VOGONS

Chicony CH-498B VER: 3.0 - No post for some CPUs

Topic actions

First post, by KayTeeFive

Posted on 
User metadata
Rank Newbie
Rank
Newbie

Hi all,

I bought a Chicony CH-498B Ver. 3.0 a few days ago with a Cx486DX2-v80GP, and I've run into a very strange issue.
When I first powered it on with the Cx486DX2-v80GP installed, it would not boot at all. There were no POST codes, as if no CPU was detected.

I then replaced it with a Cx486DX2-80 (green heatsink), the 5 V version, and changed the jumper settings accordingly. The system booted fine and has been completely stable.
Next, I reinstalled the Cx486DX2-v80GP (3.45 V version) and returned the jumpers to the 3.45 V setting. However, it still would not boot.

Oddly enough, if I lift the CPU slightly out of the socket (about 0.5–1 mm), the system starts booting normally. As soon as I push the CPU fully down into the socket, the system freezes.
The Cx486DX2-v80GP CPU pins are in perfect condition. As for the Cx486DX2-80, the pins were not in great condition, but they were straightened and aligned.

I also tried an Intel i486DX2-50 (with the black heatsink), a 5 V CPU taken from a PB430 motherboard. Because of the pin stoppers, it can only be inserted to a certain depth. Unfortunately, it did not work either. Of course, the jumper settings were adjusted according to the motherboard documentation.

To rule out cracked solder joints, I reflowed the solder connections on the CPU socket. The additional capacitors around the LIF socket and the two resistors were already present on the motherboard.

Does anyone have any idea what could be wrong with this motherboard?
Should I replace the LIF socket with a ZIF socket?

Thanks in advance.

Reply 1 of 9, by dominusprog

Posted on 
User metadata
Rank Oldbie
Rank
Oldbie

Use contact cleaner on the CPU, cache and BIOS sockets. Also, put some thermal grease under the voltage regulator.

Duke_2600.png
A-Trend ATC-1020 V1.1 ❇ Cyrix 6x86 150+ @ 120MHz ❇ 32MiB EDO RAM (8MiBx4) ❇ A-Trend S3 Trio64V2 2MiB
OPTi MAD16 / Creative Vibra16XV ❇ 8.4GiB Quantum Fireball ❇ Win95 OSR2 Plus!

Reply 2 of 9, by nuno14272

Posted on 
User metadata
Rank Member
Rank
Member

Test the voltage reulator with a multimeter to see if its outputting stable 3,5v !!!

Seems to me 5V is ok, but not 3.5.

1| 386DX40
2| P200mmx, Voodoo 1
3| PIII-450, Voodoo 3 3000

Reply 3 of 9, by KayTeeFive

Posted on 
User metadata
Rank Newbie
Rank
Newbie
dominusprog wrote on 2026-06-24, 21:50:

Use contact cleaner on the CPU, cache and BIOS sockets. Also, put some thermal grease under the voltage regulator.

Unfortunately, didn't help. But now it shines 😀

The attachment photo_2026-06-26_01-12-52.jpg is no longer available
nuno14272 wrote on 2026-06-25, 14:15:

Test the voltage reulator with a multimeter to see if its outputting stable 3,5v !!!

Seems to me 5V is ok, but not 3.5.

3.5 was verified - it's good.

I found that the pins from RGB cooler headers happen to be exactly the same diameter as CPU pins, so I used them for testing.
I inserted a test pin into each socket contact and compared the resistance/contact tension while seating it. As a reference, I used pins that are never connected on the CPU.
Based on this, I'm assuming the socket itself is faulty — the contact pressure on certain pins is insufficient.
That might actually explain why the 5V Cyrix Cx486DX2-80 CPU, which previously had bent pins, works fine in this socket.
Now I'm wondering whether I should replace the socket or just live with it as it is.
I can handle the rework, but I've never replaced a CPU socket before. I do have a spare motherboard to practice on, but I'm afraid of damaging the pads on this board.

P.S. This socket issue was already present when I first tested the motherboard with the preinstalled 3.45V Cyrix Cx486DX2-v80GP CPU. At the time, I thought the CPU itself was dead.

Reply 4 of 9, by rasz_pl

Posted on 
User metadata
Rank l33t
Rank
l33t

this type of socket is on the easy side, this will give you some ideas 486SocketBlaster on cheap turn pin headers as a replacement

edit: I pasted name of project but forgot the link https://github.com/scrapcomputing/486SocketBlaster

Last edited by rasz_pl on 2026-06-26, 01:05. Edited 1 time in total.

https://github.com/raszpl/sigrok-disk FM/MFM/RLL decoder
https://github.com/raszpl/FIC-486-GAC-2-Cache-Module (AT&T Globalyst)
https://github.com/raszpl/386RC-16 ram board
https://github.com/raszpl/Zenith_ZBIOS Zenith Z-386 MFM-300 ZBIOS disassembly

Reply 5 of 9, by KayTeeFive

Posted on 
User metadata
Rank Newbie
Rank
Newbie
rasz_pl wrote on 2026-06-25, 23:20:

this type of socket is on the easy side, this will give you some ideas 486SocketBlaster on cheap turn pin headers as a replacement

Wow, it's totally cool idea. Thanks!

Reply 6 of 9, by KayTeeFive

Posted on 
User metadata
Rank Newbie
Rank
Newbie

After looking that project https://github.com/scrapcomputing/486SocketBlaster, I've measured CPU pins. Measurement are not precise, but...

Cx486DX2-80 5V CPU ~0.51-0.55 mm
Cx486DX2-v80GP 3.45V CPU ~0.47-0.49 mm

My test pin that was used for tension tests is ~0.57-0.6mm.

So, It's pin socket issue. So for now i will stay on Cx486DX2-80 5V variant, and for future, i will prepare that 486SocketBlaster.

Reply 7 of 9, by nuno14272

Posted on 
User metadata
Rank Member
Rank
Member

Whats with that resistors on the back ?
never seen resistors on the cpu on the back. Some sort of factory fix ?

1| 386DX40
2| P200mmx, Voodoo 1
3| PIII-450, Voodoo 3 3000

Reply 8 of 9, by KayTeeFive

Posted on 
User metadata
Rank Newbie
Rank
Newbie
nuno14272 wrote on 2026-06-26, 08:41:

Whats with that resistors on the back ?
never seen resistors on the cpu on the back. Some sort of factory fix ?

It wasn't me 100% 😀 Perhaps it's factory fix, soldering was clean. I've tested MB without them - it works, but i put them back.
On thermal camera chip MC74F245N has stable temperature around 43-45 degree of celsius.

MC74F245N: 150Ohm resistor on pins 10, 18
UM9515-10: 100Ohm resistor on pins 7, 8

Also, there are 4 SMD capacitors on the socket, they are sits on the CPU power line.

UPD:
After tracing chip tracks, it was found, that this resistor on MC74F245N affects JP15 "CPU Clock Selection".
There are 3 pins:
1 - 1.8v
2 - 0V, output, goes to the CPU CLK pin
3 - with resistor 1.55V, without resistor 1.8

The attachment Знімок_20260626_134600.png is no longer available

Reply 9 of 9, by rasz_pl

Posted on 
User metadata
Rank l33t
Rank
l33t

Resistor is damping interferences like reflections. Those resistors and capacitors are bodges for bad initial design.

https://github.com/raszpl/sigrok-disk FM/MFM/RLL decoder
https://github.com/raszpl/FIC-486-GAC-2-Cache-Module (AT&T Globalyst)
https://github.com/raszpl/386RC-16 ram board
https://github.com/raszpl/Zenith_ZBIOS Zenith Z-386 MFM-300 ZBIOS disassembly

Go to top of page Go to top of page
Back to General Old Hardware
Close window
Close window