bear
Veteran Member
You can't usually read a crystal by connecting it straight to a scope. The scope inputs have too high a capacitance for the crystal to oscillate correctly.
- VCF South West - June 14 - 16, Davidson-Gundy Alumni Center at University of Texas at Dallas
- VCF West - Aug 2 - 3, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 7 - 8 2024, Schaumburg, IL
- VCF SoCal - Mid February 2025, Location TBD, Southern CA
- VCF East - April 2025, Infoage Museum, Wall NJ
-
Please review our updated Terms and Rules here
The HP Apollo 700 Series I'm Trying To Get Up And Running
- Thread starter jmetal88
- Start date
It's not a crystal, though. Well, I'm sure it's got a crystal in it, but it's one of those 4-pin modules, and I thought those were different than just a normal crystal.
paul
Veteran Member
I find it awfully hard to believe that the oscillator could be at fault.
Have you checked for 5V present on the CPU board?
Otherwise it does look like replacing that board is your best option since we don't have a schematic. I think you can use boards from a 725 or 735 as well (pics attached.)
Have you checked for 5V present on the CPU board?
Otherwise it does look like replacing that board is your best option since we don't have a schematic. I think you can use boards from a 725 or 735 as well (pics attached.)
Attachments
Last edited:
gslick
Veteran Member
According to information here, you can swap the boards with 735 version boards, as long as you swap both the CPU and core I/O boards together as a matched set:
http://www.openpa.net/systems/hp-9000_720_730_750.html
Yeah, I checked for 5V present on several chips, and it was there on all of them. It's probably not the oscillator at fault, I agree, but it's the only thing I can think to replace that's easy to get and would make a big difference if it was damaged.
And like gslick said, you can use boards from a 725 or 735 if you also swap out the I/O card.
bear
Veteran Member
730 or 735. 725 is a more traditional PC-style desktop box. Looks like a Vectra. Doesn't resemble the 720, 730, or 735 in any way.
You're right, I totally missed that about the 725. Yeah, 735 is the one I was thinking of. The 730, I think you can just use the CPU card from without replacing anything else.
paul
Veteran Member
Yes, "725" is an error on my part. The first CPU pictured is from a 730 that I upgraded to a 735 including the I/O board. I was able to combine the memory from both CPU boards.
Paul, do you still have the 730 card? Is it in use? If my last-ditch 'change the oscillator just because I can' effort doesn't work, I might be interested in buying it from you, depending on the price.
Yeah, the new crystal was no help. Oh well, that exhausts all of my cheap options now anyway.
I did notice some interesting behavior while I was testing this time, though. I can toggle the 'service' light if I start the machine up without the I/O card installed. Does anyone know if that indicates anything in particular?
I did notice some interesting behavior while I was testing this time, though. I can toggle the 'service' light if I start the machine up without the I/O card installed. Does anyone know if that indicates anything in particular?
paul
Veteran Member
Sorry, no, tossed everything out that was surplus when I moved here to escape the reign of GWB.
If I, or both of us, have learned anything from this thread it's don't troubleshoot a power supply on hardware you can't easily replace!
Ah, that's too bad.
Well, I guess I'll just hold on to the computer until either I can afford the $95 replacement CPU card or a cheaper one comes along. I guess unless someone wants to offer me something for the chassis and monitor (this monitor's gigantic, and I don't want the computer leaving unless the monitor goes with it, really).
Well guys, I went ahead and ordered a CPU card for this machine. I found a guy who was somewhat negotiable on price. I was only able to talk him down to $5 less than the $95 card I mentioned in the last post, but hey, $5 is $5. Anyway, it's supposed to be here Tuesday. Fingers are crossed that it'll end up working, but I figure there's still a good chance the I/O card could have been damaged as well, so I'll just have to see.
paul
Veteran Member
Sorry to hear that. A risk of being in this game. I assume you've checked that the memory was installed in the right slots?
You never know what might turn up in the future so the parts might be worth keeping if you have the room.
You never know what might turn up in the future so the parts might be worth keeping if you have the room.
I actually went ahead and ordered an I/O card from the same guy who sold me the CPU card, yesterday. He did me a *little* bit better in price on this one, selling at $85. Originally, I was going to give up at this point, but I got in some overtime at work this week which means I can actually afford to buy another component.
And yeah, I've tried the memory in different slots to make sure. Oddly enough when I got the machine, the memory was installed starting from the highest number slot rather than the lowest, and it actually worked that way before my power supply issue.
I do have a suspected component for the I/O card failure now, though. I was examining the card and noticed that, in addition to the battery-backed CMOS, there is also some kind of configuration EEPROM. I am wondering if the rapid cycling of the power supply somehow corrupted the contents of that EEPROM in such a manner that the machine will no longer get far enough into the boot process to self-test. I plan to test this hypothesis by copying the contents of the new I/O card's ROM chips once it gets here and seeing if the EEPROMs match or if writing the new EEPROM contents to the old EEPROM might even fix the problem. Note that when I'm looking at the file inside a hex editor (I've already copied the old EEPROM), I do see a few areas of identifiable text, but they seem to be placed throughout the EEPROM's address space between a bunch of large, blank areas. It might be that that's just how the EEPROM is supposed to look (maybe it doesn't actually have to store that much and I'm barking up the wrong tree), but it does make me suspicious.
And yeah, I've tried the memory in different slots to make sure. Oddly enough when I got the machine, the memory was installed starting from the highest number slot rather than the lowest, and it actually worked that way before my power supply issue.
I do have a suspected component for the I/O card failure now, though. I was examining the card and noticed that, in addition to the battery-backed CMOS, there is also some kind of configuration EEPROM. I am wondering if the rapid cycling of the power supply somehow corrupted the contents of that EEPROM in such a manner that the machine will no longer get far enough into the boot process to self-test. I plan to test this hypothesis by copying the contents of the new I/O card's ROM chips once it gets here and seeing if the EEPROMs match or if writing the new EEPROM contents to the old EEPROM might even fix the problem. Note that when I'm looking at the file inside a hex editor (I've already copied the old EEPROM), I do see a few areas of identifiable text, but they seem to be placed throughout the EEPROM's address space between a bunch of large, blank areas. It might be that that's just how the EEPROM is supposed to look (maybe it doesn't actually have to store that much and I'm barking up the wrong tree), but it does make me suspicious.
Well, still nothing. I'm at a loss here. I don't want to go replacing all the RAM only to find out it's an issue with my power supply replacement, and if it is an issue with my power supply replacement, I'll probably never be able to figure out what it is without access to a working original power supply to check it against.
paul
Veteran Member
Sorry, I lost track - what are you referring to by that? Can you not verify that you have the required output voltages placed on the backplane when loaded by the boards you have? Or are you unsure what those voltages are?
After my power supply blew up and I couldn't fix it, I hacked in an ATX power supply. I know I have 5V, GND, 12V, -12V, -5V, and the turn-on signal in the right places, but there are a few extra traces on the power supply which I have no idea what they're supposed to do (and I wasn't able to trace them anywhere on the backplane). So basically I know all the power supply voltages are there, but I don't know if maybe there's some signal voltage I'm missing that's preventing the I/O board from initializing.
paul
Veteran Member
OK, I see. As far as I know there is a +/-9V as well, and the fans draw off the +9. I have no idea if anything else does, and what the -9 is for, or if it's even used. RS-232C perhaps?
Is it completely dead, no diagnostic lights on? At some point further back you had some activity with that same power supply arrangement, no?
Are the memory SIMMs installed in pairs from the slots closest to the middle, then going outwards?
Is it completely dead, no diagnostic lights on? At some point further back you had some activity with that same power supply arrangement, no?
Are the memory SIMMs installed in pairs from the slots closest to the middle, then going outwards?