PC Compatibility Card Issues

[KGLlewellyn]

Hey folks,

I've got a 820-0720-B Pentium PCI Compatibility card, I believe it's the 100Mhz model. I've been using it flawlessly in my Power Mac 4400 via the GIMO slot for the last few months. Out of nowhere today when I went to boot it up it gave me a couple of short beeps and didn't respond. I rebooted the card, and now when the Mac switches to the card I get no beeps and the monitor goes to 'No Signal'. After some finagling, I was able to get it to boot, but the graphics looked a bit corrupted. But now it's back again to no-beep, and PCSetup is stuck on 'PC is Booting'. The card hasn't been recapped or anything, but is anyone aware of any common failures that hit this card?

I'm not really in a position to try this out in another Mac as I lack the Y-Cable, I only have the GIMO cable and adapter sadly.

 

[Byrd]

Is there any additional video RAM installed in the sockets? I'd remove, and while at it clean the RAM, PCI and slot connectors

 

[KGLlewellyn]

Is there any additional video RAM installed in the sockets? I'd remove, and while at it clean the RAM, PCI and slot connectors

No additional video RAM. I’ve De-Oxited the RAM slot and PCI slot to no avail sadly.

 

[KGLlewellyn]

Uhm...so out of nowhere the card decided to start working again?!

Literally I've done nothing. Friday and Saturday, it didn't work. On a whim tried it today and it did?

 

[Byrd]

Was it windy or thunderstorms outside?

Welcome to the hobby of dealing with vintage electronics

 

[KGLlewellyn]

Was it windy or thunderstorms outside?

Welcome to the hobby of dealing with vintage electronics

I mean, I’ve been in this game for 20 years at least, but this one was a bit wild.

Usually there is at least some rhyme or reason even if not initially apparent, this one…man, it took the biscuit.

I will say, it started acting up immediately after I configured a 10GB image for use as drive D: , though undoing that change didn’t fix it. Idk…maybe it could have screwed the BIOS

My other running theories are either some kind of thermal shenanigans, caps being wonky on the card or the supply not delivering stable power to the card.

…though now you mention it, it does get quite windy here!

 

[Ortho'sDeli]

Out of curiosity, do you run Windows on yours? Mine will handle DOS just fine, but will always freeze during Win 95 installation.

 

[KGLlewellyn]

Out of curiosity, do you run Windows on yours? Mine will handle DOS just fine, but will always freeze during Win 95 installation.

Well when it works it handles Windows just fine, I actually run Win98SE on mine with good results.
Unfortunately it is proving rather temperamental and will just up and not work sometime. I note that this card has a bunch of electrolytic caps, so I wonder if they may be falling under spec. I'll take a look at them with an ESR meter soon.

 

[KGLlewellyn]

Well found out why I might be having issues. Cap C111 appears to have gone 'bang'. I don't suppose any kind soul would be willing to help me get the capacitance value?

Other white caps in the area all appear to be marked 'A5' if that gives any clues? I'm thinking from a cursory Google Search of Cap Codes that A5 is 100nF.

EDIT: Okay, it's not 100nF, C120 below marked A5 is 1.09nF, so I guess a 1nF cap?

 

[Ortho'sDeli]

Tried finding it on mine. The spot on the PCB you pictured is very different from my card.

 

[nyef]

Well found out why I might be having issues. Cap C111 appears to have gone 'bang'. I don't suppose any kind soul would be willing to help me get the capacitance value?

C111, like C120, looks like a decoupling capacitor... which, as recently discussed elsewhere, is unlikely to cause a straight-up circuit failure. That said, a decoupling capacitor exploding is extremely suspect, so I'm going to hope that it's more a case of having been mechanically crushed. If it did explode, then there's something deeper wrong with the board.

I'd advise checking any tin can caps on the board for evidence of leakage, and look for signs of trace rot near them either way. Also check for any evidence of other possible mechanical damage, just in case. And check your power rails, in case you've got deeper problems with the host system.

 

[KGLlewellyn]

Well, the card is still detected by the computer and even tries to start up the PC. It -appears- that the PC does start but acts as if there’s no CPU there at all. If the CPU isn’t getting a stable voltage it’d cause that symptom I would’a thought.

Given I think that this cap is involved with a reduction of noise on power rail for the CPU, it may cause it to not work? That may fit my idea where it -has- worked but goes through times where it doesn’t.

 

[KGLlewellyn]

Bah, no luck on replacing the cap, new 1nF cap in at C111 and no change, the new cap didn't explode. I've replaced a few of the electrolytics around the CPU to no avail. I'm wondering if the CPU itself is kaput. :/ I don't really have a good scope to read out the signals from this card. I might leave it a few days and see if it randomly comes back to life.

 

[KGLlewellyn]

Okay, I think I've figured it out. It's the 4400's PSU not chucking out enough juice. I don't have the Y-Cable so I can't 'easily' try this in another machine so I put the card in my 7600 and routed the audio thru the 4400. Sure enough the 7600 was able to power up the card, and I got a POST beep (which I don't get on the 4400). So yeah, I guess the 4400's PSU is not up to task?

 

[KGLlewellyn]

Y-Cable obtained and yup, as suspected it works fine in the 7600. So tl;dr 4400's PSU is rubbish haha.

 

[Big Ben]

The 4400 should be able to power the card just fine, it is designed to handle the most power hungry PC cards and the 100MHz is not one of these, but I found its PSU to be way cheaper than the other Power Macintoshes.
Mine is due for a recap soon or later, I’m not surprised yours started failing, don’t get me wrong I love this machine, I think it was great for companies, but cost reduction on the PSU was probably not the best idea. Well we can argue it lasted quite some time to be honest.

Glad your PC card is fine, those are quite incredible piece of hardware, would be to bad to see them failed.

 

[KGLlewellyn]

Well, the story got even stranger. Because it's stopped working on the 7600 now. I tried it across different versions of the Mac OS, nada, didn't work. Same issue that hit the 4400.

So just for a crapshoot, I put it in my 9600, lo and behold it works right off the bat. Surely the odds of this card causing two PSUs to go just bad enough that they can't drive the card but leave everything else on the Mac just fine seems a little suspicious?

Just to clarify, if I put other PCI cards in, those cards work fine, so I don't think it's the PCI bus dying either.

I'm really flummoxed by this...

Oh yeah, and if I put the card in other slots in the 7600, the behavior doesn't change either. It's almost like the card is flipping a hidden switch on the Mac which then causes it to stop working for awhile. PRAM reset makes no difference either.

 

[KGLlewellyn]

Alrighty, this is definitely some kind of power issue. The card seems to be -very- sensitive over sufficient power delivery. So in my 9600 I took out four other PCI cards before I put in the PC Compatibility card. Yup, all looking good, working fine.
I put those four other cards back in, PC Card stops working, has the same issue as on the 7600 and the 4400.
Take the cards out, working just fine again.

Now it sorta leaves the question, are these aging PSU's just not up to task or are there some components that are running out of spec on the PC card itself that are causing it to be more sensitive than it otherwise should be?

 

[Durosity]

I only have the GIMO cable

Does the GIMO cable happen to have a part number on it? I’ve been looking for one for ages but never found one, I’m hoping if I had the part no that would help with the search

 

[jessenator]

Now it sorta leaves the question, are these aging PSU's just not up to task or are there some components that are running out of spec on the PC card itself that are causing it to be more sensitive than it otherwise should be?

I'd say you're right on track with the power supply, but in something like a 9600, that's fishy. I would have thought they were rather over-spec'd, but I'm not terribly familiar. I wonder if any of the developer/tech notes for both the 9600 and 7600 have any insight on the PCI bus, what its power limitations are, etc. Again, just a hunch.

I do know that the 4400/Tanzania/LPX-40 implementation was beefier closer to the PC world's standard, simply because of its original PREP/CHRP origins, right down to the PSU requirements. An LPX PSU was not terribly far off from the standard ATX PSUs of the WinTel world.

Additionally, there's documentation for the PCCompat cards that specifically says to never install the 4400's card* into a 7200 as damage will occur. I don't know how far that extends to the other Power Macs of the era.

I give the * caveat, because this warning was specific to the 4400/200 PC Compatible model's card: a Cyrix (pentium-rated) 166, which had a significantly higher power draw than the Pentium 166 or 100 cards. This was from either the +5V or +3.3V rails. However, based purely on PSU specs, the ATX and LPX supplies of the era were built around different standards than the Macintosh models'.

Hope that helps.

edit, I forgot I wrote this specific post, but kind of sums it up.

x86 card revival thread

Well, I stupidly missed out on an OrangePC 660 that I desperately wanted on eBay, but I did end up otherwise adding to my collection with an Apple 586 7" PCI card and an Apple Pentium (100 Mhz I think) 12" PCI card. That brings me to: Apple Houdini II 6100 PDS card (Intel 486 DX2 66) Apple 586...

68kmla.org