VOGONS


First post, by 386SX

User metadata
Rank l33t
Rank
l33t

[Update at the end]

Hi,
I've this mainboard that in the past was running ok but never used since, cause back then very long time ago I probably did an error mounting a 486DX4100 3,3v in the wrong way (no comments.. ) in the socket or I used ECC ram into but by the way it didn't post together with some bad smell coming (hopefully) from the cpu itself or from the ram receiving 5V instead of 3,3v, I don't know, anyway since then I didn't even try to know if I burned it or just the cpu or that ram module. The bottom ram socket has the left plastic pin (which block the simm) broken and maybe the ram did move outside from that. No components on the board seems bad, no lines burned evrything looks ok.
Now, enough time has passed to forget such error and to test if I destroyed something on the board.
Before trying to boot it first of all can you confirm two things:

1) is it a LS486e motherboard? The code is a E117895 and it's a "Rev: C" the layout looks like it.
2) if I've to try to boot it again, in that socket which is the right way to mount the cpu? I suppose the cutted cpu angle in the above-left point leaving one line of free pins on each sides right? (which is what I always thought I did back then still that smell something went wrong in that test).

Also can you give me some info about how fast was this mobo? The jumper config I remember was quite tricky.
Thanks

Attachments

  • mainboard2.jpg
    Filename
    mainboard2.jpg
    File size
    400.15 KiB
    Views
    782 views
    File license
    Fair use/fair dealing exception
  • mainboard.jpg
    Filename
    mainboard.jpg
    File size
    240.61 KiB
    Views
    799 views
    File license
    Fair use/fair dealing exception
Last edited by 386SX on 2020-05-30, 15:43. Edited 9 times in total.

Reply 1 of 12, by mpe

User metadata
Rank Oldbie
Rank
Oldbie

Might be a LS486e. There were many revisions of that board.

SiS496 is the latest gen 486 chipset so it should make for a very fast 486.

Installing 486 sideways likely destroyed either the CPU or the motherboard or both. Happened to me as well long time ago. Installing any sort of RAM should be fine (although ECC 72 pin SIMM are very uncommon).

Blog|NexGen 586|S4

Reply 2 of 12, by 386SX

User metadata
Rank l33t
Rank
l33t
mpe wrote on 2020-05-28, 17:12:

Might be a LS486e. There were many revisions of that board.

SiS496 is the latest gen 486 chipset so it should make for a very fast 486.

Installing 486 sideways likely destroyed either the CPU or the motherboard or both. Happened to me as well long time ago. Installing any sort of RAM should be fine (although ECC 72 pin SIMM are very uncommon).

That's why I felt (it was a long time ago) so bad to even think having a cpu installed in the wrong way and hopefully ruined only the cpu and not the mobo that I never even tried it again. I had and still have some 64MB ECC simms that at that far time I was thinking could be supported and now I'm sure they were HP ECC modules.
Anyway now I'm trying again. First of all the socket. I may try first with a simple Intel DX2-66 cpu surely not again an DX4-100 3,3v but I remember I spent long time into checking the jumper configs. Any advices on it? That config is probably the one left from years ago when I put it in the plastic and lost interest in it.

Reply 3 of 12, by evasive

User metadata
Rank Oldbie
Rank
Oldbie

https://iq.ul.com/pwb/Trade.aspx
E117895="EXCELLENCE ELECTRONIC CO LTD"
also used by ATL from Canada, Arvida technologies and so on and so forth...

The simple Rev: C is indeed consistent with the lack of markings on a Lucky Star board. However on the LS486E Rev C the COM ports are located between the ISA and PCI slots whereas on this board they are neatly near the others.

I'll see if I have something matching somewhere...

Reply 4 of 12, by 386SX

User metadata
Rank l33t
Rank
l33t
evasive wrote on 2020-05-28, 18:30:
https://iq.ul.com/pwb/Trade.aspx E117895="EXCELLENCE ELECTRONIC CO LTD" also used by ATL from Canada, Arvida technologies and so […]
Show full quote

https://iq.ul.com/pwb/Trade.aspx
E117895="EXCELLENCE ELECTRONIC CO LTD"
also used by ATL from Canada, Arvida technologies and so on and so forth...

The simple Rev: C is indeed consistent with the lack of markings on a Lucky Star board. However on the LS486E Rev C the COM ports are located between the ISA and PCI slots whereas on this board they are neatly near the others.

I'll see if I have something matching somewhere...

Thanks for the info! I suppose jumper configs may be similar to any Lucky Star equivalent but just to be sure I'm not using the wrong manual also.

Reply 5 of 12, by evasive

User metadata
Rank Oldbie
Rank
Oldbie

as said, will see if I find something matching...

when you got this thing to boot please reply with the POST ID string shown at the bottom. I'd love to identify this one.

Attachments

  • Filename
    SIS-486GREEN.zip
    File size
    69.47 KiB
    Downloads
    40 downloads
    File license
    Public domain

Reply 6 of 12, by 386SX

User metadata
Rank l33t
Rank
l33t
evasive wrote on 2020-05-28, 18:47:

as said, will see if I find something matching...

when you got this thing to boot please reply with the POST ID string shown at the bottom. I'd love to identify this one.

Question: can someone explain to me the cpu voltage config of this mobo? For a 5V cpu (es. Overdrive 66mhz) there's the config JP28: X W4: ON. Does it mean that JP28 is all opened (four pins are there) or all shorted? W4 is already closed/shorted.
Usually on the manual when there is more pins (1-2-3-4) they write 1-2,3-4 and not that "X"....
Thanks

Reply 8 of 12, by 386SX

User metadata
Rank l33t
Rank
l33t
evasive wrote on 2020-05-29, 18:09:

I believe that means: does not matter/don't care. So basically if W4 is in place (ON) it is 5V and there's no adjustment possible with JP28.

Thanks, I begin to suspect something is wrong with the mobo I cannot get it to boot. I tried both Intel and Amd configs with DX2, DX4, 5V or 3V cpus and no signal get on screen, the psu start everything get warm (cpu, cache, chipset) even if the power regulator on the right corner above stay strangely too cold..
I tried with two different ISA video cards and a PCI ones, nothing. But I can't see anything damaged on board...
I'll go with trying everything different rams I have but I don't think that's the problem..

Reply 9 of 12, by evasive

User metadata
Rank Oldbie
Rank
Oldbie

Do you have a POST diagnostics card? Something like this:
https://www.amazon.com/Optimal-Shop-Computer- … c/dp/B008BZBKXC

edit: another question, do you have the possibility to check the flashrom chip? If the contents are corrupted the board will not boot either.

Reply 10 of 12, by 386SX

User metadata
Rank l33t
Rank
l33t
evasive wrote on 2020-05-30, 08:24:

Do you have a POST diagnostics card? Something like this:
https://www.amazon.com/Optimal-Shop-Computer- … c/dp/B008BZBKXC

edit: another question, do you have the possibility to check the flashrom chip? If the contents are corrupted the board will not boot either.

EDIT: found that card! I'll try and update here as soon as possible!
I have no flash eeprom tools to check it, I was thinking to the old trick to flash it on another board but it's certainly dangerous for the eeprom chip pin if I've to change voltages, I'm gonna check the eeprom model first.

Reply 11 of 12, by 386SX

User metadata
Rank l33t
Rank
l33t

Update:
The pci post card give the code " 00 " that from the manual means: (only for AMI bios table) code copying to specific areas is done, passing control to INT 19h boot loader next.
The voltages leds that are on are, -12, +12, 5v while NOT the 3,3v or the others CLOCK, FRAME, IRDY, RESET. But I'm not sure about the 3,3v led accuracy cause it may be broken on the post card I already wrote in hte past on the manual that it didn't light up.

Also I was reading some old threads on the same mobo (it looks like someone here had the same exact model in an old thread), could it be also the voltage regulator marked HSMC TIP127 gone? I read online that it should be a PNP transistor one and with some youtube guide I suspect there may be something wrong on it? It's not shorted but both the Base and the Emitter seems to gives values reading on the multimeter in both ways Positive on B, Negative on E and the opposite (?) while I understand it should gives reading only in one way (Negative on the Base and Positive on the Emitter, right?) All the combination gives me readings sometime from .300 to .600. Or maybe I should test it desoldered from the mainboard?
Anyway I also checked the bios eeprom is a 27C1000PC-15 (should be a MX27C1000, 5V supply, 12.5v programming voltage supply).

Update 2:
Some progress... I measured the TIP127 voltage from the emitter to the collector and measured 3.00v with the config for the AMD DX4 100 so I think it may be ok (?). Anyway I tried using another cpu, a Intel DX2 66Mhz, configured for the jumpers (hoping the manual revision is the right one..) and we did some progress.
Before the cpu was cool (the AMD) now the Intel DX2 become quite hot and with this config (and the W4 jumper shorted to 5v) I get the PCi post card to gives a "C7" code! And the IRDY led light up. But in the diagnostic pci card manual there's nothing on that code under the AMI bios name. But online with the WinBios AMI codes (it should have that I suppose) I found this: C7 - CMOS shutdown register test. but in some others manuals also found "ROM BIOS checksum passed. CMOS shutdown register test to be done next.". In both cases may it be a eeprom fault at the end? No beep from the mobo speaker at all but the keyboard three leds light up for a moment.

Reply 12 of 12, by 386SX

User metadata
Rank l33t
Rank
l33t

Update 3: I've downloaded specifications for the eeprom and seems it require the 12v VPP last pin. I've a Socket 7 board that need to be flashed too but boot in the boot block mode and I may try to program this one with that. It has a Atmel AT29C010A-15PC that is 5V VCC only (last pin is NC and not VPP) but I suppose I may use the external 12V voltage pin trick or not? I remember I flashed in the past a board in that way.
What do you suggest?