VOGONS


First post, by rpocc

User metadata
Rank Newbie
Rank
Newbie

Hi to everyone.

I have one PCChips M396F V2.7 motherboard, based on SARC RC2016A5 chipset for 386SX. The battery is removed. I test it as well as my other MBs on an open stand without enclosure if that counts. Being otherwise clean It has one annoying malfunction:

Rarely it may coldstart normally but most of the time it requires holding reset shorted for 1-2 seconds (sometimes couple of times, or few short presses), and even after that sometimes it passes the VGA BIOS entry point and proceeds to boot, but sometimes it types few characters or lines from the Trident 9000 boot screen and then just hangs up. Without holding reset it doesn't even start outputting any POST codes. My card shows presence of all voltages, inactive reset, inactive IRDY, active CLK and FRAME.

This board clocks from a standard 14.31818 crystal and achieves 40 MHz with PLL chip.
So far I've tried: another crystal, another crystal overtones bypass capacitors of 22, 30 and 47 pf (original are disk 30 pf), another PLL, another keyboard BIOS (it reboots just before memory test with not-6042 keyboard BIOS but the instable boot remains the same), putting the video card and POST card into different ISA slots, testing continuity and voltage between CPU reset pin and the chipset, continuity between reset jumper and ground, chipset, continuity between powergood and chipset. Found absolutely nothing.

Didn't recapped the MB though it shows total of 100 uF at 5V and 3x10 uF at other power rails, having total 13x 10uF capacitors, so at least the initial values are normal and all of them appear to have no visible signs of failure.

In theory I could put some MCU like ATMEGA13A onto reset pin and put a simple timer-driven forced reset as a workaround but there's no simple pattern providing algorithm for guaranteed start without handing at one or another moment.As usual there's a simple choice: to forget and tear it down for parts (Not so much of universally useful parts) for other MBs or keep trying searching the cause of the problem but I'm out of ideas. Don't have another SARC chipset and really wouldn't liek the idea of replacing the CPU.

Any ideas, why motherboard could act like that?

m396f-2-7-car-632d595415792793370795.jpg

Reply 1 of 5, by ediflorianUS

User metadata
Rank Member
Rank
Member

Does that happen when you have a battery in it? (did you properly clean the board? alcohol or contact/brake/multipurpuse cleaner ?)
did you check with a magnifiyng glass for cracked solder spots/joints?

My 80486-S i66 Project

Reply 2 of 5, by rpocc

User metadata
Rank Newbie
Rank
Newbie
ediflorianUS wrote on 2023-11-14, 15:04:

Does that happen when you have a battery in it? (did you properly clean the board? alcohol or contact/brake/multipurpuse cleaner ?)
did you check with a magnifiyng glass for cracked solder spots/joints?

Although I’m not so bad at cleaning leakage, broken tracks, etc and have a whole barrel of alcohol in my bathroom as well as vinegar acid, acetone, and other happy chemicals, this one wasn’t even cleaned by me and the area around the battery looks quite nice like it was removed before it started leaking. I’ll try to start it up with battery put on but usually the very absence of battery isn’t an issue unless it’s Dallas.

Reply 3 of 5, by ediflorianUS

User metadata
Rank Member
Rank
Member
rpocc wrote on 2023-11-14, 15:20:
ediflorianUS wrote on 2023-11-14, 15:04:

Does that happen when you have a battery in it? (did you properly clean the board? alcohol or contact/brake/multipurpuse cleaner ?)
did you check with a magnifiyng glass for cracked solder spots/joints?

Although I’m not so bad at cleaning leakage, broken tracks, etc and have a whole barrel of alcohol in my bathroom as well as vinegar acid, acetone, and other happy chemicals, this one wasn’t even cleaned by me and the area around the battery looks quite nice like it was removed before it started leaking. I’ll try to start it up with battery put on but usually the very absence of battery isn’t an issue unless it’s Dallas.

I am just saying.... check evrything , my Toshiba(IBM spec) would simply "not compute" above a _ without a 3.6v li-tion charging batt inserted... (it's a 286).

Before salvageing for parts and recaping , try a hot air reflow of evrything.(after cleaning - if not functioning properly).
A 5 min burst in back should do fine , and second run on the front. (always moving the nozle).

My 80486-S i66 Project

Reply 4 of 5, by rpocc

User metadata
Rank Newbie
Rank
Newbie
ediflorianUS wrote on 2023-11-14, 15:24:
I am just saying.... check evrything , my Toshiba(IBM spec) would simply "not compute" above a _ without a 3.6v li-tion chargin […]
Show full quote
rpocc wrote on 2023-11-14, 15:20:
ediflorianUS wrote on 2023-11-14, 15:04:

Does that happen when you have a battery in it? (did you properly clean the board? alcohol or contact/brake/multipurpuse cleaner ?)
did you check with a magnifiyng glass for cracked solder spots/joints?

Although I’m not so bad at cleaning leakage, broken tracks, etc and have a whole barrel of alcohol in my bathroom as well as vinegar acid, acetone, and other happy chemicals, this one wasn’t even cleaned by me and the area around the battery looks quite nice like it was removed before it started leaking. I’ll try to start it up with battery put on but usually the very absence of battery isn’t an issue unless it’s Dallas.

I am just saying.... check evrything , my Toshiba(IBM spec) would simply "not compute" above a _ without a 3.6v li-tion charging batt inserted... (it's a 286).

Before salvageing for parts and recaping , try a hot air reflow of evrything.(after cleaning - if not functioning properly).
A 5 min burst in back should do fine , and second run on the front. (always moving the nozle).

Will try that and report, thanks.

Reply 5 of 5, by rpocc

User metadata
Rank Newbie
Rank
Newbie
ediflorianUS wrote on 2023-11-14, 15:24:

I am just saying.... check evrything , my Toshiba(IBM spec) would simply "not compute" above a _ without a 3.6v li-tion charging batt inserted... (it's a 286).

Before salvageing for parts and recaping , try a hot air reflow of evrything.(after cleaning - if not functioning properly).
A 5 min burst in back should do fine , and second run on the front. (always moving the nozle).

I’ m a bit surprised but simple expose to 330 degrees C at both sides raised percentage of successful runs from 5% to over 90%, but yet didn’t solved the issue completely. Nevertheless, thanks for the advise. I’m a bit confused about what could cause this: microcracks, whiskers, oxidation or something else.