PDA

View Full Version : IBM 5155 Jittery Display



Shadow Lord
April 8th, 2012, 01:55 AM
I am afraid I know what the answer to this one is going to be but just to be sure:

My 5155 has a jittery display (jumps, bounces, flickers). Could it be something simple like a bad CGA card? I've already ruled out other sources of interference and the power to the unit is good (going through an APC UPS). Or is it the PSU going bad? I am guessing a PSU problem but want to be sure. Thanks.

pearce_jj
April 8th, 2012, 02:53 AM
I'd just offer a word of caution in dealing with it. Be VERY careful poking about inside the CRT part, the voltage will kill if you make direct contact with it.

kiyotewolf
April 8th, 2012, 03:10 AM
I have the book on repairing computer monitors.



~Paul

SpidersWeb
April 8th, 2012, 03:51 AM
Not sure if this helps at all, but similar generation gear so might be useful to know. My IBM 5151 monochrome display was flickering, well rather jumping around occasionally.

A little while later it started squealing and wouldn't come back up afterwards. It was a regulator transistor in the PSU for the display, technician replaced that one part and now it's like new again.

Shadow Lord
April 8th, 2012, 09:21 AM
Yeah, I am thinking PSU myself on this. I just hate to take the thing apart. Inside it is mint. All the original screws, original parts, etc. Whoever had it did a good job taking care of it. They even did the 640KB hack (http://www.ibmmuseum.com/ohlandl/5155/5155.html#640K_Hack) on it! Too bad they din't use a jumper block. They went w/ a straight short.

Shadow Lord
April 8th, 2012, 09:23 AM
Paul is it in PDF format? Or you literally have the book?

modem7
April 8th, 2012, 03:26 PM
http://www.minuszerodegrees.net/images2/5155_p12_connected.jpg

The CRT unit is supplied with +12V from the 5155 power supply. From that +12V, the CRT unit generates the other voltages that it uses (+16V, +50V, ... 12500V).

If I was in your situation, I would supply the +12V to the CRT unit via a second power supply. That would quickly inform me as to whether the problem cause was in the 5155 power supply, or in the CRT unit itself.
The second power supply needs to be able to supply the +12V at at least 1.5 amps (the CRT unit draws up to 1.5 amps). Any PC power supply should do. Even the 63W IBM one can supply 1.5 amps on the +12V line.

I of course don't know what resources are available to you, but below is how I would connect the second power supply up to the CRT unit - via leads that have parrot clips on the end.
I have done this myself, and (as expected) works.

http://www.minuszerodegrees.net/images2/5155_p12_parrot.jpg

Shadow Lord
April 9th, 2012, 09:43 AM
Modem7,

Thats a great idea. I did not realize it was that easy to access the power supply of the CRT. Your sugegstion will help narrow the problem down pretty quickly to CRT vs PSU. Thanks.

Shadow Lord
August 4th, 2012, 07:38 PM
Well,

After a long hiatus I finally got back to this project. I did as modem7 suggested (see below for a big tip) and supplied the 12V directly to the monitor and viola the jittery display was gone. Just for giggles I reconnected the 5155's PSU to the monitor again and no jittery display. No idea why, but running on its own PSU and running on the external 12V produces the same performance characteristics now so "problem fixed"....

TIP: The 5155 is cramped beast of a case to work in. And taking the PSU and monitor apart is no fun. However, it is totally unnecessary for this particular test. As it turns out the power "cord" to the monitor runs in a cable conduit underneath the 5155 and can be easily accessed by opening up just two blue screws. If your 5155 has been modified your cables maybe running elsewhere but stock thats how IBM ran them. Wish I has known that before I took the whole thing apart! ;)

modem7
August 4th, 2012, 08:12 PM
After a long hiatus I finally got back to this project. I did as modem7 suggested (see below for a big tip) and supplied the 12V directly to the monitor and viola the jittery display was gone. Just for giggles I reconnected the 5155's PSU to the monitor again and no jittery display. No idea why, but running on its own PSU and running on the external 12V produces the same performance characteristics now so "problem fixed"....
To me, that sounds like the P12 connector had developed a poor connection to its mating conector.

Franklin
October 8th, 2012, 10:23 AM
Hi,
I have the same problem. A question: is it possible to use the PSU from a 5160 to do this test?

Shadow Lord
October 8th, 2012, 10:34 AM
Hi,
I have the same problem. A question: is it possible to use the PSU from a 5160 to do this test?

As I understand it any 12V power supply should work. Just make sure the amperage. Although, honestly a 5160 PSU should have enough juice (as long as it is not powering anything else) to power a small monitor. Other's with actual EE backgrounds can chime in and verify.

modem7
October 8th, 2012, 10:26 PM
I have the same problem. A question: is it possible to use the PSU from a 5160 to do this test?
Welcome to these forums.
The cause of Shadow Lord's problem was most likely a poor connection in the P12/J12 plug/socket, fixed when he unplugged them from each other and then reconnected them.

http://www.minuszerodegrees.net/images2/5155_p12_connected.jpg

If you haven't done that already, that should be your first attempted repair action.

modem7
October 8th, 2012, 10:52 PM
As I understand it any 12V power supply should work. Just make sure the amperage. Although, honestly a 5160 PSU should have enough juice (as long as it is not powering anything else) to power a small monitor. Other's with actual EE backgrounds can chime in and verify.
Per post #7, the CRT unit of the 5155 will draw up to 1.5 amps from the +12V line.

The +12V line on the IBM PC's 63W power supply can deliver up to 2 amps (source: IBM Technical Reference).
The +12V line on the IBM XT's 130W power supply can deliver up to 4.2 amps (source: IBM Technical Reference).

So, any power supply designed for PC, XT or AT should be fine.

Whatever power supply is used, there is the possibility that only having its +12V line used, will not be enough for the power supply to start. Something, such as a floppy drive, may need to be connected up to one of the power supply's +5V/+12V plugs to enable the power supply to start.

Franklin
October 9th, 2012, 05:36 AM
Thanks for these advices. I think i will start with the P12/J12 connection.
half an hour later: I did not find this P12/J12 connection. On that spot was another connector (not sure where this is for it seems related to a noise killer). If i disconnect it and restart then everthing seems to work normally (without jittery!). If anyone has got an idea what this connector is for, please let me know. For now it is working so thanks.

10535

modem7
October 9th, 2012, 11:40 PM
Thanks for these advices. I think i will start with the P12/J12 connection.
half an hour later: I did not find this P12/J12 connection.
See the "TIP" left by Shadow Lord in post #9. You can get to P12/J12 from underneath the unit.


On that spot was another connector (not sure where this is for it seems related to a noise killer). If i disconnect it and restart then everthing seems to work normally (without jittery!). If anyone has got an idea what this connector is for, please let me know. For now it is working so thanks.
The P13/J13 connection that you've pulled apart is part of the +12V line from the power supply to the power supply's fan.

It appears that a previous owner has 'tapped into' that +12V line in order to, I'd say, power that white rectangular box that they have added (top left of photo).

So I'm guessing that the white rectangular box was causing a problem to the +12V line, which in turn affected the CRT unit.

Right now, with the P13/J13 connection pulled apart, the power supply has its fan off and thus the power supply may overheat.

What is the writing on the white rectangular box ?

Franklin
October 10th, 2012, 01:21 AM
ok, i will try to find the P12. The white box has got the name 'noise killer'. As i understood from your reaction this is not an original part of the IBM 5155. If that is the case then i will try to remove it, it is not original and the noise is still present, what was the previous owner thinking :-). Beside that the noise is part of these machines. I will make an update if i am able to remove this appendix

modem7
October 10th, 2012, 01:57 AM
The white box has got the name 'noise killer'. As i understood from your reaction this is not an original part of the IBM 5155. If that is the case then i will try to remove it, it is not original and the noise is still present, what was the previous owner thinking :-). Beside that the noise is part of these machines. I will make an update if i am able to remove this appendix
'Noise' will be in regard to electrical noise, not audible noise.
I'm guessing that the 'noise killer' is one or two capacitors in a box, a filter.

So the question now is, why would the previous owner add it? It's as if the owner determined that there was a more-than-normal amount of 'noise' on the +12V line, and then added the filter to quiet things down (electrically speaking).

When you unplugged J13/P13, you were disconnecting both the fan and the 'noise killer' from the power supply's +12V line. Just that act got rid of the CRT unit jitter. That suggests that the source of noise is either the 'noise killer' or fan.

Perhaps the fan went faulty during the ownership of the previous owner, faulty in way that didn't stop it turning (completely), and induced noise into the +12V line, thus affecting the CRT unit (assumption: CRT unit runs from same +12V source in power supply). The owner added the 'noise killer' to remove the noise, but enough noise remained to still result in some minor CRT jitter.

Franklin
October 10th, 2012, 07:18 AM
Hi Modem7,
Thanks for the clarification. I just contacted the previous owner. He could remember that in the past a 220V fan (this is not correct i think, it was before also 12V) was replaced by a 12V, since then the screen was jittery. I suspect that the original fan is not working (assume this is part of the PSU), the current solution is crappy, i think i have to find another PSU then (or a proper fan).

Anyway good analysis!

Springbok
November 24th, 2018, 07:19 AM
This is an old thread but I found it because my 5155 developed jitter. I am STUNNED that simply unplugging and replugging J12/P12 fixed my issue. It wasn’t even loose. What the hell is this voodoo magic?

T-R-A
November 24th, 2018, 11:19 AM
It wasn’t even loose. What the hell is this voodoo magic?

Any connector (especially ones near anything hot) can develop a degree of corrosion over the years. Likely the "unplugging-plugging" loosened some of that and the J12/P12 connector is marginally "cleaner" than before.