h-a-l-9000: I do not know which pin on the Super I/O is the enable disable pin. And does the Southbridge control this? The datasheets for, both, the southbridge and superio are not available.

It's called reverse engineering. You can find out by measuring continuity from the jumper pin to the pins of the chip until you find zero ohms. Other sides of the jumper may connect directly to 5V or GND, so check this out first.

h-a-l-9000 wrote:

h-a-l-9000: I do not know which pin on the Super I/O is the enable disable pin. And does the Southbridge control this? The datasheets for, both, the southbridge and superio are not available.

It's called reverse engineering. You can find out by measuring continuity from the jumper pin to the pins of the chip until you find zero ohms. Other sides of the jumper may connect directly to 5V or GND, so check this out first.

I am confused by your use of the term 'jumper'. This motherboard does not contain a jumper in the traditional sense of the word for enable/disable of the Super I/O.

The ISA card does contain jumpers? But forget it...

Oh, yes, I wasn't thinking about the ISA card anymore for some reason. It does have floppy enable/disable, as well as serial, game, LPT, and IDE enable/disable. The issue is that on the motherboard, the BIOS is controlling enable/disable. So perhaps I can copy what the ISA card is doing and force the floppy to enabled with external hardware on the motherboard?

GPA: The next time I decide to tackle this board, I will do a simple check of the few diodes, SMD caps, resistors, and resistor packs. This will not take long. I will also remove the serial IC, check for FDD, then solder on a new serial IC. All the electrolytic caps have been replaced and there aren't any through hole resistors. I may also force the Super I/O to enabled via hardwire intervention by copying the ISA card's connection.

I won't be doing this today, but will report back when I have.

But the idea of tracing the ISA card's jumpers is good. I'm pretty sure the enable/disable jumpers are simply putting a pull up or pulldown resistor. Maybe those pins are in a floating state at the moment on the motherboard because of some component that failed.

Indeed - some component in the southbridge, which is back to my original suspicion. I believe the BIOS controls the pull-ups/downs on the southbridge.

This is what I did tonight:

1) checked all SMD capacitors for a short
2) checked all SMD resistors for a short and for open circuit
3) checked all resistor packs around the Super I/O and floppy connector
4) checked all diodes and inductors
5) ensured that the Super I/O's floppy enable pin (pin 99) is pulled high (5v) through a 4.7 Kohm resistor. This is actually done via a resistor pack and is hardwired for enabled.
6) ensured that GND is connected to the Super I/O. There are 4 locations
7) ensured that 5V is connected to the Super I/O. There are 2 locations
8) re-verified that all appropriate pins on the Super I/O connect to the ISA bus. There are 33 connections on the defective motherboard and on the Super I/O ISA card.
9) ensure that all pins of the floppy cable header are connected to the Super I/O.

I suppose all I have left to do is cut the serial IC off, UM8667. I was hoping to avoid that, but I am running out of ideas. Perhaps I should determine if the Super I/O is receiving a clock. I think it is pin 105.

The issues has been resolved. Onboard floppy access is now working fine.

So what solved it?

yawetaG wrote:

So what solved it?

Such a long and interesting topic can't just end like this!
Unless it was just a faulty floppy cable or something...

feipoa wrote:

The issues has been resolved. Onboard floppy access is now working fine.

congratulations!

The 8667?

feipoa wrote:

The issues has been resolved. Onboard floppy access is now working fine.

I tried to decide between these 2 posts:

1: https://www.youtube.com/watch?v=Z0YIJQ1jgEI
2: http://www.urbandictionary.com/define.php?ter … e&defid=1338492

Sorry, I came down with my first cold/flu/whatever sickness in about 4 years and had to disappear into bed.

I believe the issue was directly related to the Super I/O chip. When I went to solder the new chip on, I ensured that each pin was touching the solder pad by verifying connectivity on the ISA slot. However, for one or two pins, the slightest pressure of the multimeter probe on the Super I/O pins was causing it to make contact, even though there was no visible deflection in the pins. This was very difficult to troubleshoot. But alas, it is working fine now with the Super I/O replacement. I'm glad I did not remove the serial IC or replace the southbridge. The serial ports also work fine now.

I will probably go around to my stack of 486 motherboards and replace the Super I/O or serial IC's for those which are defective.

If a Super I/O can spontaneously fail like that, I assume many other IC's can as well.

I would also like to point out that there were two pins on the Super I/O which, even when contacted, were not making contact with the ISA slot. No matter how much I pushed the pins onto the pad, the connections were not being made. For these two cases, I had to solder wires onto the Super I/O and route them to the ISA slot. I identified these early though, so it was not what was causing me issues.

So, it is possible that the original Super I/O was working fine and that there was a break somewhere along the path of the original traces. I will never know if it was a bad solder point/trace somewhere, or if it was the Super I/O circuit which went bad.

Like that soldering job, a masterpiece!

Looks like some nice drag soldering!

feipoa wrote:

So, it is possible that the original Super I/O was working fine and that there was a break somewhere along the path of the original traces. I will never know if it was a bad solder point/trace somewhere, or if it was the Super I/O circuit which went bad.

SuperIO_Wires.jpg

I will re-flow my Super IO chip and see if it helps.I think I am a good candidate for testing if it's a contact issue or a chip issue.
IF you want to give me more details on the broken traces, that will help me.

The possibly broken traces are the ones with the wires soldered to them.

feipoa wrote:

The possibly broken traces are the ones with the wires soldered to them.

I checked and both pins are making contact with the first ISA slot on my motherboard.
Reflowed the chip, still issue. Next is to swap the chip.