Micron Xceed SE/306-48 and SE/30 issue

jimbojones

Well-known member
I recently acquired a Micron Xceed SE/306-48 (sold untested, so a gamble) to go in my SE/30.

Unfortunately, when I put the card in my SE/30 it makes a bong and shows this screen:

20231001_202649.jpg

Does anyone have any pointers on if there is anything I can do/try?

My SE/30 with 68MB of RAM runs fine without it.

Here are some pics of the card (on the bottom of the card it looks like there is rework - just up from the bottom where it says 3754 and a bit to the left - but I don't know if that's from the factory):

20231001_203053.jpg
20231001_203319.jpg
 

Phipli

Well-known member
Check for shorts or trace breaks, then check for shorts over the capacitors. Check for bent pins on the main connector.

Did you disturb your ROM while you were at it?

Could you take a closer, clearer, photo of this area :

20231001_203319.jpg
 

Phipli

Well-known member
Also, could we have a closeup of here ?
20231001_203053.jpg
Actually, I think this one just looks dark because the neighbouring chip is socketted and the rest aren't, so that's probably fine.
 

tvj

Member
I have the same card, and that rework on the back doesn't look factory to me. My card has no rework. Are those traces cut? It also looks like the gouge is so deep where the reworked wire is that the internal traces might be compromised. I'd check with a multimeter in continuity mode to make sure.

Can you check the voltage on the SE/30 from the floppy port with and without the card slotted? If the SE/30 power supply is marginal, you'll get jailbars from the increased demand on the power rail.
 

joshc

Well-known member
Doesn't look like a factory bodge. Someone has knocked/damaged it previously and tried to repair it.

Have you done a continuity test of those traces?
 

jimbojones

Well-known member
Thanks all. I'll check the power supply and try without the hard disk etc.

I'll check the continuity of the traces as well.
 

jimbojones

Well-known member
I'm a novice in this area, so bear with me :)

I tested the traces around the affected area using a multimeter from the blobs at each end of the trace - all seem to have continuity.

I unplugged the floppy and hard disks and tried again, but to no avail. I'll try and do some voltage testing.

I can't find a manual for this card. It's a long shot, but could my memory configuration be an issue (4x1MB + 4x16MB)?
 

Phipli

Well-known member
It's a long shot, but could my memory configuration be an issue (4x1MB + 4x16MB)?
No, that isn't anything to do with the video card.

Did you check the main connector for bent pins?
Did you check all the yellow capacitors to see if any were a dead short?
 

joshc

Well-known member
May also be worth taking the ROM off the card, spray contact cleaner into the ROM socket, and checking the ROM pins look OK as well.
 

jimbojones

Well-known member
Pins are all good. Will have a go with the ROM and some contact cleaner.

Can I check the capacitors somehow without taking them off the board?
 

Phipli

Well-known member
Pins are all good. Will have a go with the ROM and some contact cleaner.

Can I check the capacitors somehow without taking them off the board?
Yes, just check if there is a short over them using the continuity tester, or the resistance meter
 

jimbojones

Well-known member
Thanks again.

I checked all of the capacitors with my multimeter and I think they seem to look ok (they don't allow all current through).

I took the ROM out, sprayed some contact cleaner in and tried again, but no luck. I tried the card without the ROM in and got the same screen.

I wonder if I'm going to have to try and find another board now and try and swap the ROM. Hmmmm.

I've probably reached the limit of my electronics skills.
 

croissantking

Well-known member
I'm now in possession of OP's card, so I'll continue sharing my progress in this thread. I have the same issue (simasimac) with the card installed in either of my SE/30s.

I dumped the contents of the ROM, which claims to be version 1.1. Would be good to compare to another ROM dump of the same version. Do we know which other versions exist for this card? @tvj have you dumped yours?

 

Attachments

  • XCEED_SE306_48_v1.1.BIN
    8 KB · Views: 4

zigzagjoe

Well-known member
I'm now in possession of OP's card, so I'll continue sharing my progress in this thread. I have the same issue (simasimac) with the card installed in either of my SE/30s.

I dumped the contents of the ROM, which claims to be version 1.1. Would be good to compare to another ROM dump of the same version. Do we know which other versions exist for this card? @tvj have you dumped yours?

CRC Is valid, so this is fine. It's unlikely to be an issue with the DeclROM as you'd have to try very hard to prevent a machine from booting via DeclROM. You could verify by removing the card ROM and trying to boot with the card installed, symptom should remain the same.

Here's an assortment of schematics courtesy of the Bolle.

For troubleshooting, I suggest removing the system ROM and watching bus activity post-reset.
If no /AS ever then the CPU is stuck due to a critical issue like missing clock, /halt or /reset held low, and other incompatible with life scenarios.
If you see one/few /AS then it stops: possibly a stuck data line/defective data buffer on the card.
If you see continuous /AS and bus walking behavior: things are really weird.
You might also check that /STERM isn't held low or something.
 

Attachments

  • XCeed 306-48.zip
    2.1 MB · Views: 7

croissantking

Well-known member
CRC Is valid, so this is fine. It's unlikely to be an issue with the DeclROM as you'd have to try very hard to prevent a machine from booting via DeclROM. You could verify by removing the card ROM and trying to boot with the card installed, symptom should remain the same.

Good to know. I think the previous owner also said the symptoms were the same with ROM removed.

I always get a chime, and I have made it to a grey screen with mouse pointer once or twice - but then the screen corrupts with black and white dots and the computer freezes. Mostly it’s a simasimac though.

For troubleshooting, I suggest removing the system ROM and watching bus activity post-reset.
If no /AS ever then the CPU is stuck due to a critical issue like missing clock, /halt or /reset held low, and other incompatible with life scenarios.
If you see one/few /AS then it stops: possibly a stuck data line/defective data buffer on the card.
If you see continuous /AS and bus walking behavior: things are really weird.
You might also check that /STERM isn't held low or something.

Should I do this with a logic probe?
 

zigzagjoe

Well-known member
Good to know. I think the previous owner said the symptoms were the same with ROM removed.

I always get a chime, and I have made it to a grey screen with mouse pointer once or twice - but then the screen corrupts with black and white dots and the computer freezes. Mostly it’s a simasimac though.



Should I do this with a logic probe?
Ah, OK, I missed that it is in fact chiming. So don't bother with those troubleshooting steps; you're in case 4, things are really weird.

I'd definitely recommend confirm that the behavior remains constant without a ROM as a next step. Do you have a logic analyzer?
 

croissantking

Well-known member
Ah, OK, I missed that it is in fact chiming. So don't bother with those troubleshooting steps; you're in case 4, things are really weird.

I'd definitely recommend confirm that the behavior remains constant without a ROM as a next step. Do you have a logic analyzer?
No, I’ll get one.
 

croissantking

Well-known member
The cheap salae logic clones go a long ways - Not super fast, but enough for simple needs and the price can't be beat. You may want some test clips to go with it.
I can’t easily tell which ones are the clones - would this be one of them?

And then this for the clips?

Here's an assortment of schematics courtesy of the Bolle
I had a look at these. Very very useful; he has done quite a lot of work to draw these up. Was it for a particular project, maybe to create a replica card? Or just for fun to understand it better? It would probably be quite easy to produce replicas as the card is a simple two-layer affair, with off the shelf parts and programmable GALs. No custom ASICs that I can tell. The only issue would be getting ZIP VRAM, but I suppose you’d redesign it to take SOJ package chips.
 
Last edited:
Top