Deunan wrote on 2024-08-12, 12:36:

Terminator power is a tricky thing. I'd try to disable it on the drive side. Then try to disable the termination on the drive completly - see if it changes anything. Try disabling Ultra on the drive as well, via A9 jumper (but leave the termination on) if the card doesn't seem to offer link downgrade options. I only have AHA-2940UW and UW2 cards, these do have plenty of link options.

A9 says "enable Ultra-SCSI". Does that mean jumpering A9/B9 would disable it? I can try taking off the termination jumpers and trying A9 instead, as I haven't tried that combo before, but I do know I had the exact same problems on the previous adapter (AHA-2940) with no termination jumpers on the drive at all.

My assumption was A9 was enabled by default. But if it's disabled now then enable it - try everything you can change to see what happens. In fact try A4 to disable parity checking as well. It seems the 0x40/0x8x sense code is "A correctable ECC error occurred; x equals the length of the error." - that would suggest some communication issue.

Terminator power usually comes from the host side, enabling it on the device(s) can lead to power backfeed and is not safe, even on SCSI which is pretty robust in general.

BTW what ID is that card set to now? Preferably 7, because older devices can't address SCSI IDs 8-15.

Deunan wrote on 2024-08-12, 13:22:

My assumption was A9 was enabled by default. But if it's disabled now then enable it - try everything you can change to see what happens. In fact try A4 to disable parity checking as well. It seems the 0x40/0x8x sense code is "A correctable ECC error occurred; x equals the length of the error." - that would suggest some communication issue.

Terminator power usually comes from the host side, enabling it on the device(s) can lead to power backfeed and is not safe, even on SCSI which is pretty robust in general.

BTW what ID is that card set to now? Preferably 7, because older devices can't address SCSI IDs 8-15.

The card is set to 7 by default, yes. I'll check out some configurations and see what happens. As for device age, the biggest MO drives were Sony's latest model and I think they were released around 2002 or 2003, so it's by no means what I'd call an old SCSI device. Ofc now I think it's a shame they didn't do these drives in IDE, since now in hindsight I can easily say that's a very comparatively problem-free standard .

Ok, I've tried:
- no jumpers at all
- jumpers on 4 (disable SCSI parity) and 11 (terminating)
- jumpers on just 11
- jumpers on 9 and 11
- restritcing the speed in scsiselect to 5 MB/s
- several other settings such as PnP, enable or disable treating removable disks as hdds etc

nothign worked, keep getting the same error. I'm thinking maybe I should pick up a new cable? Mine is pretty old and crude looking (like those early IDE cables, thick bands etc) so maybe it's too slow for the device? I've also been trying to find what the error codes mean, Wikipedia says that C 40 is "DRAM parity error", but Q of 80 and higher is supposed to be "vendor specific".

It would be nice if you had another scsi device to try.

just an FYI to you and deunan about the scsi ID's 8-15... Narrow SCSI never ever has those, they are only available on wide SCSI.

I found this manual: https://www.manualslib.com/manual/926403/Sony-Smo-F551.html

it's for one generation older model than my drive, but it looks like many things apply 1:1, including a detailed explanation of the function pins in the back. Stumbled on a manual of e-501 before that, and that one was from 1990. Sony has made a lot of generations of these things. Anyhow, hoping to find an answer about what the codes mean, if it's a physical problem with the drive then I'll just return it and pick up another one (hopefully much cheaper). So far everything points to the drive being broken, even though it was true that the first adapter I had also couldn't run SCSISelect for some reason, so I'm glad I replaced that one. But yeah, it would seem there is some physical problem with the drive - especially given that I was able to have it format one disk and write something to it before it failed again. So I'm starting to think the SCSI setup is not the culprit here. I think I'll just return it.

The only question is whether I could have physically damaged the drive during my testing, which I doubt but it'd be good to make sure so I can let the seller know when returning the product.

Spiffles wrote on 2024-08-12, 18:00:

I'm thinking maybe I should pick up a new cable? Mine is pretty old and crude looking (like those early IDE cables, thick bands etc) so maybe it's too slow for the device? I've also been trying to find what the error codes mean, Wikipedia says that C 40 is "DRAM parity error", but Q of 80 and higher is supposed to be "vendor specific".

Look for a different cable. Actually the older ones are usually better quality, using thicker wires, but one or more could be internally broken. Although in my experience it's usually the connectors that go bad, if the cable was stored in humid place (or near salt water body). I've had IDE and FDD cables go bad due to connector oxidation.

Here's a decent list of error codes for 2930U, but it's pretty universal for Adaptec devices: https://storage.microsemi.com/en-us/support/s … event_codes.htm

It is possible the issue is with the drive itself but in general I'd expect error code 3 (media) not 4 (HW) error if the problem was just with reading or writing. I still think it's comm issue. But such issues can also be caused by dried out or leaked caps in the drive itself. If it was never serviced properly you might want to look into that. 2002 or 2003 is still 20 years old now, and the mid to late '90 and early 2000 electrolytic caps are not exactly known for being very robust. I've seen '80 caps in better shape than these. Same goes for your PC PSU, excessive ripple and tired caps in the drive can just add up and cause instability. Do make sure you're not fighting with the wrong issue.

Deunan wrote on 2024-08-12, 19:58:

Look for a different cable. Actually the older ones are usually better quality, using thicker wires, but one or more could be internally broken. Although in my experience it's usually the connectors that go bad, if the cable was stored in humid place (or near salt water body). I've had IDE and FDD cables go bad due to connector oxidation.

Here's a decent list of error codes for 2930U, but it's pretty universal for Adaptec devices: https://storage.microsemi.com/en-us/support/s … event_codes.htm

It is possible the issue is with the drive itself but in general I'd expect error code 3 (media) not 4 (HW) error if the problem was just with reading or writing. I still think it's comm issue. But such issues can also be caused by dried out or leaked caps in the drive itself. If it was never serviced properly you might want to look into that. 2002 or 2003 is still 20 years old now, and the mid to late '90 and early 2000 electrolytic caps are not exactly known for being very robust. I've seen '80 caps in better shape than these. Same goes for your PC PSU, excessive ripple and tired caps in the drive can just add up and cause instability. Do make sure you're not fighting with the wrong issue.

Thanks for the suggestions. I've already returned the drive, and I think it's a good call even if the problem wasn't it, because I can pick one up from overseas for literally 2,5x cheaper. If the 2nd one has the same problems, I'll definitely start looking at the cable. Though looking at it, the cable does look robust and has no visible damage on it. It's also fairly short, only supporting 2 devices, and I hear with SCSI shorter cables are generally advised. I'll keep you posted when I pick up a second drive.

I think you better find the 50 pin cable that have the termination build-in on the cable.
or you can use the 68 pin to 50 pin adapter and using the 68 pin lvd cable with build-in terminator
using those cable will eliminate one of the painful issue of bus termination.

also made sure to check the blue led on the front of the drive
solid blue: disc is recognized and can be read
blinking blue: disc is not recognized and can't be read.

Hey, so my new drive just arrived, tested it out and it works like a charm. Turns out the problem actually was the previous drive, it had some sort of internal damage that prevented it from reliably formatting and reading the discs. There was nothing wrong with the SCSI setup - even my first host adapter that didn't have a BIOS would have made this new drive work. Of course it's better to have an adapter where you can access the SCSI Select utility, but turns out that was not the primary problem in the first place.

So yeah, just like that my project of having a PC that can read legacy media under Windows 10 is complete and successful 😁