Oh My God. My 5150...is dead :(

[crazypigeon]

So, today, I got one of the original PCs, an IBM 5150, complete with a ton of software and all of the manuals...for free.

The guy said that last time he used it, it worked fine, but, he hasn't turned it on since 1990, so, he's not sure if it still actually works.

I've been tinkering with it for an hour or so, and, it dosn't seem to work. I looked at the manuals, and it seemed that the PSU wasn't working, as, I can't get it to put out voltage across pins 1 and 5. I looked on ebay, and the cheapest psu for the 5150 was $20. I figured I'd get it, but, try some more tinkering before hitting buy.

Well, I remembered that I had a psu from a 386 in my closet. I checked the voltage across pins 1 and 5, 5.25 volts, perfect. NOT.

I plugged it in to the mobo, nothing happened for a couple of seconds, then POP, the capacitator at C6 EXPLODED, hitting me, and leaving a burning ember of what once was a capacitor burning on the mobo. I quickly unplugged the psu, and realized what I had just done. I just ruined a IBM classic, I want to die.

But once I got past that, I examined tho mobo, and it seems the only thing visibly wrong with it is capacitor C6. I figured I'd ask one of you guys that have a ibm 5150 with the 16k-64k mobo, could tell me the details on the capacitor at C6, so that I could consider replacing it with a new one when I get a chance, or even sell me a 16-64k mobo so that I can have an original mobo, as the only one on ebay is a 64k-256 mobo

[modem7]

If you search these forums for the word "tantalum", you'll see a few posts where tantalum capacitors have blown, either on old motherboards or other old equipment such as floppy drives.

On the 16-64KB board, I count six capacitors designated C6. They are identical: 10uF/16V tantalum.
The combination of C6 capacitors filter (cleans up) the +12V line that runs from the PSU to the expansion slots. The motherboard will operate without them. In your case only one is dead which is less of a problem.

So why did C6 blow? Probably old age.

The blown C6 may have gone short circuit in which case it will be preventing the PSU from working properly. Even if the C6 has gone open circuit, it's best to remove it.

With the blown C6 removed, if other C6's immediately blow on power up, check the +12V. The 5150 expects an approximate range of +11.5 to +12.5 (source: 5150 Tech Ref). If other C6's are blowing, a severe overvoltage is happening on the +12V line (in which case I would have expected a decent PSU to shut itself off).

Note that the +5V line out of a 5150 PSU can range from approx. +4.75 to +5.25 (source: 5150 Tech Ref)

A related thread on this forum (concerning an XT motherboard though) is: http://www.vintage-computer.com/vcforum/showthread.php?t=5323

[tezza]

I plugged it in to the mobo, nothing happened for a couple of seconds, then POP, the capacitator at C6 EXPLODED, hitting me, and leaving a burning ember of what once was a capacitor burning on the mobo. I quickly unplugged the psu, and realized what I had just done. I just ruined a IBM classic, I want to die.

But once I got past that, I examined tho mobo, and it seems the only thing visibly wrong with it is capacitor C6. I figured I'd ask one of you guys that have a ibm 5150 with the 16k-64k mobo, could tell me the details on the capacitor at C6, so that I could consider replacing it with a new one when I get a chance, or even sell me a 16-64k mobo so that I can have an original mobo, as the only one on ebay is a 64k-256 mobo

Yea, don't be too discouraged. Capacitors commonly explode if the units haven't had juice through them for a while. Providing they don't take anything else with them when they burst, a replacement may be all that's needed.

This has happened to me three times. The first time was with my beloved Osborne when I first switched it on. I also wanted to die. However, the capacitors were easily replaced. (it then developed another fault which is now fixed, but that's another story).

In fact, when I boot up an old computer now I almost expect it to happen. I find myself stepping back a few paces and placing my hand near the wall socket switch for an emergency shutdown when I hear the "pop". :)

[JDT]

I put a hard drive in to my 5150, unfortunatly the 63? watt power supply was not pumping out enough juice to spin the drive up, so I gutted the power supply chassis and stuffed the guts of another AT power supply in the casing of the original IBM one, had to use a little "creativity" but its worked fine like that for almost a year now!

I also had a cap blow up at me, one on my sixpakplus card, may have to look in to getting that replaced.

[Yzzerdd]

Oh my God! You killed a 16-64K mobo? Those are so rare! Well, I hope you get it fixed. I've never(NEVER) had a cap blow up on me yet. But I usually go for vintage PCs that are in prisine condition from their original owners, unless it is cheap or free. I know my day will come, and I will crap myself(out of fear) when it happens...LOL

--Ryan

[per]

Oh my God! You killed a 16-64K mobo? Those are so rare! Well, I hope you get it fixed. I've never(NEVER) had a cap blow up on me yet. But I usually go for vintage PCs that are in prisine condition from their original owners, unless it is cheap or free. I know my day will come, and I will crap myself(out of fear) when it happens...LOL

--Ryan

I'm in the same situation as you, but I haven't got/used too many vintage computes either. The oldest one I have is an XT and the second oldest is a NES (PAL).

[tezza]

....I've never(NEVER) had a cap blow up on me yet. But I usually go for vintage PCs that are in prisine condition from their original owners, unless it is cheap or free.

It's more the length of time they have been shut down more than how well they have been cared for. Hence you can still get mint-looking units blow a cap if suddenly the juice flows after decades of inactivity.

Tez

[crazypigeon]

If you search these forums for the word "tantalum", you'll see a few posts where tantalum capacitors have blown, either on old motherboards or other old equipment such as floppy drives.

On the 16-64KB board, I count six capacitors designated C6. They are identical: 10uF/16V tantalum.
The combination of C6 capacitors filter (cleans up) the +12V line that runs from the PSU to the expansion slots. The motherboard will operate without them. In your case only one is dead which is less of a problem.

So why did C6 blow? Probably old age.

The blown C6 may have gone short circuit in which case it will be preventing the PSU from working properly. Even if the C6 has gone open circuit, it's best to remove it.

With the blown C6 removed, if other C6's immediately blow on power up, check the +12V. The 5150 expects an approximate range of +11.5 to +12.5 (source: 5150 Tech Ref). If other C6's are blowing, a severe overvoltage is happening on the +12V line (in which case I would have expected a decent PSU to shut itself off).

Note that the +5V line out of a 5150 PSU can range from approx. +4.75 to +5.25 (source: 5150 Tech Ref)

A related thread on this forum (concerning an XT motherboard though) is: http://www.vintage-computer.com/vcforum/showthread.php?t=5323

Thanks so much for the information, and the reassurance. I know longer am pacing around my room in disbelief that I have broken it.

To the person that used an AT power supply in theirs, how did you power the monitor? The monitor just uses one if those extention 120volt cables, but, I'm not sure if it takes 120 in, and, I really don't want to actually break it.

Thanks again everyone.

[fred333]

Not worry everything will work out.

[JDT]

To the person that used an AT power supply in theirs, how did you power the monitor? The monitor just uses one if those extention 120volt cables, but, I'm not sure if it takes 120 in, and, I really don't want to actually break it.

I actually disassembled the power supplies from thier housing and put the stronger (250 or 300watt) guts in to the original power supply's housing. This was last year some time.. im not 100% sure, but I think* i just spliced wires from the input to the monitor.. but that doesnt sound right. I could probably take a picture if you need me to, mind you its a PITA to get the PSU out and then take the cover off, after I dig it out of my closet... but if you need me to, I will... I remember the process being pretty straight forward. Had to do some "clever" engineering though...

[flavio75]

I plugged it in to the mobo, nothing happened for a couple of seconds, then POP, the capacitator at C6 EXPLODED, hitting me, and leaving a burning ember of what once was a capacitor burning on the mobo. I quickly unplugged the psu, and realized what I had just done. I just ruined a IBM classic, I want to die.


I can't believe what I'm reading... this EXACTLY the same thing that happened to my 5160 motherboard some time ago. Is the C6 a three-legs capacitor or a two-legs one ? If it is a three-legs capacitor, I bought a few replacement caps here: www.mouser.com.

It was some time ago, though... so I do not know if they still have them. I haven't had time to remove the blown cap from my motherboard and put the new one in yet, so I do not know if the repair would be successful or not.

My cap blew up following exactly the same procedure you described. The PC was not starting up, and I thought the PSU was the problem since there was a weird noise coming from it. So I used an AT power supply in its place. When I switched it on, the cap blew up after a couple of seconds. I had disconnected everything else fortunately, only the video card was installed, no drives or anything else. The weird thing is that after the cap exploded, I reconnected the original XT PSU and my motherboard did boot up! The only problem is that it doesn't seem to recognize the hard drive anymore. The hard drive certainly is not damaged because it was totally disconnected when the cap blew. The controller was not installed either. The computer now works (and the blown cap is STILL there!) but I have to start it up using a floppy disk... anyone has any idea if it's possible to get the hard drive working again ?

[tezza]

The computer now works (and the blown cap is STILL there!) but I have to start it up using a floppy disk... anyone has any idea if it's possible to get the hard drive working again ?

Well, if the hard drive stopped working after the cap was blown, I'd look at replacing the cap first off. It must have served some purpose and perhaps it had a role in the hard disk circuitry.

Tez

[JohnElliott]

This happened to my IBM 5271 last week as well. Power supply dead, motherboard dead (when I tried powering the motherboard from an AT power supply I got a brief display of actual flames from the dead capacitor). I've checked the hard drive and the 3270 emulation cards in other XTs and they seem to have survived. It was quite amusing to see a Sinclair PC200 displaying its boot messages on a 3270PCDA rather than its built-in CGA.

Now I'm wondering what to do with the bits - in particular whether it's worth trying to find a replacement IBM XT motherboard and PSU, and reconstruct the thing. I don't really have the skills to take a soldering iron to the thing myself. Fitting a baby-AT motherboard isn't really practical, because the 3270PCDA needs three adjacent 8-bit slots for the full effect.

[Druid6900]

Probably cost you less to have the two items repaired rather than replace them. The PSU probably just needs the fuse replaced and the cap is a 3 minute replacement job.

On the one where the HD doesn't work, tantalum caps, when they go, usually fuse the tantalum oxide innards into a conductive path between the two legs (ie: high resistance short) and, depending on what the particular cap is being used for, this may not be too good.

Try clipping the cap out and see if the HD works.

[nige the hippy]

"cor blimey what a fuss!"

Tantalum capacitors pop, end of story.
snip it out, and if the problem goes away great, if it doesn't, there's probably another one somewhere that needs snipping.

get your meter out, look for a power supply line that's low, and then trace it to the point of lowest resistance (to ground), then snip again!

Sooner or later you're going to find the problem.

As an illustration, the MZ80K I've got on the bench has had 3 faults already, 1 capacitor duff, 1 TTL IC duff, and I'm just trying to get the memory sorted.

You've just got to hold your nose & jump in!

[JohnElliott]

So far, so good; without the blown capacitor, the motherboard boots, which it didn't before. Now all I need to do is find a replacement, and I see the question of whether and how to replace C56 has come up before (http://www.vintage-computer.com/vcforum/showthread.php?t=5323).

As for the power supply, I'll have to see where I can get hold of something that can undo security Torx bolts.

[Druid6900]

As for the power supply, I'll have to see where I can get hold of something that can undo security Torx bolts.

Channel-locks, then replace them with standard phillips head bolts.

[JohnElliott]

OK. For future reference, a normal (16v, 10uF) two-leg capacitor seems to work fine as a replacement for C56; the positive leg goes in the middle hole, and the other one I put in the hole nearest the edge of the planar (though I think either should do).

Not sure what to do about the power supply, though. The fuse is fine. The first time I switch on, the fan turns about 1/8 of a revolution before cutting out; on subsequent attempts nothing happens except an ominous click from somewhere near the switch.

(this is without any load, but that shouldn't matter... should it? I'm reluctant to hook it up to the motherboard in case it was the PSU killed C56 rather than the other way round).

[Druid6900]

Did you measure the fuse with an ohm-meter? Sometimes, the little buggers break at the ends, under the metal caps where you can't see it.

If it is good, you really don't want to try and repair a switching power supply and you're probably better off slapping the case around a working supply of the same dimensions as, I believe, someone else mentioned.

[JohnElliott]

Did you measure the fuse with an ohm-meter?

Yes. I'll gut the PSU as suggested.

[tezza]

Not sure what to do about the power supply, though. The fuse is fine. The first time I switch on, the fan turns about 1/8 of a revolution before cutting out; on subsequent attempts nothing happens except an ominous click from somewhere near the switch.

Some modern power supplies DO require a load. So maybe it does? I used a modern form-factor power supply once to supply power to an external floppy drive. At first the powerpack didn't work except for a slight turn of the fan. I thought it was broken, so I tried again with another one. Same result.

After some web searching I found a discussion which talked about newer power supplies requiring to be plugged into a motherboard in order to activate them (presumably by supplying a load). The advice was that if you wanted to use it just for a periferal device you short out two of the pins on the motherboard connector, (I think it was any black to green). The PSU thinks there is a motherboard attached and activates.

Anyway, I tried this with a paper clip and it worked. The PSU sprang into life and functioned without a problem.

Now please don't follow this advice without doing some more web checking or verifying it somewhere. As far as electronics go, I'm just a very short step up from illiterate. It worked for me with a MODERN power supply, but you might have an entirely different problem.

Tez

[modem7]

Yes, most switch mode power supplies (modern or old) require some form of load in order to start up.

Regarding the 5160, the thread http://www.vintage-computer.com/vcforum/showthread.php?t=8636 is where I calculated the individual power consumption of devices in my 5160, and determined that for the genuine IBM 130W PSU, a device that draws at least somewhere between 6W and 15W (exactly what is unknown to me) is required to enable the PSU to start.

[modem7]

After some web searching I found a discussion which talked about newer power supplies requiring to be plugged into a motherboard in order to activate them (presumably by supplying a load). The advice was that if you wanted to use it just for a periferal device you short out two of the pins on the motherboard connector, (I think it was any black to green). The PSU thinks there is a motherboard attached and activates.
PSUs for the 5150/5160/5170 are switched on/off only by the power switch on the PSU.

In modern PCs, the PSU is switched on/off by a signal from the motherboard, and that signal was probably what you were simulating with the paper clip. If one looks at the pinout for ATX type PSU's (eg. http://pinouts.ru/Power/atxpower_pinout.shtml), by shorting green and black, you were doing just that. Even so, the PSU would still have probably required an adequate level of load to actually start up.

[JohnElliott]

Well, that saved a certain amount of time and trouble. The PSU works with a load (I used a 486 motherboard) so I don't need to gut it. Thanks to everyone on this thread for their help.

[vwestlife]

I plugged it in to the mobo, nothing happened for a couple of seconds, then POP, the capacitator at C6 EXPLODED, hitting me, and leaving a burning ember of what once was a capacitor burning on the mobo. I quickly unplugged the psu, and realized what I had just done.
I just had a similar episode upon pulling my own PC out of storage today. Capacitor C9 on the original IBM CGA board failed upon power-up, letting out a fair amount of smoke and leaving a nice big black burn mark on the inside of the case cover, as well as on the CGA board itself. Thankfully the rest of the system, including the motherboard and other expansion boards, weren't damaged.

So, does anyone know the correct value of C9? The original one is three-pin. I'd rather just get an aftermarket "snow-free" CGA card, but if nothing else on the IBM CGA board got damaged, then it'll be an easy fix.

[Mike Chambers]

that's harsh.. ugh. it's not your fault though. any AT power supply even one you go buy brand new today would be just fine to power an XT motherboard. nothing changed PSU voltage and pinout wise between the very first XT board made and the very last AT board made.

i can't really blame it on it being old either. it all depends how it was taken care of. my mustang has 140k miles and has been taken care of meticulously... my dad's has 115k and it barely works lol. burns more oil than gas. you can literally watch it blow out of the back in the rear view.

if that 5150 was somewhere that sees some moisture, that's the problem. what you should probably consider is getting a nice cheap XT clone motherboard and sticking it in there, unless you really do care whether or not it's a true original IBM. i'd sell you a spare "TURBO-XT" board i have, but something tells me i should hang on to it... you never know what could happen to the other board i have.

[dongfeng]

I've just had the same things happen on a couple of 5150 motherboards. One literally spat fire and another popped with an almighty bang. They have since been replaced and the boards are working. Just dodgy memory to sort out...

[modem7]

So, does anyone know the correct value of C9? The original one is three-pin.
On my IBM CGA board, C9 is the same as C1, C2 and C3 (all .047) That matches the schematic diagram I have. The caps have two leads.

[modem7]

nothing changed PSU voltage and pinout wise between the very first XT board made and the very last AT board made.
There was a subtle difference. In the move the AT, an unused pin in one of the power connector plugs/sockets became used for an extra +5V line. Obviously only a possible problem if one tries to run an AT board from a 'true' XT PSU.

[vwestlife]

Oops! I spoke too soon about the rest of my PC. After removing all the expansion boards, I flipped it on and one of the 13 capacitors on the motherboard marked as "C7" shot itself apart like a firecracker, complete with bright yellow flying embers!

These C7's on the 64KB-256KB PC motherboard appear the serve the same purpose as the six C6 capacitors on the 16KB-64KB mobo, as described by modem7 earlier in the thread. In my case, the one nearest U5 (74LS30 chip) blew itself half-apart.

I'm using an aftermarket 200W power supply, as this PC has been ever since I installed a hard drive in it some 15 years ago. I tested the output, and with just the full-height IBM floppy drive and a Seagate ST-225 hard drive connected, I get:

Red: +4.92V
Yellow: +11.15V
Orange: +4.92V
Blue: -11.31V
White: -5.10V

The +/-12V line is a little below spec, but that's certainly better than being above-spec, and the PS continues to perform flawlessly despite the two capacitor blow-ups so far!

I also have an early 16KB-64KB PC with the original black-case 63.5W power supply, but with this experience on my newer PC, I'm now hesitant to just plug it in and fire it up. On that one, I think I'll get a variable power supply and slowly bring each supply voltage up to spec. That gradual ramping-up of voltage is known to help capacitors in old radios "re-form" after years of nonuse. (But yet I recently powered up an even older Commodore VIC-20 after years of storage and it works flawlessly... go figure, IBM quality!?)

[tezza]

Those are scary stories *shudder*

Hmmm...that 5150 that Eric (Frozenfire71i) gave me hasn't been run for a month. I just might fire it up and run it for a few hours tonight while we have dinner.

You guys have got me worried now! :)

[vwestlife]

Regular (but not necessarily constant) use is the key. A car that's driven once a week or even once a month is going to start a lot easier than a car that has been sitting for the past five years... same thing with electronics. But just like a car, once you get it running, it's best to keep it running for a while, until it gets fully warmed up. (Of course, turn down the monitor brightness or use a screen saver to prevent burn-in.) On a PC, keep it cookin' at least until the air coming out of the PS fan is warm to the touch.

[Yzzerdd]

Yup, I run my computer every day. My 5150(B) or whichever is setup(usually it or the AT&T) get to check my mail, word process, do finances, etc. So clearly it is vital that I have one running. OK, I can technically got ALL that on a new computer. At the same time. So what.

Anyhow, I click all computers that are on the desk when I wake up. When I go to bed, they are all turned off. So they have about 16 hours on time per day and 9 hours of being off. Plenty of time to get to a stable temperature in both ways, heat and cooling.

Due to the vitality of their existance in my life, I also need to make sure they are in top shape at all times. Regular cleaning, maintenance(drive cleaning), etc are all part of it's life. Usually once a month it gets a tune-up. Of course, I follow alot of the general guidelines out there to prolong computer life. I hear it is best to run the PC constantly unless it will be off for 2 or more days, and even though it makes sense in the form of the electricity jolts when turned on/off as well as the uneven heating(until stable temp in a few hours) and uneven cooling(the very same). Although making sense, it makes no sense in that it wastes electricity and is hard on the bearings in the fan and if applicable, HDD.

Which is why I prefer external HDDs. You can turn them off and leave the system on. Wait. That wasn't the point at all.

--Ryan

[Druid6900]

I must not live right.

I turn on stuff that hasn't been fired up in twenty years and nothing explodes. Superior electrolytics, I suppose. And I KNOW that this stuff hasn't been powered on because it's been sitting in my basement all that time.

No exploding chips, no flames, no snowstorms. Possibly the occasional blown fuse, but, it's replaced and away it goes.

Maybe if I throw a handful of change into some of them before I turn them on.....

[vwestlife]

I must not live right.

I turn on stuff that hasn't been fired up in twenty years and nothing explodes. Superior electrolytics, I suppose. And I KNOW that this stuff hasn't been powered on because it's been sitting in my basement all that time.
Who knows, maybe IBM cheapened out and only specified their capacitors to last 20 years, instead of forever. ;) It's nothing new. The IT guys where I work just had to replace a whole bunch of Dell motherboards due to bad capacitors. It was all covered under warranty, but still a big hassle.

And in this case, it's actually the tantalum capacitors that are failing, not electrolyics. There are no electrolytic capacitors anywhere on a PC motherboard.

[Yzzerdd]

I turn on stuff that hasn't been fired up in twenty years and nothing explodes. Superior electrolytics, I suppose. And I KNOW that this stuff hasn't been powered on because it's been sitting in my basement all that time.

No exploding chips, no flames, no snowstorms. Possibly the occasional blown fuse, but, it's replaced and away it goes.

LOL, no snowstorms, eh?

I've never had any of that happen, either. Hmm... No blown fuses either. Well I have good luck too, I suppose. That, and I only buy 1 or 2 owner systems, unless the price is too good. System is all rusty and looks like it has been in an unfit enviroment for a few years? No thanks, I'll pass.

Sounds like bad luck you've been having there, VWestlife. That system sounds like alot of trouble. Have you tried calling an exorcist for that system? It sounds like a ream demon. I guess with some work, it could be a diamond!

Oh me and my lame jokes....

--Ryan

[Druid6900]

And in this case, it's actually the tantalum capacitors that are failing, not electrolyics. There are no electrolytic capacitors anywhere on a PC motherboard.

Yes, I was thinking more of the PSU, where there are some nice, big ones :)

[modem7]

I can't believe what I'm reading... this EXACTLY the same thing that happened to my 5160 motherboard some time ago. Is the C6 a three-legs capacitor or a two-legs one ? If it is a three-legs capacitor, I bought a few replacement caps here: www.mouser.com. It was some time ago, though... so I do not know if they still have them.
Yes they do - I recently bought and received some.
If others are seeking these 10uF/16V tripple-lead tantalums, the Mouser details for the part are:

Mouser part number: 80-T398E106M16AT
Manufacturer part number: T398E106M016AT
10/16V TRIPPLE LEAD
Kemet Dipped Radial Tantalum Capacitor

[paul]

I've also had the orange tantalum cap fail on an IBM CGA card. Top corner near the bracket screw, sits across the 5V bus. Mine simply shorted while in storage and so the PS would not power-up, leading me to be quite concerned until I unplugged the card. Just clipped it out for now.

But having two 64-256 5150's that have been sitting around the house for more than two years without being powered-up does concern me a bit, and I think it would be wise to replace the few present. I don't think the short itself is a big deal but even minor physical damage to the machine would bug me.

It would not hurt us to compile a list of preventative maintenance replacement parts (and locations) for these machines, including electrolytics in the PS.