Reviving a Power Macintosh 7500/100

[fjm_48]

I pulled an old PM 7500/100 out of storage a couple of weeks ago, and I found that I had lost/thrown out the keyboard, mouse, and Apple monitor. Using a DB15 to VGA adapter, I connected the Mac to a monitor. It powered up to the welcome screen showing system 8.(something). I ordered a Wombat USB/ABB adapter which arrived today. I attached the adapter and a USB keyboard and mouse. After applying power, I got no chime and no video. The HDD did start spinning and I could hear the heads shifting across tracks for about 5 seconds. Then nothing else. I detached the Wombat adapter with no change. One by one I disconnected the HDDs, the FDD and CDROM drive - no change. The PRAM battery is dead. I unplugged the processor board, pressed and held the power switch, waited 10 minutes and jumpered in 3.6V to the PRAM battery. Still no change. I checked the voltages at the PSU connections to the main board; I got 5V, 12V -12V, 3.3V and 3.6 +/- 5%.

Am I correct in expecting a chime even without a boot drive connected. Is there a schematic available that I could use in troubleshooting? Are there any suggestions on what I should try next?

 

[Byrd]

Hi,

It’s time to strip it down to the basics, with it probably being a bad connection from the CPU daughtercard, cache, RAM. I’m finding the “Outrigger” 7x00 Macs are more temperamental to run these days, because of the sheer amount of plug in components if one fails or is marginal, there goes a reliable Mac. Clean all the connectors, make and female, with electronic solvent spray. Blast out any dust too.

 

[fjm_48]

Well, since I'm tearing into it anyway, I'll clean all the connections. Thanks

 

[Phipli]

Another thing to check is the voltages - the power supplies in those can fail. I'd start by checking the voltages on a hard disk power connector - they should have 5v and 12v on available on them.

 

[fjm_48]

Phipli, the voltages on the HDD power connectors (with one HDD connected) are 5.06 and 11.81. Both of those are easily within +/- 5%. But, I do suspect the PSU also. Sometimes during power up, the power LED comes on for a second or two then shuts down. I'm going to replace the PSU and try again. The replacement will arrive tomorrow.

 

[techstep]

I have a spare 7500 logic board that is working if you are interested/need it. Would only ask what I paid for it+ shipping. Feel free to DM me.

 

[fjm_48]

Hi,

It’s time to strip it down to the basics, with it probably being a bad connection from the CPU daughtercard, cache, RAM. I’m finding the “Outrigger” 7x00 Macs are more temperamental to run these days, because of the sheer amount of plug in components if one fails or is marginal, there goes a reliable Mac. Clean all the connectors, make and female, with electronic solvent spray. Blast out any dust too.

I've removed all plugged components from the motherboard, cleaned the contacts, and reinstalled. I thoroughly cleaned everything with solvent cleaner. Still no joy. I get about 3 seconds of hash (white noise) from the speaker right after power up. Then nothing else. Do you know what the minimum hardware/software configuration needed to receive the chime?

 

[fjm_48]

Another thing to check is the voltages - the power supplies in those can fail. I'd start by checking the voltages on a hard disk power connector - they should have 5v and 12v on available on them.

Phipli, the voltages on the HDD power connectors (with one HDD connected) are 5.06 and 11.81. Both of those are easily within +/- 5%. But, I do suspect the PSU also. Sometimes during power up, the power LED comes on for a second or two then shuts down. I'm going to replace the PSU and try again. The replacement will arrive tomorrow.

 

[Byrd]

Static noise on a recent 7300 restoration (two of them, also suspected the PSU but all well), was due to the processor card. Check this over well and maybe hold it down firmly or at a fraction of an angle from 90 deg then start up. The CPU core is also highly fragile check it looks OK by removal of the heatsink and reapply thermal paste.

 

[fjm_48]

I've ordered ceramic caps to replace all of the aluminum electrolytics on both the logic board and the processor card. I'll post again with the results in probably about a week.

 

[s_pupp]

I've ordered ceramic caps to replace all of the aluminum electrolytics on both the logic board and the processor card. I'll post again with the results in probably about a week.

Ceramic caps did not work for me in a PM8600, MDD G4, nor in one of my SE/30's. I've read in a Kemet publication that the effective capacitance of ceramic caps decreases with application of DC voltage, so when exposed to DC, ceramic caps need to be higher capacitance than the ones they are replacing. They also decrease in capacitance over time, although the decrease slows down as the capacitor ages. These tendencies occur to varying degrees based on dielectric used. Because of this, I've moved on to tantalums (need to have at least double the voltage rating than the voltage applied to them, so use 25V for 12V circuit) and aluminum polymer caps because of this.

 

[obsolete]

Yeah, specing ceramic caps is a pain because you really only get their rated capacitance when you measure it with a meter. In a circuit, there's voltage derating, thermal derating, dc bias aging...you need to read the datasheet carefully to try to get a part that will be close to the capacitance you want in actual use.

 

[fjm_48]

Yeah, specing ceramic caps is a pain because you really only get their rated capacitance when you measure it with a meter. In a circuit, there's voltage derating, thermal derating, dc bias aging...you need to read the datasheet carefully to try to get a part that will be close to the capacitance you want in actual use.

Thanks for the reply (s_pupp too). Yes, I'm aware of ceramic cap issues... dc bias, dielectric re-crystallization, etc. I've sized the replacements accordingly. As these devices are for power rail holdup/bypassing, precise capacitance is not critical as long as it is enough.

 

[fjm_48]

This is a cross post of a message that I put on Vintage Computers Federation Forums:

Can anyone tell me for sure whether a Model 8500/9500 cpu card, 820-0849-A, is compatible with the 7500 logic board?

 

[Phipli]

This is a cross post of a message that I put on Vintage Computers Federation Forums:

Can anyone tell me for sure whether a Model 8500/9500 cpu card, 820-0849-A, is compatible with the 7500 logic board?

We can't no.

The assembly part number you want to use to search for that card is the "630-2491" or similar label rather than "820-0849-a". I think the 820 one might be the PCB part number, rather than the specific processor card model and be used for multiple models.

The rest of this reply assumes it is the 180MHz 630-2491, but you need to check your card specifically for a 630- number. The 630-2491 was the first match I got on Google, but there are other models that use the same PCB, hence it not being the number to use.

We can tell you what Apple said, which was that that card was only suitable for the 7300 if it is a 630-2491.

The issue is they never said  why so it might be a stupid reason like "we don't want you to", or a reason you don't mind like "we didn't budget enough power on that PSU if you have three high power PCI cards.

The latter is my guess.

Alternatively - perhaps there is a serious reason that we don't know about. But it is fairly unlikely. If you haven't noticed, the 7500 and 8500 logic boards are actually the same, just with different parts fitted. Doesn't mean they function identically (the 8100 board was able to operate at two different core voltages and as a X2 and a X3 bus multiplier, for example).

But generally I don't worry about switching cards EXCEPT for the "High Performance" ones. They're wired differently. So you shouldn't switch cards from a slower speed computer with one from a 250MHz or faster 8600 and 9600.

See attached

 

[fjm_48]

We can't no.

The assembly part number you want to use to search for that card is the "630-2491" or similar label rather than "820-0849-a". I think the 820 one might be the PCB part number, rather than the specific processor card model and be used for multiple models.

The rest of this reply assumes it is the 180MHz 630-2491, but you need to check your card specifically for a 630- number. The 630-2491 was the first match I got on Google, but there are other models that use the same PCB, hence it not being the number to use.

We can tell you what Apple said, which was that that card was only suitable for the 7300 if it is a 630-2491.

The issue is they never said  why so it might be a stupid reason like "we don't want you to", or a reason you don't mind like "we didn't budget enough power on that PSU if you have three high power PCI cards.

The latter is my guess.

Alternatively - perhaps there is a serious reason that we don't know about. But it is fairly unlikely. If you haven't noticed, the 7500 and 8500 logic boards are actually the same, just with different parts fitted. Doesn't mean they function identically (the 8100 board was able to operate at two different core voltages and as a X2 and a X3 bus multiplier, for example).

But generally I don't worry about switching cards EXCEPT for the "High Performance" ones. They're wired differently. So you shouldn't switch cards from a slower speed computer with one from a 250MHz or faster 8600 and 9600.

See attached

Your reply led me to some further investigation. On the back of the card is a number 600-4830-A. A search if this number brought up only the eBay listing from which I purchased the card. Nothing else on the card resembles an Apple part number. But I detected that the label (600-4830-A) seems to be on top of another label. Underneath is a label 630-2240-A. That card is listed on the attachment from your reply. It is shown to be marked 8500/9500 but as being compatible with the 7500. That same chart (at http://absurdengineering.org/library/Service References, Assorted/Compatibility Charts & Matrices/Processor Card Matrix.pdf) does not show 7500 compatibility.

I think that I may have 2 bad (or 1 bad and 1 non-compatible) cpu cards. Next, perhaps I can find someone to test my 630-1219-A.

Thanks for the help. I may eventually get this solved.

 

[Phipli]

I think that I may have 2 bad (or 1 bad and 1 non-compatible) cpu cards. Next, perhaps I can find someone to test my 630-1219-A.

This is unlikely - "Incomparable" cards still work, just apple say you shouldn't. I really believe it is to do with the case thermal design and power supply rating.

But I detected that the label (600-4830-A) seems to be on top of another label. Underneath is a label 630-2240-A

Interesting - could you share some photos of both sides? And the two stickers (the top one and one under it).

You should be able to tell what speed the card is from the resistors as per my old thread here :

Overclocking 604 Processor Cards

I thought I'd just post this info because I've been looking at it this morning and while I have the datasheet out and heatsinks off... The following table shows the bus multiplier settings for the 604 and 604e. Note "PLL_CFG" 0 through to 3. These PLLs are represented by 4 pairs of resistor...

68kmla.org

 

[fjm_48]

This is unlikely - "Incomparable" cards still work, just apple say you shouldn't. I really believe it is to do with the case thermal design and power supply rating.

Interesting - could you share some photos of both sides? And the two stickers (the top one and one under it).

You should be able to tell what speed the card is from the resistors as per my old thread here :

Overclocking 604 Processor Cards

I thought I'd just post this info because I've been looking at it this morning and while I have the datasheet out and heatsinks off... The following table shows the bus multiplier settings for the 604 and 604e. Note "PLL_CFG" 0 through to 3. These PLLs are represented by 4 pairs of resistor...

68kmla.org

The R1, R2, R3, R9 configuration seems to be 1010. The melted adhesive obscures those a little, but they're still visible. My cpu card seems to be the second (newer) one shown on your attachment.

I see a 45MHz oscillator on my card. I assume that is the bus frequency. That would make the processor core frequency 180MHz. Is that compatible with my 7500 logic board?

The attached are the photos you requested.

Thanks for the ongoing help.

 

[Phipli]

The R1, R2, R3, R9 configuration seems to be 1010. The melted adhesive obscures those a little, but they're still visible. My cpu card seems to be the second (newer) one shown on your attachment.

I see a 45MHz oscillator on my card. I assume that is the bus frequency. That would make the processor core frequency 180MHz. Is that compatible with my 7500 logic board?

The attached are the photos you requested.

Thanks for the ongoing help.

Yeah, it's 180. It's on the FCC sticker, and the CPU itself.

 

[fjm_48]

Yeah, it's 180. It's on the FCC sticker, and the CPU itself.

The board that I photographed is the one that I purchased on eBay. I'm thinking that the melted adhesive may be a sign of overheating of the processor. I think I should ask for a refund before it's too late. I assume from prior posts that the 180MHz core clocking would not be the root of the failure that I am seeing.