PDA

View Full Version : Attemping to fix an IBM PS/2 Model 50Z


Allen
April 16th, 2008, 02:22 PM
I have an IBM PS/2 Model 50Z (type 8550) with data I would like to retrieve. Unfortunately, the power supply appears to be dead. Since it would seem difficult to get another compatible power source, is there anyway I can revive the power supply or interface the hard drive to a modern PC?

The system seems to be in good shape otherwise, including many of the original documents. I just want to get rid of the system, but it would be nice if I can get to some of the data. Any suggestions are welcome.

strollin
April 17th, 2008, 07:42 AM
It might be easier to remove the HD and attempt to temporarily install it in a working machine to be able to recover the data.

IBMMuseum
April 17th, 2008, 09:33 PM
I have an IBM PS/2 Model 50Z (type 8550) with data I would like to retrieve. Unfortunately, the power supply appears to be dead. Since it would seem difficult to get another compatible power source, is there anyway I can revive the power supply or interface the hard drive to a modern PC?

The system seems to be in good shape otherwise, including many of the original documents. I just want to get rid of the system, but it would be nice if I can get to some of the data. Any suggestions are welcome.

The 50Z had a "DBA ESDI" HDD, proprietary to a selection of PS/2 models:

50Z
N51SX, N51SLC (a smaller size, so not an easy swap)
53SLC2 (it can be made to work)
55SX
CL57SX (smaller size again)
70
P70
8590 (It can be made to read, not to boot from; the 9590 removed the planar DBA ESDI connections)
ThinkPad 720 (smaller drive)

Unless you have another one of the models (and the normal design for DBA ESDI has only a single HDD interface) the other step might be to get another Model 50 or Model 70 PSU (I have seen them appear on eBay).

Allen
April 18th, 2008, 09:37 AM
I appreciate the information about similar PS/2s. The power supply was working recently, but the system had problems booting because of error codes 161 and 163 (battery failure and time/date not set). Even after I installed a new battery, I would get the same error codes above. What would continue to cause these error codes, even with a newly installed battery?

If I replace the power supply, I may get the same errors. I was about to give up and strip the unit and give the parts away on this forum, unless I can figure out what to do next.

Vlad
April 18th, 2008, 09:44 AM
When the battery dies it loses its setup information. You need to use its Setup Disk do restore the information it needs in the CMOS. There are members here that can link you to the disk image so you can make a floppy and get back in business. This is a pretty common thing with errors 161 and 163, just needs the setup disk is all.

-Vlad

Allen
April 18th, 2008, 09:49 AM
Actually I tried using the reference diskette, the original and backup. Still received those error codes. Does a bad power supply cause problems with the BIOS or CMOS?

modem7
April 18th, 2008, 04:08 PM
Could the replacement battery (6V) been low, or perhaps not properly connected. Do you have the means to measure the battery voltage?

modem7
April 18th, 2008, 04:51 PM
Does a bad power supply cause problems with the BIOS or CMOS?
Not normally, but I have seen it in a few situations where a motherboard in an XT was replaced by a clone AT motherboard (and keyboard). The machine suffered from an inability to hold the CMOS setup for more than a few days (on each days, machine turned on in morning then off in afternoon). The problem didn't go away until the power supply was replaced by a modern one (as used in AT machines).

That problem could have been that the old XT power supply was underrated for the job, but there was talk at the time that AT power supplies were different to XT ones. Sure the AT power supply uses a pin that is unused in the XT power supply (pin 2 - extra +5V line), but there was talk at the time that other things may be happening such as delayed activation of the 'power good' line.

Anyway, that's an example of a situation where the power supply affected the CMOS setup. So maybe you can't rule out a faulty power supply as a cause.

Allen
April 18th, 2008, 06:35 PM
I appreciate the explanation about the possible power supply issue. In regards to the battery, I tested it with a multimeter and it showed +6 V. I didn't see any physical damage to the red and black wires connecting the battery compartment to the system board. It's possible the system board is bad somehow.

Druid6900
April 18th, 2008, 07:40 PM
I recall there being a wire jumper block on some of the 50s, I think around where the battery attached in circuit, that you had to reverse to clear the CMOS when you replaced the battery, then you re-reversed it for normal operation.

modem7
April 18th, 2008, 07:42 PM
A partial check of the power supply is to check that the output voltages are within specs.
Page 141 of the following document shows the power supply connector and the expected voltages:
http://ps-2.kev009.com:8081/ohlandl/books/ps2-hmm.pdf

Even though the 50Z power supply may fire up when removed from the computer, the measurement should be done with the power supply connected (under load).

modem7
April 18th, 2008, 08:32 PM
I recall there being a wire jumper block on some of the 50s, I think around where the battery attached in circuit, that you had to reverse to clear the CMOS when you replaced the battery, then you re-reversed it for normal operation.
What you're taking about is covered on pages 376/377 of the document I pointed to. That section is about removing a power-on password, however sentences there such as, "Make sure you either backup, or make note of the current configuration before you erase the password.", suggests that the entire CMOS data is being cleared, not just the password. Worth a try.

IBMMuseum
April 18th, 2008, 09:38 PM
What you're taking about is covered on pages 376/377 of the document I pointed to. That section is about removing a power-on password, however sentences there such as, "Make sure you either backup, or make note of the current configuration before you erase the password.", suggests that the entire CMOS data is being cleared, not just the password. Worth a try.

I know that the trick works on a 55SX, but don't remember with the 50Z (no for the regular 8550 non-Z: it had a slot connector on the system planar for the speaker and battery module). Dunno why the 161 & 163 errors remain if you let it go all the way through configuring the system and setting the time. You have to do that to get past those errors.

Allen
April 19th, 2008, 01:15 PM
I received this IBM PS/2 Model 50Z from the original (previous) owner who could not boot the system because of the 161 and 163 error codes. He replaced the 6-volt battery and tried many boot diskettes, including the original reference disk, backup reference disk, and various DOS boot disks. Even with a new 6-volt battery installed, the system would still display error codes 161 and 163. Because it was hung-up on these errors, the system would not boot, whether from the floppy or hard disk. The original owner was only able to get the system to display IBM basic.

He gave up, and let me have the system. I tried everything he did, and using my multimeter I verified that the system board was receiving 6 volts from the battery (I checked the red and black wires and the connections from the battery compartment to the plug on the system board). Still, I cannot get past those error codes, and pressing F1 on the keyboard only brings me to IBM basic. Therefore I cannot verify if the floppy drive or hard drive are functional.

Now the power supply appears to be dead. When plugged in and turned to the on position, the power supply does nothing, except that the yellow light yields one quick blink, but neither the yellow nor green light stays illuminated.

Because the power supply is dead, I cannot post pictures of the screen, but if you go to
http://www.neowin.net/forum/index.php?showtopic=624470 you can see a similar problem (if someone is registered on neowin.net, tell the original poster to replace the battery). This appears to be the most common issue with IBM PS/2s taken out of storage.

To respond to previous replies, I see no jumpers in or around the battery compartment, therefore I see no way to clear or reset the CMOS. In addition, I cannot access the PS/2 hardware maintenance manual (referenced site not functioning).

modem7
April 19th, 2008, 04:28 PM
I cannot access the PS/2 hardware maintenance manual (referenced site not functioning).
A copy of the manual is now at: http://members.dodo.com.au/~slappanel555/misc/ps2_hardware_maint_oct94.pdf

To respond to previous replies, I see no jumpers in or around the battery compartment, therefore I see no way to clear or reset the CMOS.
Page 377 of the manual explains the situation - some of the mode 50's have no jumper and an alternate method needs to be used.

Now the power supply appears to be dead. When plugged in and turned to the on position, the power supply does nothing, except that the yellow light yields one quick blink, but neither the yellow nor green light stays illuminated.
From memory, one of those lights is the power supply light and the other is the hard drive access light (it was about '92 the last time I worked on a model 50Z).
The power supply was dead in your original post, then came good, and is now dead again. I bet that it will come good again, but only for a while. You might find that it is related to temperature. I've seen that a few times (in various bits of equipment) and in all cases was caused by a capacitor in the power supply (one whose ESR I suspect deteriated to the point where the ESR was borderline for the power supply to start up).
Even if the power supply now never fires up, there is a good chance that replacing the low value electrolytic capacitors in it will get it going again. But, maybe the power supply is shutting down because of a faulty component on the motherboard.

Still, I cannot get past those error codes, and pressing F1 on the keyboard only brings me to IBM basic. Therefore I cannot verify if the floppy drive or hard drive are functional.
Okay. Your earlier, "Actually I tried using the reference diskette, the original and backup. Still received those error codes" suggested to me (and probably others) that you were able to boot from the reference disk and therefore go through the setup process.

So your course of action would be:
1. Get power supply going, then
2. Determine/rectify why the machine isn't booting from the diskette drive (power supply might be cause of that), then
3. Boot from Reference Diskette and go through setup process.

Allen
April 19th, 2008, 06:06 PM
I should clarify two points:

modem7 said "The power supply was dead in your original post, then came good, and is now dead again." Actually it was functioning fine, then it became dead. My last post should have been my original post. I was just giving more details in my last post. The original owner didn't have any problems with the power supply; I did not have any problems for the first few days, until the system shut down (powered off) by itself. Now I cannot turn on the system. Clearly, the power supply is dead.

In regards to using the reference disk or any boot disk, I inserted the disks, but nothing happened as a result. No setup screen--nothing. The only two screens I ever came across were the ones showing the error codes and IBM basic (like the pictures in the link I provided).

I'm concerned if I replace the power supply, I will still get the same error codes, unless a bad or failing power supply has something to do with those error codes (in addition to a bad 6-volt battery).

modem7
April 19th, 2008, 08:02 PM
I'm concerned if I replace the power supply, I will still get the same error codes, unless a bad or failing power supply has something to do with those error codes (in addition to a bad 6-volt battery).
Yes, no guarantees. You could have two faulty components now - power supply and motherboard.

IBMMuseum
April 19th, 2008, 09:42 PM
...In regards to using the reference disk or any boot disk, I inserted the disks, but nothing happened as a result. No setup screen--nothing. The only two screens I ever came across were the ones showing the error codes and IBM basic (like the pictures in the link I provided)...

You may have to press F1 for older PS/2s. It didn't run the configuration, so it didn't clear the errors. Hard to tell if it is anything other than the PSU at this point.

Allen
April 19th, 2008, 11:08 PM
I disconnected the power supply from the system board, plugged it in, and to my surprise the power supply powered on with both the green and yellow lights illuminating. Previously, I could not get the power supply to come on when connected to the system board (see my previous two posts).

Page 141 of the referenced PS/2 maintenance manual (PDF page 155) shows the power supply voltage check for Models 50 and 70 (also attached here as Manual.jpg). Using my multimeter, I tested the power supply and got these results:

Pin 2, Pin 1: -11.35
Pin 2, Pin 3: +11.44
Pin 2, Pin 15: +5.35

The second and third results are not within the range specified by the manual. I'm confused about the values in the manual and the values on the power supply label (attached as Label.jpg)--they appear to be different. Do I have a bad power supply?

modem7
April 20th, 2008, 03:46 AM
Interesting to see a switch-mode power supply that doesn't need a load to start up.

Based on the power supply label, it looks like someone goofed when creating the manual.
Pin 3 is now obviously -12V in which case the minimum value is probably +9V based on the +12V specs (although 9V in my opinion is too low for a 12V line).
Pin 15 is now obviously +5V in which case the minimum value is probably going to be about the +4.5V mark.

The figures you measure look good. The manual doesn't tell us whether to make the measurements with the power supply connected or disconnected. Normally power supply measurements are made with the power supply connected.

Does the power supply stop working again when you plug it into the chassis?

If so, does the power supply then power up with the hard drive removed, or the diskette drive removed?

Allen
April 20th, 2008, 02:44 PM
The power supply has only one connector, and that is the 50-pin in the picture in my last post. It connects to the system board, and the drives, fan, and system board receive power from this 50-pin connector. It would be nearly impossible to test the power supply if it were connected to the system board.

The power supply still does not work when connected to the system board. I tried without the drives connected, but it still does not work. It only seems to work when it's not connected to the system board.

I can only conclude something on the system board is telling the power supply not to power on. I didn't see anything unusual about the system board except I had never seen a fuse on one before (picture attached). What does this fuse do? This is the only thing I see that I can easily replace.

modem7
April 21st, 2008, 03:33 AM
The power supply still does not work when connected to the system board. I tried without the drives connected, but it still does not work. It only seems to work when it's not connected to the system board.
I can only conclude something on the system board is telling the power supply not to power on.
You can't rule out the power supply. I've seen power supplies that fail under load.

I didn't see anything unusual about the system board except I had never seen a fuse on one before (picture attached). What does this fuse do? This is the only thing I see that I can easily replace.
On some motherboards, the +5V to the keyboard connector is fused. It might be the same for the model 50Z. Someone else might know for sure.

Allen
April 26th, 2008, 10:19 AM
You're correct about the fuse, it's related to the PS/2 keyboard port.

The power supply might be failing under load, or maybe the system board is bad. I've run out of ideas; I don't know what else to do. I would prefer to restore the system instead of disassembling the system for parts, but I don't really want to spend too much more time or money on it. My heart really isn't into IBM PS/2s, so I wouldn't mind giving the parts or the whole unit to anyone who can restore it.

Druid6900
April 26th, 2008, 10:57 AM
Troubleshooting 101;

Remove and/or disconnect everything but the video card and see if the supply powers up

If so, it's one of the things you've disconnected, if not, remove the video card and see what the PSU does.

If it powers up, it's the video card, if not, it's the motherboard.

If it's the motherboard, remove any options (co-pro) and try it.

If it still doesn't power up, remove the ram, one bank/set at a time and try it. Save the ram in the order you took it out.

When there is no ram left, put the bank/set that you took out FIRST in (intelligent substitution) and try it.

If available, substitute any CPU that will work in it with only minimal ram.

If it STILL doesn't work, it isn't going to.

Allen
April 27th, 2008, 02:35 PM
I appreciate the troubleshooting tips, but I tried all that, and I did it again today. The power supply tests fine if not connected to the computer, but if the power supply is connected to the system board without any the drives, fan, internal speaker, expansion cards, memory, etc. plugged in, the computer will not turn on. The lights on the power supply do not illuminate (although I can hear a humming sound as if it was on).

It sounds to me the system board is bad. If the power supply is failing under load, shouldn't it still power up with the lights on if the system board has no drives, fan, internal speaker, expansion cards, memory, etc. connected to it?

In short, it doesn't seem to matter what is connected or disconnected from the system board, except for the power supply itself.

IBMMuseum
April 27th, 2008, 02:37 PM
I appreciate the troubleshooting tips, but I tried all that, and I did it again today. The power supply tests fine if not connected to the computer, but if the power supply is connected to the system board without any the drives, fan, internal speaker, expansion cards, memory, etc. plugged in, the computer will not turn on. The lights on the power supply do not illuminate (although I can hear a humming sound as if it was on).

It sounds to me the system board is bad. If the power supply is failing under load, shouldn't it still power up with the lights on if the system board has no drives, fan, internal speaker, expansion cards, memory, etc. connected to it?

In short, it doesn't seem to matter what is connected or disconnected from the system board, except for the power supply itself.

I've got other 50Z planars, if you want to go down that path...

Druid6900
April 27th, 2008, 08:15 PM
I appreciate the troubleshooting tips, but I tried all that, and I did it again today. The power supply tests fine if not connected to the computer, but if the power supply is connected to the system board without any the drives, fan, internal speaker, expansion cards, memory, etc. plugged in, the computer will not turn on. The lights on the power supply do not illuminate (although I can hear a humming sound as if it was on).

It sounds to me the system board is bad. If the power supply is failing under load, shouldn't it still power up with the lights on if the system board has no drives, fan, internal speaker, expansion cards, memory, etc. connected to it?

In short, it doesn't seem to matter what is connected or disconnected from the system board, except for the power supply itself.

If you went through the list systematically and you still don't have PSU activity, then one of the chips probably has a dead short to ground.

You could probably track down which one it is by placing a fingertip on each chip on the board. The one that burns the chip logo into your finger would be the one. :)

If you aren't handy at unsoldering chips from multilayer boards, I'd take IBM up on his suggestion.

Sharkonwheels
April 27th, 2008, 08:52 PM
No to ThreadJack, but Dru - I was planning on unsoldering the CPU off of one of my Morrow MD-3 mainboards, and installing a socket.

Will this be an issue? Anything I need to know? It's a standard 40-pin DIP NEC D780C Z80 clone...


T

linemanduke
April 27th, 2008, 09:03 PM
if worse comes tpo worse i have a 70 and a p70 with parrallel port cd writer i could put your stuff on to cd for ya

linemanduke
April 27th, 2008, 09:14 PM
also any one think his fdd may be bad? when the one went on my p70 nothing would happen except the power supply would hum. I thought the power supply had went, but like Allen's would also test good off system. it may be your drive is shorted or something. try running the computer with floppy drive disconnected. floppy disk drive being bad would also explain the reference disk problem. the drive i replaced in my second p70 would act like it was accessing. but wasn't and would some times give me an error and kick over into ibm basic.

btw sorry for lack of grammar its late i'm tired and i went to a charter school with questionable grammar teaching methods LOL.

Terry Yager
April 27th, 2008, 10:19 PM
also any one think his fdd is bad when the one went on my p70 nothing would happen except the power supply would hum i thought the power supply had went but would also work off system may be your drive is shorted or something try with floppy disconected. would also explain the reference disk problem my previous p70 drive would look like it was accessing but was not and would some times give me an error aand kick over into ibm basic.

???:roll:

--T

Terry Yager
April 27th, 2008, 10:23 PM
also any one think his fdd is bad when the one went on my p70 nothing would happen except the power supply would hum i thought the power supply had went but would also work off system may be your drive is shorted or something try with floppy disconected. would also explain the reference disk problem my previous p70 drive would look like it was accessing but was not and would some times give me an error aand kick over into ibm basic.

Windage & elevation, d00d! No, wait...fall back and regroup...make that gramm'rage & punctuation!

--T

linemanduke
April 27th, 2008, 10:35 PM
sorry corrected grammer and punctuation .............i think lol:rolleyes:

linemanduke
April 27th, 2008, 10:37 PM
Troubleshooting 101;

Remove and/or disconnect everything but the video card and see if the supply powers up


i need to learn to read lol

hdd copy offer still stands

Allen
April 27th, 2008, 11:35 PM
The unit will not power on even if just the floppy drive is disconnected. Thank you linemanduke for the offer, but since the original owner said it was not absolutely necessary to retrieve the data, I won't spend any unnecessary time and money to read the disk. I might be able to find another compatible PS/2 in my area to connect the hard disk. Otherwise, it's not a big deal.

Since I'm not an expert in electronics and have no experience in soldering and unsoldering connections, I will not attempt to fix the system board. I just want to be sure the real problem is the system board and not the power supply. I didn't know a system board can stop a power supply from working (or tell it not to power on properly). This appears to be the case.

Terry Yager
April 27th, 2008, 11:44 PM
sorry corrected grammer and punctuation .............i think lol:rolleyes:

Tnx! I actually read thru the post a few times, and still was unable to de-cypher what you were trying to say.

--T

linemanduke
April 28th, 2008, 04:27 AM
Tnx! I actually read thru the post a few times, and still was unable to de-cypher what you were trying to say.

--T

ok lol i was saying i thought floppy drive may have been bad. and that i had had somthing similar happen. in short terms

Druid6900
April 28th, 2008, 12:33 PM
No to ThreadJack, but Dru - I was planning on unsoldering the CPU off of one of my Morrow MD-3 mainboards, and installing a socket.

Will this be an issue? Anything I need to know? It's a standard 40-pin DIP NEC D780C Z80 clone...


T

Several.

You will have to apply heat to the pin longer than with, say, a double sided board to get the solder to melt all the way up the tube (via) and any given pin may be attached to several lands (I hate to say this) via the via.

Too little heat and it won't melt all the way and you won't extract all the solder. Too much heat and the pad you are working on will lift.

When you think the tube is clears, put the soldering tip against the pin (don't touch the pad) and push it one way then the other several times until it wiggles easily and the top side pad of the pin shows no solder. Then that pin is free. Repeat for all the other pins.

Power and ground are a bitch, but, you can keep the tip in contact for longer without damaging the pads as they are usually much bigger.

If the tube doesn't completely clear on the first attempt, don't keep re-heating and trying to remove the solder. Let it cool, re-solder it, let it cool and try the de-soldering again.

When you THINK you have all the tubes clear, gently pry up one end of the chip and then the other, but, not too far or too hard. If you do, you'll rip the the plating right out of the hole and you'll be screwed.

Rock the chip end to end while observing the pins. Any that aren't moving should be re-soldered and desoldered again.

And, finally, if you don't HAVE to remove a chip from a ML board, don't.

You guys thought this stuff was easy, didn't ya :)

Druid6900
April 28th, 2008, 12:39 PM
I didn't know a system board can stop a power supply from working (or tell it not to power on properly). This appears to be the case.

Most switching power supplies contain what is called a "crowbar circuit". The purpose of this is to "crowbar" the PSU into shutdown if the draw on any of the voltages is abnormally high, as in the case of a dead short or high resistance short to ground, to keep the supply from going nova.

This is normal protective operation for a switching supply and reacts much more quickly than a fuse could. Waiting for the fuse reaction time would have you digging shrapnel out of walls, ceiling and body.

Allen
August 20th, 2008, 10:32 PM
I thank everyone for the help and useful information--I learned quite a lot. I don't want to spend any more time and money trying to restore the system, so I will give away the PS/2 50Z's parts and options to anyone who would like them...

I originally wanted to give up on this IBM PS/2 Model 50Z, but I had a change of heart. As I was disassembling it to give some of the parts away (in early May), I just felt it needed a second chance at life. At the very least it needed a new system board and diskette drive, but I wasn't sure if the hard drive was good or not. It would not have been worth spending any more money and time on it if the hard disk was bad, but I decided to take a chance.

I acquired a working system board back in May or June and recently found a working floppy drive. As a result, I now have a working IBM PS/2 Model 50Z with the original hard disk functioning and booting the system. I was able to retrieve the files for the original owner.

I acknowledge the encouragement and help from those of you who replied, including the offline encouragement and help from IBMMuseum, johnorun, linemanduke, and Lou - N2MIY. Thanks very much--I couldn't have done it without all of you.

linemanduke
August 23rd, 2008, 08:56 AM
you are very welcome. I'm glad you got a new system and was able to get the stuff off your hard drive. what did it boot to os2, windows, or just dos ?

Allen
August 23rd, 2008, 11:13 AM
The hard disk still has the original, preloaded software: IBM DOS 4.0, Microsoft Windows 3.0, and other applications including Microsoft Excel and Word. I'm lucky that the hard drive did not need replacing, unlike the diskette drive and system board.

I've used or worked on the XT, AT, PS/1, PS/Valuepoint, Aptiva, and newer PCs from IBM, but this was the first time I've ever worked closely with a PS/2. I'm glad I had this chance to do so.

Druid6900
August 23rd, 2008, 12:11 PM
Yeah, I have a couple of Aptiva towers, including the one with the sliding panel, and, although it came complete with EVERYTHING that come with it, including the original CDs and restore CD, from the original owner, I haven't put either of my website because it's doubtful that anyone would want to pay the shipping charges on the damn thing.