PDA

View Full Version : 5150 not booting. Please help?



Fire-Flare
April 30th, 2010, 07:54 PM
I tried to start it up this morning, and it threw a fit of long beeps.

I've removed the drives and all the cards except for video. And now it's not beeping at all. It just repeats the video card's start-up screen.

The hardware looks and smells fine, switches are as follows for 0 disk drives, 256k onboard memory, and VGA video card:

Block 1: On-off-off-off-on-on-on-on

Block 2: On-off-off-on-on-off-off-off

Tetrium
April 30th, 2010, 08:10 PM
You mean it was working fine before and then the very next time you tried to fire it up, it did what you just described?

Fire-Flare
April 30th, 2010, 08:22 PM
It was stored in another room for about half a year.

modem7
April 30th, 2010, 10:57 PM
The VGA card is a possible problem. Do you have any other card that you could use instead?

If not, try the following. It will help us to determine the state of the motherboard.

Remove the VGA card, and leave the all switch settings as they are. so all there should be connected to the motherboard is the power supply and the speaker.

Power up. After about 45 to 50 seconds, there should be a single beep. That beep indicates that the POST has successfully finished. Do you hear the beep?

If so, that gives us confidence that the motherboard is good. It doesn't prove that the motherboard is good though (the POST is a crude test only).

If you didn't hear the beep, we can be very confident that something is wrong with the motherboard or PSU. Why the PSU? Some of the clone power supplies send a fake POWER GOOD signal to the motherboard (e.g. they generate POWER GOOD even if the 5 volt line is sitting at 4 volts). If your 5150 has the original IBM power supply, that would tend to lay fault at the motherboard.

Tetrium
May 1st, 2010, 12:18 AM
It was stored in another room for about half a year.

Relocating a computer sometimes causes expansion cards to unseat a lil bit, causing a computer to seemingly fail without reason. My guess was also the PSU btw, but frankly my knowledge of old IBM's is quite limited. I do understand basic hardware troubleshooting though, I've troubleshooted a couple hundred computers in my life so I consider myself savvy in that respect.
I hope you get her working again :)

Fire-Flare
May 1st, 2010, 06:32 AM
The VGA card is a possible problem. Do you have any other card that you could use instead?

If not, try the following. It will help us to determine the state of the motherboard.

Remove the VGA card, and leave the all switch settings as they are. so all there should be connected to the motherboard is the power supply and the speaker.

Power up. After about 45 to 50 seconds, there should be a single beep. That beep indicates that the POST has successfully finished. Do you hear the beep?

If so, that gives us confidence that the motherboard is good. It doesn't prove that the motherboard is good though (the POST is a crude test only).

If you didn't hear the beep, we can be very confident that something is wrong with the motherboard or PSU. Why the PSU? Some of the clone power supplies send a fake POWER GOOD signal to the motherboard (e.g. they generate POWER GOOD even if the 5 volt line is sitting at 4 volts). If your 5150 has the original IBM power supply, that would tend to lay fault at the motherboard.

Removing the video card didn't help, neither did reseating the power supply connectors.

The PSU from an AT clone, it's made by Enhance. Are those known to give problems? (or known at all?)

Raven
May 1st, 2010, 06:44 AM
Flip the first switch in the switch bank once, then back to re-seat the switch, so-to-speak. I found that the DIP switches can become wonky if the machine is moved about (mine was transported and the floppy switches got weird). It sounds like yours is in POST-loop, which is controlled by the first switch (if it's the same as the 5160, and I assume that is).

Fire-Flare
May 1st, 2010, 07:11 AM
I switched all of them back and forth, no change but it seems to be looping. The screen was resetting and I can hear a faint click at about the same interval.

What else would cause this?

Chuck(G)
May 1st, 2010, 08:36 AM
Got a meter? I'd start troubleshooting by checking the voltages at the motherboard--in particular, see if things change when the board resets itself..

vwestlife
May 1st, 2010, 08:42 AM
Perhaps a bad RAM chip somewhere is causing the system to restart when it's tested. Try setting the DIP switches to 16K RAM (SW2 1-5 on, 6-8 off), so that the minimum amount of RAM will be recognized and tested. The PC should now almost instantaneously beep and boot up to Cassette BASIC (if you have it set to 0 floppy drives) when turned on.


Flip the first switch in the switch bank once, then back to re-seat the switch, so-to-speak. I found that the DIP switches can become wonky if the machine is moved about (mine was transported and the floppy switches got weird). It sounds like yours is in POST-loop, which is controlled by the first switch (if it's the same as the 5160, and I assume that is).

On the 5150 PC, the first DIP switch controls whether it will boot directly to Cassette BASIC (on) or will try to boot from the floppy drive first (off).

On the XT, the first DIP switch was changed so that (off) is "normal operation," since it is assumed that an XT will always have at least one floppy drive installed, and (on) now puts the self-test into an endless loop.

Source: http://www.rci.rutgers.edu/~preid/pcxtsw.htm

Chuck(G)
May 1st, 2010, 08:52 AM
If you can burn your own EPROMs (remember that the 5150 takes 24-pin 8Kx8 ROMs), I can offer you a diagnostic ROM image.

TNC
May 1st, 2010, 09:50 AM
The PSU from an AT clone, it's made by Enhance. Are those known to give problems? (or known at all?)

Didn't the AT- class machines have another pinout for the power connector?

Chuck(G)
May 1st, 2010, 10:15 AM
Didn't the AT- class machines have another pinout for the power connector?

No, it's the same, but in particular, pay attention to the Power Good signal--if it's not stable, the system will keep rebooting. This just sounds so much like a power supply issue.

The 5150 BIOS (IIRC) tests enough memory to get the BIOS and display setup, and then resume testing the rest. If that first block doesn't pass the test, the system just halts with no indication.

Raven
May 1st, 2010, 10:16 AM
Perhaps a bad RAM chip somewhere is causing the system to restart when it's tested. Try setting the DIP switches to 16K RAM (SW2 1-5 on, 6-8 off), so that the minimum amount of RAM will be recognized and tested. The PC should now almost instantaneously beep and boot up to Cassette BASIC (if you have it set to 0 floppy drives) when turned on.



On the 5150 PC, the first DIP switch controls whether it will boot directly to Cassette BASIC (on) or will try to boot from the floppy drive first (off).

On the XT, the first DIP switch was changed so that (off) is "normal operation," since it is assumed that an XT will always have at least one floppy drive installed, and (on) now puts the self-test into an endless loop.

Source: http://www.rci.rutgers.edu/~preid/pcxtsw.htm



Ah, forgive my lack of knowledge about the 5150 then.

Fire-Flare
May 1st, 2010, 11:34 AM
No, it's the same, but in particular, pay attention to the Power Good signal--if it's not stable, the system will keep rebooting. This just sounds so much like a power supply issue.

Okay then I'll start with that. What tool and settings should I use?

framer
May 1st, 2010, 12:39 PM
I'll second the PS as the first place to look. It really helps if you can get another PS to try. I rebuilt a PC PS with the innards of a mini AT PS. It's a quick cheap solution if thats found to be the problem.

framer

Chuck(G)
May 1st, 2010, 01:34 PM
A digital multimeter is the tool for this. If you don't have one, you can get an adequate cheapie from your local Harbor Freight.

Fire-Flare
May 1st, 2010, 01:37 PM
I doubt that any of the shops in my area have one. I'll put an ad on Craigslist asking to borrow one.

Fire-Flare
May 1st, 2010, 01:39 PM
A digital multimeter is the tool for this. If you don't have one, you can get an adequate cheapie from your local Harbor Freight.

I know a guy who will let me borrow his. Which wires should be positive and negative?

Chuck(G)
May 1st, 2010, 01:45 PM
I've seen these things at my local hardware store, as well as a couple of big-box stores. The last one cost me $2.99:

http://images.harborfreight.com/cpi/photos/90800-90899/90899.gif

modem7
May 1st, 2010, 02:22 PM
Didn't the AT- class machines have another pinout for the power connector?
Technically yes. In the move to the AT (5170), IBM used an unused pin (pin 2 of P8 - the pin next to POWER GOOD) to supply an additional +5 volts line.

However, fitting an IBM AT power supply to a IBM PC or IBM XT isn't an issue because on the IBM PC and IBM XT motherboards, pin 2 of the corresponding socket isn't connected to anything.

modem7
May 1st, 2010, 07:57 PM
I know a guy who will let me borrow his. Which wires should be positive and negative?
Refer to the diagram at http://members.dodo.com.au/~slappanel555/misc/IBM_5150_PSU_pinout.jpg (from 5150 technical reference).

1. Keep motherboard connected to power supply (so power supply is loaded).
2. Set multimeter to DCV (DC volts).
3. Negative lead of multimeter (black lead) to the metal case of the power supply (good enough reference point in this situation).
4. Move positive lead (red lead) about as appropriate to measure voltage.

As for the POWER GOOD line (wire colour usually orange). That should be sitting at approximately +5V. If that line is dipping periodically, each dip would cause your motherboard to reset, I don't know if a digital multimeter will be fast enough to pick any dips. It will probably depend on how long the dip is.