I also had D467 out of circuit in addition to TS467. The voltage rail measured correctly as 125.5 volts.

I was hoping that the main HT voltage was OK.

Unfortunately, it seems that the post office lost the replacement for D467 in transit. This means that I have to order a new one and we will have to wait for another week or so.

Why am I not surprised. I too have had problems here with Australia Post.

Huge thank you for the thorough answer! This helps me understand what is going on and why.

You're welcome.

While we wait for the post office to try to deliver the replacement for the replacement of D467, I would like to ask you another question: where does the signal, that capacitor C471 couples, originate?

A pulsed signal originates from Pin 4 of IC402 into TS461, which is the horizontal driver transistor, which amplifies the signal into a transformer which AC couples It to the horizontal output transistor. C471 is there to block any DC, but only allows the AC signal generated to pass to the horizontal coils within the yoke assembly.
It's actually more complicated than the description above, but due to my age, it's been eons since I did any substantial circuit analysis'.
Sorry if i'm not precise.
I hope that it gives you the gist of what's going on.

Regards,
Relayer

 

[horttanainen]

Hello @Relayer,

and oh boy do I got good news for you!

Post office found the lost packet of replacement diodes for D467. I replaced TS467 along with D467. Finally the monitor starts up (very flickery video warning):

However, there is still some problem: the picture won't center horizontally and the TS467 heatsink becomes really hot after just a couple of minutes of being on.

More video on me trying out the different knobs to center the image (very flickery video warning):

Could it be that I ordered wrong replacement part for TS467? Left is the original part and right is the replacement.

Or could it be that some other replacement part was of wrong type? Should we double check replaced parts? Or is there some other more obvious reason behind the display not centering? To my understanding there is a voltage missing somewhere (or too much voltage).

A pulsed signal originates from Pin 4 of IC402 into TS461, which is the horizontal driver transistor, which amplifies the signal into a transformer which AC couples It to the horizontal output transistor. C471 is there to block any DC, but only allows the AC signal generated to pass to the horizontal coils within the yoke assembly.
It's actually more complicated than the description above, but due to my age, it's been eons since I did any substantial circuit analysis'.
Sorry if i'm not precise.
I hope that it gives you the gist of what's going on.

Thanks I got the gist of what's going on of the above description!

Best Regards,
Horttanainen

 

[BradtheRad]

Congratulations, this is progress!

It looks as though you have the monitor on free run, no signal coming in. This is a default mode of operating. It's not following instructions of any frame length or frame height or frame rate from an incoming video signal.

So does anything change when you connect a computer? Does the crt show the entire computer image? Then does the rightmost edge remain dark?

Until you know what amount of heat is normal, for the time being avoid leaving it on too long lest heat destroys a component.

 

[Relayer]

Could it be that I ordered wrong replacement part for TS467? Left is the original part and right is the replacement.

The big question is; did you order the part from China?
If so, then I'd say that is the problem of TS467 overheating.
Unfortunately China is notorious for producing fake semiconductors. Well below their ratings.
I cannot discern the manufacturer symbol on the new horizontal output transistor. Got to be a clone for sure.
If you did purchase the transistor from China, then you need to go to a reputable spare parts dealer and get
an original.
One other thing, the transistor you obtained shouldn't be an insulated type.

It looks as though you have the monitor on free run, no signal coming in.

As BradtheRad has stated, you're allowing the monitor to free run. You need to input it with a proper signal. Something
static would be desirable. Then we'd be able to see exactly if you are getting the verticals and the horizontals
correct, or they are just free running. eg. Runaway vertical frames (Vertical Hold problems) or runaway horizontal
or both.

So does anything change when you connect a computer? Does the crt show the entire computer image? Then does the rightmost edge remain dark?

Exactly as BradtheRad has stated.

Or is there some other more obvious reason behind the display not centering? To my understanding there is a voltage missing somewhere (or too much voltage).

There is a good possibility that there may be a fault with one of the integrated circuits. IC302 to be exact.
What I need you to do is to check the Vcc voltage on IC302. Which is Pin 9 of the IC. It should be around
26 volts.
If you find it OK, then start measuring all the other pins of the IC to determine if any of them are low or
high.
As I've said already you really need a static signal whilst doing those measurements. A pattern generator
would be an ideal signal source, but that isn't practical in this instance.

Please let us know how you get on.
Regards,
Relayer

 

[horttanainen]

Hello @Relayer and @BradtheRad,

It looks as though you have the monitor on free run, no signal coming in. This is a default mode of operating. It's not following instructions of any frame length or frame height or frame rate from an incoming video signal.

The monitor filled the whole screen with "background light" on free run prior to me breaking it. This is why I thought there would be something still wrong with the monitor. However, lets try feeding the monitor a signal...

So does anything change when you connect a computer? Does the crt show the entire computer image? Then does the rightmost edge remain dark?

When I connect a rasperry pi to the monitor (from some old freebsd project of mine) the monitor appears to show the whole signal but the rightmost edge remains dark:

Here is a picture of the same signal displayed on an Apple 2c monochrome monitor:

The big question is; did you order the part from China?
If so, then I'd say that is the problem of TS467 overheating.
Unfortunately China is notorious for producing fake semiconductors. Well below their ratings.
I cannot discern the manufacturer symbol on the new horizontal output transistor. Got to be a clone for sure.
If you did purchase the transistor from China, then you need to go to a reputable spare parts dealer and get
an original.
One other thing, the transistor you obtained shouldn't be an insulated type.

I ordered the part from europe but apparently the origin of the component is china. The funny thing is that I tried to google the manufacturer prior to ordering this part but I too could not find any manufacturers with this "SH" logo. I think this should have put me off but I could not resist ordering a component with the same initials as me.

I ordered a new replacement part for TS467. This time an original Philips BU508A. The item should arrive before christmas.

There is a good possibility that there may be a fault with one of the integrated circuits. IC302 to be exact.
What I need you to do is to check the Vcc voltage on IC302. Which is Pin 9 of the IC. It should be around
26 volts.
If you find it OK, then start measuring all the other pins of the IC to determine if any of them are low or
high.
As I've said already you really need a static signal whilst doing those measurements. A pattern generator
would be an ideal signal source, but that isn't practical in this instance.

I measured all the pins on IC302 while feeding the monitor with the same composite signal as in the video. The voltages were:

PIN	Indicated Voltage	Measured Voltage
1	7	6.9
2	1.4	1.4
3	5.5	6
4	2.4	2.1
5	25	25.2
6	15	12.7
7	7.5	5.6
8	GND	Did not measure
9	26	25.3
10	1.8	2.1
11	0.24	6.1
12	???	21
13	4.5	4.1

Pins 6, 7, and 11 have voltage levels differing most from the indicated voltage levels.

Best Regards,
Horttanainen

 

[BradtheRad]

You gained success by persevering! The display is stable. Characters are crisp and bright.

Now that it's working is a good time to record more voltage readings to create a reference of normal operation. At the horizontal output transistor, etc.

The right edge is still dark. Yet you can move the image right and left. In fact almost so it reaches the right hand edge.

So near and yet so far.

The beam needs greater voltage swing to reach all the way to the right. We might wonder if the adjustment potentiometer needs to be replaced. However that won't help if your high voltage is insufficient at the start. To diagnose further requires more understanding of the high voltage section.

 

[Relayer]

OK, since your monitor is showing a horizontal shift fault, I got you
to test the wrong IC.
It's IC402 that I need you to check all the voltages on.
I deeply apologise for this.
By the way, could you please provide a part number for IC402 as well
as the part number for IC302.
Thanks.
Regards,
Relayer

 

[fbone]

Guys just wanna say this thread is superb and helped me a lot as well! I registered to the forum because of it and I am amazed from the amount of knowledge found here!

I wont hijack the thread, maybe when you finish repairing your monitor and I truly hope so - these are beauties, I will then ask why mine has some fade vertical banding on the left side.

Congratulations for your persistance.

 

[horttanainen]

Hello again @Relayer,

OK, since your monitor is showing a horizontal shift fault, I got you
to test the wrong IC.
It's IC402 that I need you to check all the voltages on.
I deeply apologise for this.

No need to apogolise, I should have guessed you meant horizontal combinator IC402 since I had already replaced it earlier when we were debugging the short.

Here are the voltages for IC402:

PIN	Indicated Voltage (on the monitor schematic)	Measured Voltage
1	-	0
2	0.2	0.4
3	6.4	6.9
4	0.9	0.9
5	GND	Did not measure
6	1.6	1.5
7	-	4
8	6.2	6
9	???	1.5
10	2.8	5.4
11	0.13	5.4
12	0.4	5.8
13	???	11.5
14	6	5.8
15	11.8	5.8
16	6	7
17	4	1.9
18	6	5.5

As you can see many of the voltages differ greatly from the indicated values. I suspect this this normal as the picture is almost perfect. I could not determine which pin should drive the horizontal deflection circuit. Maybe 4? Here is the schematic: http://pdf.datasheetcatalog.com/datasheet/philips/TDA2595.pdf

By the way, could you please provide a part number for IC402 as well
as the part number for IC302.

I replaced IC402 earlier when we debugged the short causing the PSU voltages to drop. Here are the part numbers exactly as they are written on the component.

original IC402:
PHILIPS
TDA2595
359590G
JSD8823 7

Current IC402:
PHILIPS
TDA2595
215230
DSS9505 9

IC302:
TDA2653A
DSD8744 5
025920

Guys just wanna say this thread is superb and helped me a lot as well! I registered to the forum because of it and I am amazed from the amount of knowledge found here!

I wont hijack the thread, maybe when you finish repairing your monitor and I truly hope so - these are beauties, I will then ask why mine has some fade vertical banding on the left side.

Congratulations for your persistance.

Thank you. I suggest you start a new thread as I am planning to mark this thread as "solved" when we solve the remaining horizontal black bar problem. Please post a link to your thread here when you do!

Best Regards,
Horttanainen

 

[Relayer]

OK. Pin 15 is way too low for my liking. As far as I can determine it should be the IC's main power rail.
I'll get you to check R443 (8.2Ω), R436 (8k2Ω), R438 (10kΩ), R441 (120kΩ) and if you have a capacitance
and/or a ESR meters, remove and test C433 (100μF 16V), C446 (1μF 63V) and C447 (4.7μF 25V). If you have
neither meters, then replace them instead.
You can measure the resistors in circuit, but if you get strange readings, pull one leg out and remeasure.

Please let us know how you get on.
Regards,
Relayer

--- Updated Dec 21, 2020 ---

@fbone
Could you please take a photo of the screen so I can see exactly what is being displayed?
Can you please make sure the picture is static.
Thanx.
Regards,
Relayer

 

[fbone]

@horttanainen thank you for your understanding! I created a new thread here: https://www.edaboard.com/threads/commodore-1084s-monitor-with-faint-vertical-lines.396322/

@Relayer thank you for your interest - I have attached a photo in the new thread!