Toshiba T3100 BIOS help

[silence012]

Hi all,

I am trying to fix a Toshiba T3100. It shows error code AA on the screen which, according to maintenance manual, is "DMAC #2 data bus error".



Toshiba laptops also output error codes on LPT1 port, and with "Printer Port LED" I am getting the same error.

As I am not familiar with this error, I am trying to locate in the BIOS rom part of the code where this is actually output to parallel port, and start troubleshooting from there. The idea is to see what BIOS checks before it outputs the code, and why is it not happy However, I cannot locate it, although I found lot of other places where some other codes are output to parallel port. For example:



I was able to locate in the rom, where the error message is stored, but I cannot find where it is used



I would appreciate any help in finding the code snipped, or any kind of explanation about the error, or an idea what bios could actually test for DMAC.

Thanks in advance!

Attached is a interleaved rom from both BIOS roms.

 

[modem7]

DMAC = DMA controller.

In the T3100, there are two of those chips.

From what I have seen, the typical test that the POST does of the 8237/82C37 chip is a read/write test of the chip's first 8 registers (using various data values).

The DMAC #1 test passed, and then the DMAC #2 test failed.

I suggest that you do a close-up look at the two 82C37 chips, and the area surrounding them. I see that the battery connector is close to the chips. Maybe a leaking battery has damaged the circuit board.

 

[silence012]

@modem Thanks for the feedback.

Do you know if DMACs have some well known I/O addresses? Like 64h for 8042, for example.

There is no visible damage to any of the chips, but if I remember correctly the previous owner build his own "battery pack" using AA batteries, so could be that higher voltage was applied or something similar. But I guess in that case RTC chip would failed.

 

[silence012]

I found it in Toshiba T1200 flyer .

 

[silence012]

It would be also useful if anyone has the datasheet for TC17G032AF which is gate array used for 82C37 and data bus as well. I cannot find it on the internet.

 

[modem7]

Do you know if DMACs have some well known I/O addresses? Like 64h for 8042, for example.

Of course, it is up to the motherboard maker as to which I/O ports are used for what.

For compatibility reasons, Toshiba may have used the same I/O port ranges as used for the IBM AT:
0000-001F used by DMA 1 (first 8237)
00C0-00DF used by DMA 2 (second 8237)

A breakdown of those two port ranges is at [here].

There is no visible damage to any of the chips, but if I remember correctly the previous owner build his own "battery pack" using AA batteries, so could be that higher voltage was applied or something similar. But I guess in that case RTC chip would failed.

The battery that I was referring to is the CMOS backup battery. There is a tendency for those to leak, with the 'leakage' making its way down the cable to the system board, resulting in damage to the battery connector there and the surrounding part of the system board. The 'leakage' can open up traces on the system board (with the traces still looking as intact). It happened to various degrees on all three of the Toshiba T1100 laptops that I owned.

It would be also useful if anyone has the datasheet for TC17G032AF which is gate array used for 82C37 and data bus as well. I cannot find it on the internet.

The TC prefix on the part number is what Toshiba use. Gate arrays are programmed/customised devices. I am guessing that TC17G032AF is the unique number that Toshiba had stamped on the gate array after it was programmed/customised for Toshiba. I presume that you saw appendix C in the T3100 Maintenance Manual.

 

[silence012]

@modem7 I've already found few of "out" instructions in the rom that points to some of the I/O addresses in these range, and assuming that it is the same for T3100 as they used for T1200, probably for IBM compatibility as you've mentioned. This breakdown is really useful!

I am aware that CMOS batteries leak through cables, saw it on many vintage laptops that I fixed so far but I don't see any signs of leakage anywhere. The board looks really clean. I bought this computer more than year ago, and if I remember correctly the previous owner removed the original CMOS battery and tried to use either 3xAA or 4xAA battery instead. So could be that voltage was too high and that this burned some chips. I have another T3100, so I can try to replace 8237 chips and see fi this fix the issue, but this would be an almost blind guessing approach
In my experience, it is more likely that some of the logic (74xx) chips will die before others but you never know.

For TC, you are right there. I saw appendix C, and it seems that this is the only info that we have. But should be good enough. The annoying thing with this chips is that pins a re too small and attaching lids from logic analyzer is almost impossible

Thanks once more...and I definitely owe you a beer

 

[silence012]

It seems that DMA controller chip itself was broken. I pull one from working machine and put it on the board and it fixed the problem. Now computer boots properly.

However, I am still missing one chip. I couldn't find that anyone is selling it online. If anyone in the Europe had some spare parts 3100 and willing to sell me a chip I would really appreciate.

The chip is Mitsubishi 82C37AFP4. I think is is SOP-40.

Cheers,
Branko

 

[silence012]

BTW, if anyone else experience the similar issue:
DMAC#1 is chip IC53
DMAC#2 is chip IC49

 

[Istarian]

The TC prefix on the part number is what Toshiba use. Gate arrays are programmed/customised devices. I am guessing that TC17G032AF is the unique number that Toshiba had stamped on the gate array after it was programmed/customised for Toshiba. I presume that you saw appendix C in the T3100 Maintenance Manual.

Nope.

These components are Toshiba manufactured gate array chips of some sort and are not unique to their laptops (or this particular line). E.g. VTech used them in the Socrates "game console" to generate the video output. In that case the chip is also marked as follows:

VTEL 27-0769
TC17G032AF
0248 8835EAI/JAPAN

https://upload.wikimedia.org/wikipedia/commons/2/23/VTech-Socrates-Motherboard-Flat.jpg

According to the document linked below, The TC17G032(AF) is a member of the TC17G series/family of gate arrays with a "Gate Complexity" of 3,200 gates.

https://tvsat.com.pl/PDF/T/TC17Gxx_tos.pdf

The following shot (of a Toshiba 3100 motherboard) shows that there are four such gate array in use, all TC17G032AF parts with slightly different markings.
0115 8707E
0116 8711EAI
0117 8711EAI
0118 8705E