SPD Error: Possible To Override?

[Paul]

Hi.

I have a couple of identical ECC PC100 memory modules which have been
working fine for years, but have recently started playing up. This
occurred after my computer had developed a seemingly unrelated fault
(I stripped and reassembled it as I've done dozens of times before and
then the memory problem appeared). When I boot up, the POST begins
and gives me an SPD error --- it's the one where the frequency is
reported as 66MHz (although I *know* that the RAM's 100MHz). I
imagine that the only problem is that (for some reason) there's duff
information in the SPD EEPROM. My motherboard is an old SE440BX-2,
and the BIOS doesn't appear to have any means either of ignoring the
SPD report or of manually entering the timing data (at least as far as
I can determine). Does anyone know of a way around this, please, or
are the modules now unusable with this board? ...Or indeed unusable,
period.

Thanks,


Paul.

 

[Franc Zabkar]

Hi.

I have a couple of identical ECC PC100 memory modules which have been
working fine for years, but have recently started playing up.

Both modules?

This occurred after my computer had developed a seemingly unrelated fault
(I stripped and reassembled it as I've done dozens of times before and
then the memory problem appeared). When I boot up, the POST begins
and gives me an SPD error --- it's the one where the frequency is
reported as 66MHz (although I *know* that the RAM's 100MHz). I
imagine that the only problem is that (for some reason) there's duff
information in the SPD EEPROM.

Damaged socket?

My motherboard is an old SE440BX-2,
and the BIOS doesn't appear to have any means either of ignoring the
SPD report or of manually entering the timing data (at least as far as
I can determine). Does anyone know of a way around this, please, or
are the modules now unusable with this board? ...Or indeed unusable,
period.

Thanks,


Paul.

I'd try to read the SPD data using an appropriate utility, eg CTSPD.
That may at least tell you whether the serial EEPROM is accessible, in
which case it could be a data corruption issue.

AFAIK the supply rail (VDDSPD, pin 184) for the SPD EEPROM need not be
the same as for the DRAMs. Could this rail be faulty?

- Franc Zabkar

 

[Paul]

Hi Franc --- thanks for the response.

Here goes with your questions...

Both modules?

Yup. Weird, eh?

Damaged socket?

Well, I've tried the modules on another SE440BX-2 mobo, and got the
same error: I imagine it's unlikely that both mainboards would have
the same fault. My guess is that somehow the EEPROMs got zapped and
the info contained therein defaulted to 66MHz.

I'd try to read the SPD data using an appropriate utility, eg CTSPD.
That may at least tell you whether the serial EEPROM is accessible, in
which case it could be a data corruption issue.

Thanks --- I'll go and have a look for that.

AFAIK the supply rail (VDDSPD, pin 184) for the SPD EEPROM need not be
the same as for the DRAMs. Could this rail be faulty?

Possibly --- I'm not an electronics whizz, I'm afraid! If it were the
case that this rail alone were faulty, presumably it'd mean that the
memory chips would be all right (though the SPD would be dead)?

A bit more detail about the background of the problem... I'd taken
the machine apart initially because the CD writer stopped responding
(the drive light remains on continuously and the tray won't eject),
and when I put the machine back together, the CD still didn't work
(and still doesn't) and the POST reported an error with the floppy
drive as well as the SPD problem. Naturally, I thought all these
failures must be due to a problem with the motherboard, but swapping
the board for another SE-440BX-2 didn't help. Since then, the problem
with the FDD has disappeared: I still don't know what to do about the
CD writer (I'd guess it's probably had it), but I'm hoping to salvage
the memory, somehow.

Anyway, thanks again for your reply.