• Please review our updated Terms and Rules here

IBM 5155 Portable PC

Besides plus and minus 12V, P8 also supplies the POWER GOOD signal from the PSU. Without the POWER GOOD signal, the motherboard won't start. Obviously, because your 5155 is now booting to BASIC, you must have reconnected P8.
Yes, it it quite okay to have P8 plugged in without C56 being present. Ideally, you should fit a new C56 later.

Hmmm... I only have P9 plugged in and it's booting to BASIC. I'm still getting the same problem with the power when I plug the P8 connector in. I also reconnected the floppy power connectors back in and the floppy drives flashed at me and appeared to be searching for a disk. I was curious about not having the P8 connector plugged in and am scratching my head now about it booting to BASIC without it being plugged in.
 
I'm still getting the same problem with the power when I plug the P8 connector in.
So did you remove C56 ?
If not, that symptom is what we expect until you either remove C56 or fit a new C56.
If you did remove C56, then there must be another short, on either the +12V line or the -12V line.

Hmmm... I only have P9 plugged in and it's booting to BASIC.
That is not the norm for a 5155/5160 motherboard. The POWER GOOD signal from the PSU (activated when all voltages are present and stable) is what releases the motherboard from a reset state. Either someone has modified your IBM motherboard, or it has a second fault, one associated with the POWER GOOD signal.
Is it of concern? It might be if the motherboard is starting before the +5V has stabilised. If you later see intermittent starting problems, this POWER GOOD issue may be the cause.
 
So did you remove C56 ?
If not, that symptom is what we expect until you either remove C56 or fit a new C56.
If you did remove C56, then there must be another short, on either the +12V line or the -12V line.
Yes, I did remove C56. I did not remove the motherboard, though. I just clipped the top wires off of C56 so the solder is still probably on the bottom of the motherboard if that matters at all.

That is not the norm for a 5155/5160 motherboard. The POWER GOOD signal from the PSU (activated when all voltages are present and stable) is what releases the motherboard from a reset state. Either someone has modified your IBM motherboard, or it has a second fault, one associated with the POWER GOOD signal.
Is it of concern? It might be if the motherboard is starting before the +5V has stabilised. If you later see intermittent starting problems, this POWER GOOD issue may be the cause.

I don't think it's really of concern since it appears to be booting fine without the P8 connector being plugged in. I'll have to power the computer off and on a few times to see if there's intermittent starting problems. The last 3-4 times it started just fine, but I did have one occasion where the screen was just flickering and did not boot to BASIC. That was before I removed C56. All of the dates that I've seen on components are from 1984 (the PSU has a test date on it) so it would be hard to say with my knowledge of older PC's whether the motherboard is a standard 5155/5160 board. Really, I'm happy to get it to turn on at all.
 
Yes, I did remove C56. I did not remove the motherboard, though. I just clipped the top wires off of C56 so the solder is still probably on the bottom of the motherboard if that matters at all.
The solder in the holes won't be a problem. There is a possibility that C58 also went short. Do you have a multimeter? If so, you can use that to determine which line has the short. I'm sure that you don't want to cut off C58 unnecessarily.

I don't think it's really of concern since it appears to be booting fine without the P8 connector being plugged in.
The plus and minus 12V from P8 are fed by the 5155/5160 motherboard through to the expansion connectors, for possible use by an expansion card.

All of the dates that I've seen on components are from 1984 (the PSU has a test date on it) so it would be hard to say with my knowledge of older PC's whether the motherboard is a standard 5155/5160 board.
I'm using the term "5150/5160 motherboard". A 5155 motherboard is actually an early 5160 motherboard. One is pictured at http://www.nadbor.pwr.wroc.pl/yesterpc/Hardware/ISA motherboard/IBM XT/item.htm
 
It has occured to me that you have plugged at least one expansion card back into the motherboard. The second short could be on an expansion card. Remove all expansion cards and then see in plugging in P8 stops the PSU from running.
 
It has occured to me that you have plugged at least one expansion card back into the motherboard. The second short could be on an expansion card. Remove all expansion cards and then see in plugging in P8 stops the PSU from running.
Absolutely right. I've seen TWO seperate CGA cards(like the one in the 5155) with shorted Tantalums at C8. That's a three legged capacitor on the +12v line. If you check for a short between the middle leg of C8 and ground, it'll tell you if it's ok. If it shows short, it could also be C9 which is right below C8.
 
I've had several shorted tantalum capacitors recently on motherboards and expansion cards.
Most recently on a Tecmar Captain card. Replaced the capacitor and it works fine.

One on a 386 motherboard ..... lit up like a bulb when I applied power. I have a couple of scrap
motherboards that I steal replacements from.
 
It has occured to me that you have plugged at least one expansion card back into the motherboard. The second short could be on an expansion card. Remove all expansion cards and then see in plugging in P8 stops the PSU from running.

I removed all of the expansion cards this morning. I turned the computer on after each card was taken out and still had the same power problem if P8 was plugged in. I do have a multimeter; however, (you can tell I've never used one before) it belongs to my father-in-law and it's an older analog one so I'm not sure if I tested the voltage correctly.

According to the technical reference for the PSU about the P8 connector, it looks like the pins I need to test are 3(+12vdc) and 4(-12vdc). I put the black probe in pin6 for the ground pin and then used the red probe to test the other pins. Pin 3 looks like it's outputting around 10v and then I don't get anything when I test pin 4. So either the -12v is the problem or I'm doing something wrong with the multimeter. I wasn't sure how to check the resistance on this multimeter.
 
Last edited:
I removed all of the expansion cards this morning. I turned the computer on after each card was taken out and still had the same power problem if P8 was plugged in.
So the short is definitely on the motherboard.

I do have a multimeter; however, (you can tell I've never used one before) it belongs to my father-in-law and it's an older analog one so I'm not sure if I tested the voltage correctly.
According to the technical reference for the PSU about the P8 connector, it looks like the pins I need to test are 3(+12vdc) and 4(-12vdc). I put the black probe in pin6 for the ground pin and then used the red probe to test the other pins. Pin 3 looks like it's outputting around 10v and then I don't get anything when I test pin 4. So either the -12v is the problem or I'm doing something wrong with the multimeter. I wasn't sure how to check the resistance on this multimeter.
As I posted before, four capacitors are involved:

+12V line: Capacitors C56 (10 uF tantalum) and C55.
-12V line: Capacitors C58 (10 uF tantalum) and C57.

C55/C57 are of a very reliable type. C56/C58 are of type tantalum which fails relatively regularly in vintage computers. You have removed physically damaged C56. The voltage check you done just now suggests a short on the -12V line (a resistance check will prove it). Since you've already cut off C56, I suggest that you take a chance and do the same now for C58. Or do you want to do a resistance check first?
 
As I posted before, four capacitors are involved:

+12V line: Capacitors C56 (10 uF tantalum) and C55.
-12V line: Capacitors C58 (10 uF tantalum) and C57.

C55/C57 are of a very reliable type. C56/C58 are of type tantalum which fails relatively regularly in vintage computers. You have removed physically damaged C56. The voltage check you done just now suggests a short on the -12V line (a resistance check will prove it). Since you've already cut off C56, I suggest that you take a chance and do the same now for C58. Or do you want to do a resistance check first?

What should I be looking for with the resistance check? I was using this page (http://www.techrepublic.com/article/put-your-power-supply-to-the-test-with-a-multimeter/5034541) as a reference for checking capacitors and according to that it looks like I should be getting a high resistance; however, when I test both the +12 and -12 pins I get 0 resistance.
 
What should I be looking for with the resistance check? I was using this page (http://www.techrepublic.com/article/put-your-power-supply-to-the-test-with-a-multimeter/5034541) as a reference for checking capacitors and according to that it looks like I should be getting a high resistance; however, when I test both the +12 and -12 pins I get 0 resistance.
What should I be looking for with the resistance check? I was using this page (http://www.techrepublic.com/article/put-your-power-supply-to-the-test-with-a-multimeter/5034541) as a reference for checking capacitors and according to that it looks like I should be getting a high resistance; however, when I test both the +12 and -12 pins I get 0 resistance.

I was positive that since you removed C56, that the +12V line was not going to measure zero ohms (or near that).

Take a look at http://www.minuszerodegrees.net/5150_5160/MDC/no_beeps_2_2.htm
In that way you can verify which pins are used for the +12V and -12V resistance measurements.
Are those the pins you are using?

And per the web site you pointed to, did you have your old meter set to a low ohms range, e.g. 0-100 ohms ?

So that you are confident about measuring resistance, with power off, measure the resistance of R1. It is next to slot 8. You should measure about 500 ohms.
 
I was positive that since you removed C56, that the +12V line was not going to measure zero ohms (or near that).

Take a look at http://www.minuszerodegrees.net/5150_5160/MDC/no_beeps_2_2.htm
In that way you can verify which pins are used for the +12V and -12V resistance measurements.
Are those the pins you are using?

And per the web site you pointed to, did you have your old meter set to a low ohms range, e.g. 0-100 ohms ?

So that you are confident about measuring resistance, with power off, measure the resistance of R1. It is next to slot 8. You should measure about 500 ohms.

Ok, I was doing the resistance test wrong. I was putting the probes inside the power connector P8 instead of testing the motherboard. My multimeter only has one OHM setting which is X1k, I'm guessing that means I just multiply whatever I get by 1k. When I test R1 I get 1 on the multimeter so it's putting out 1000 ohms. When I test the +12 line I get 3.5, so 3500 ohms. When I test the -12 line I get .2 on the multimeter which would be 200 ohms; however that's what confused me because my father-in-law said that was a 0 resistance because when I touch the two probes together I get 200 ohms too. So maybe this multimeter is off?
 
however that's what confused me because my father-in-law said that was a 0 resistance because when I touch the two probes together I get 200 ohms too. So maybe this multimeter is off?
There is normally a 'zero adjust' adjustment on the front panel of the analogue meter. When you change resistance ranges, one normally shorts the two probes, then adjusts the 'zero adjust' adjustment so that a reading of zero shows.
The other possibility is that the battery in the meter is low (although I can't remember what symptom that produces).
 
How does clipping C56 and C58 effect the PC? I just got it to boot up now, after clipping C58, with P8 plugged in so that's a positive!
C56 (together with C55) help remove noise on the +12V line.
C58 (together with C57) help remove noise on the -12V line.

The 5155/5160 motherboards do not use the +12V or -12V from the PSU. The +12V and -12V from the PSU is routed via the motherboard to the expansion slots, for possible use by an expansion card.

I myself would not be fussed by leaving C56 and C58 off the board, since I expect that any expansion card that uses either/both +12V or -12V will have its own filter capacitors, rated such that the card does not depend on the motherboard's C56/C58. Other people will have a differing opinion.

But ideally, C56/C58 should be replaced.
 
C56 (together with C55) help remove noise on the +12V line.
C58 (together with C57) help remove noise on the -12V line.

The 5155/5160 motherboards do not use the +12V or -12V from the PSU. The +12V and -12V from the PSU is routed via the motherboard to the expansion slots, for possible use by an expansion card.

I myself would not be fussed by leaving C56 and C58 off the board, since I expect that any expansion card that uses either/both +12V or -12V will have its own filter capacitors, rated such that the card does not depend on the motherboard's C56/C58. Other people will have a differing opinion.

But ideally, C56/C58 should be replaced.

Ok. I won't worry about it for now. I doubt that I'll put any of the expansion slots to use anyway. If I did want to replace C56/58 I'm looking for the Three-legged 10uF/16V replacement referenced http://www.minuszerodegrees.net/failure.htm here?
 
Ok. I won't worry about it for now. I doubt that I'll put any of the expansion slots to use anyway. If I did want to replace C56/58 I'm looking for the Three-legged 10uF/16V replacement referenced http://www.minuszerodegrees.net/failure.htm here?
I have been meaning to revise that web page (post research) because I had noticed that IBM did something odd.
For example, on a 640K 5160 motherboard I have, the connection points for C58 are wired -|+|-, but the connection points for C56 are wired +|-|+
That means that although both C56 and C58 are 10uF/16V, their pinouts are different.

Also, you can use a 2-legged tatalum instead of a 3-legged one. 2-legged ones are easier to source, but they must be inserted in the correct orientation, negative leg into negative hole and positive leg into positive hole.

When you're ready to get replacements, revisit these forums.
 
Back
Top