VOGONS


First post, by NC Mountain Man

User metadata
Rank Newbie
Rank
Newbie

I am trying to build a 486 PC for retro gaming. Below are the parts being used. When I put everything together and switch on the power, nothing happens. I don't get any picture on the screen. I don't get any beeps from the PC speaker. Nothing. All of the jumper setting are appropriate for the current configuration. I removed and reseated the RAM, Cache, processor, and bios chips. I checked the voltages coming from the power supply adapter and they are all correct. All of the ISA slots are getting power. There is 5 volts at the bios chip. I have removed all of the ISA card except for the video card with no change. I even bought another VI15G motherboard, and still get the same results. Does anyone have any recommendations? I have ordered once of the ISA diagnostic cards that will show me the post codes, but it hasn't arrived yet.

Parts:
- AOpen VI15G motherboard
- Diamond Stealth 64 VLB Video Card
- QD6580 Multi I/O VLB Controller Card
- Sound Blaster 16 CT2940
- 16 MB of Ram
- 256K Cache
- Corsair 750 watt power supply with ATX to AT converter (with -5w)

Reply 1 of 15, by Nvm1

User metadata
Rank Member
Rank
Member

What kind or RAM?
I suspect you are using EDO ram instead of FPM ram. Alot of boards don't beep then but simply refuse to start.
Check if you have FPM ram and try it again.

Reply 2 of 15, by NC Mountain Man

User metadata
Rank Newbie
Rank
Newbie

Thanks Nvm1! I appreciate the advice. The RAM I am using was advertised as FPM when I ordered it, but I will Google the actual part numbers tonight to make sure that is what I actually received.

Reply 4 of 15, by NC Mountain Man

User metadata
Rank Newbie
Rank
Newbie

I appreciate all of your help with this! I may have found the issue with my motherboard. When I purchased this board it came complete with RAM, Cache, and Processor. The processor had a heat sink firmly attached to it, so I couldn't tell exactly what it was. The motherboard came with the jumpers set for an "Intel DX4" processor so that is what I assumed it was. I was able to get the heat sink off last night, and found out that the processor is actually an Intel DX2 66 mhz. The adhesive under the heat sink looks pretty burnt, so I am wondering if the processor is toast as well.

1. Will a motherboard post without a processor in it?
2. Would whatever happened to this processor have damaged the motherboard as well?

IMG_2447.JPG
Filename
IMG_2447.JPG
File size
1.69 MiB
Views
1222 views
File comment
Burnt Processor
File license
Fair use/fair dealing exception

Reply 5 of 15, by Deksor

User metadata
Rank l33t
Rank
l33t

A DX2 can live with just a metal heatsink. I think it's just heat that did that, but I doubt the chip's dead. I've got a cyrix CPU that had a sticker underneath with that kind of mark and it was ok.

A DX4 needs 3.3v, but a DX2 need 5v so obviously it won't post. A motherboard definitely need a CPU to POST, how do you think it could work otherwise ? ^^

Trying to identify old hardware ? Visit The retro web - Project's thread The Retro Web project - a stason.org/TH99 alternative

Reply 6 of 15, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t

Weren't there also 3.3V DX2s? How do we know that this isn't one of them? I seem to recall many 3.3V DX2s could overclock to DX4-100 with ease. Best check the s-spec.

"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium

Reply 8 of 15, by Windows9566

User metadata
Rank Member
Rank
Member
NC Mountain Man wrote:
I appreciate all of your help with this! I may have found the issue with my motherboard. When I purchased this board it came com […]
Show full quote

I appreciate all of your help with this! I may have found the issue with my motherboard. When I purchased this board it came complete with RAM, Cache, and Processor. The processor had a heat sink firmly attached to it, so I couldn't tell exactly what it was. The motherboard came with the jumpers set for an "Intel DX4" processor so that is what I assumed it was. I was able to get the heat sink off last night, and found out that the processor is actually an Intel DX2 66 mhz. The adhesive under the heat sink looks pretty burnt, so I am wondering if the processor is toast as well.

1. Will a motherboard post without a processor in it?
2. Would whatever happened to this processor have damaged the motherboard as well?

IMG_2447.JPG

i have that same CPU, a DX2 66, pulled it from a laptop with a bad keyboard and bad screen, which motherboard should i look out for, a one with built in IDE/Floppy controller port, one with PCI slots, or one with ISA/VLB only.

R5 5600X, 32 GB RAM, RTX 3060 TI, Win11
P3 600, 256 MB RAM, nVidia Riva TNT2 M64, SB Vibra 16S, Win98
PMMX 200, 128 MB RAM, S3 Virge DX, Yamaha YMF719, Win95
486DX2 66, 32 MB RAM, Trident TGUI9440, ESS ES688F, DOS

Reply 9 of 15, by amadeus777999

User metadata
Rank Oldbie
Rank
Oldbie

-insert a cpu(best is a DX50), gfx card and ram that you KNOW to work
-put a finger on the northbridge chip and see how warm it gets... if it gets HOT fast it's most likely a more serious issue
-corrupted BIOS is an option, but unlikely

The diagnostics card may be able to solve the mystery.

Reply 10 of 15, by NC Mountain Man

User metadata
Rank Newbie
Rank
Newbie

Unfortunately, the ISA slot diagnostic card wasn't much help. It is just showing dashes, which I take to mean that the POST process doesn't even start. I have ordered another 486 processor to try. I have already swapped out RAM, Video Cards, Cache, and Bios chips so that is about all that is left to try.

While searching through the Vogons' posting history, I did find another thread from a long time ago where someone else had a VI15G with the same issues, and their fix ended up being buying an actual AT power supply. I am using one of the ATX to AT adapters, and my multimeter and the diagnostic card both confirm the correct voltages from it, but maybe there is some issue with amperage.

IMG_2445.JPG
Filename
IMG_2445.JPG
File size
1.83 MiB
Views
1175 views
File license
Fair use/fair dealing exception

Reply 11 of 15, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t
Deksor wrote:

I thought only AMD made them

Apparently not. Check out the s-spec "SK080".
Although I can't confirm it, these should have 16kb write through cache just like the first DX4 chips.

"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium

Reply 13 of 15, by NC Mountain Man

User metadata
Rank Newbie
Rank
Newbie

To update: I have had some unexplained success! I spent over an hour with rubbing alcohol and a plastic scraper getting off the old thermal adhesive tape. It eventually came off and I only lost a little of the 486 label paint.

IMG_E2455.JPG
Filename
IMG_E2455.JPG
File size
840.96 KiB
Views
1104 views
File license
Fair use/fair dealing exception

I know cleaning this adhesive off wouldn't have anything to do with the operation of the motherboard, but for some reason the next time I tried starting the motherboard, it worked. The first time it started the POST process, but came up with errors and beep codes. I then restarted the motherboard without any changes, so I could count the beep codes, and it started normally. I have no idea why. It has worked every time after.

The only issues I am having now is that it wont access a floppy disk, despite using two different controllers and disk drives, and the Sound Blaster 16 (CT2940) I was planning on using, makes the motherboard not POST again when it is installed. At least I am making progress.

Reply 14 of 15, by Deunan

User metadata
Rank Oldbie
Rank
Oldbie
NC Mountain Man wrote:

I know cleaning this adhesive off wouldn't have anything to do with the operation of the motherboard, but for some reason the next time I tried starting the motherboard, it worked. The first time it started the POST process, but came up with errors and beep codes. I then restarted the motherboard without any changes, so I could count the beep codes, and it started normally. I have no idea why. It has worked every time after.

I assume you rubbed it while in the socket still to prevent pin bending? If so you might have applied enough force to actually move the CPU around a bit and rub agains the socket. So if there was some dirt that prevented good contatct it's now gone. Inspect the CPU pins and the socket to see if it's clean and not gummed up with something. The dashes on your test card mean CPU has not run enough code to output even the first code, but the reset line is not asserted. So that leaves the CPU or socket to blame I think (assuming it's not a serious problem with mobo chipset)

Reply 15 of 15, by treeman

User metadata
Rank Oldbie
Rank
Oldbie

either that or some cracked via or cold solder joint started making contact. I got 1 or 2 boards like this pressing chipset or flexing it in certain spots brings it to life... then screwing it back in a case and inserting cards kills it again