VOGONS


Intel TC430HX Socket 7 motherboard repair

Topic actions

First post, by Kahenraz

User metadata
Rank l33t
Rank
l33t

I picked this motherboard up on eBay to add to my collection, which has only a few Socket 7 boards. I thought that it was particularly interesting on that it has an onboard S3 ViRGE, as well an an integrated Yamaha YMF-701B-S with OPL3 and General MIDI YMF-704-S. This makes for a great motherboard with everything you would need for a fully-featured DOS experience, with everything already integrated.

Unfortunately, the seller shipped it without removing the heatsink from the CPU socket. It detached in shipping and bounced around, knocking off a couple SMD components.

I've had this in my repair bin for a while now, and I think it's time to see if it can be repaired. Note that in the photos, the red stickers are there to help me locate damage as I find it, in the event that I set it aside for later.

The attachment 20221027_233751_resize_79.jpg is no longer available
The attachment 20221027_234856_resize_34.jpg is no longer available
The attachment 20221027_234911.jpg is no longer available
The attachment 20221027_233812_resize_55.jpg is no longer available
The attachment 20221027_233818_resize_88.jpg is no longer available

Reply 1 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

Some parts that, from a distance, looked like they might also be damaged, but were revealed to be just a little off-kilter when viewed under a microscope.

The attachment 20221028_000904.jpg is no longer available
The attachment 20221028_001000.jpg is no longer available

Reply 2 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

These first few SMD capacitors are located adjacent the CPU cache. I removed one of them to measure, and found it to be 22 pF. It's a 0603 capacitor (6mm x 3mm), quite small, but not the smallest I've replaced. It was large enough that I could still hand solder with a fine tip, rather than needing to use solder paste and hot air.

I placed a layer of tacky flux, positioned the capacitor, and attached it from one side, then the other. To ensure a good connection, I apply flux on top of the capacitor afterwards, and reflow the solder again. This is to make sure that the joint isn't spoiled by any oxide on the SMD part. It's harder to do this in the beginning, because the part is so small, as it tends to stick to the flux applicator, and get carried away. Likewise, it's difficult to place if it's covered in flux, and it wants to stick to the tweezers.

I only took photos of the first two capacitors I replaced here, as the others are exactly the same. See if you can tell which ones are new from the final photo. (Hint: it's the four shiny ones.)

Also see my thumb, provided for scale.

The attachment retouch_1666939210370.JPEG is no longer available
The attachment 20221028_002531.jpg is no longer available
The attachment 20221028_015147.jpg is no longer available
The attachment 20221028_020510.jpg is no longer available
The attachment 20221028_022104.jpg is no longer available
Last edited by Kahenraz on 2022-10-28, 06:47. Edited 2 times in total.

Reply 3 of 78, by TrashPanda

User metadata
Rank l33t
Rank
l33t

Im following this with great interest, always like to see old hardware repaired and bought back to life and a nice integrated HX board is worth the effort.

Reply 4 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

I've seen this board crop up a few times, but never with both the integrated video and General MIDI option. It must have been very expensive at the time, if it's this hard to find one among those that still exist.

Reply 5 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

Soldering an SMD resistor looks exactly like soldering an SMD capacitor, only with different colors. And they're usually shorter. Here are some photos of before and after, as well as some more entertaining compositions to make things a little bit more interesting.

This resistor was measured at 1K Ohms. I felt a little guilty separating them, as they seemed to have grown attached.

The attachment 20221028_024835.jpg is no longer available
The attachment 20221028_025024.jpg is no longer available
The attachment 20221028_031613.jpg is no longer available
The attachment 20221028_032644.jpg is no longer available
The attachment 20221028_033034.jpg is no longer available
Last edited by Kahenraz on 2022-10-28, 07:43. Edited 1 time in total.

Reply 6 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

Here are a few more, for entertainment. I also included an uncropped photo, so you can see how I take photos through my microscope by holding my camera up to the lens. This is why they're a bit washed out and not always in focus. My microscope does not have a camera attachment, so this is the only way that I can capture a picture.

The attachment 20221028_030321.jpg is no longer available
The attachment 20221028_025405.jpg is no longer available
The attachment 20221028_030846(0).jpg is no longer available
The attachment 20221028_030839_resize_67.jpg is no longer available

Reply 7 of 78, by TrashPanda

User metadata
Rank l33t
Rank
l33t

I wonder how hard it would be to rig up a camera attachment using a USB camera.

Reply 8 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

I've been wanting to start up a YouTube channel and also stream my repairs for some time. I don't yet have the proper equipment for it, and would definitely need a new microscope so that other people can see what I'm doing. I'll get there some day.

Bad news with the motherboard. It can't seem to get past POST code E6 EA, which, according to Mr. BIOS, the flash memory read OK and it's bootstrapping, but it gets stuck from there. The fact that it gets this far, means that the board isn't dead. It should just be a matter of diagnosing what's wrong with it. I don't think it's related to any of my repairs.

http://mrbios.com/techsupport/award/postcodes.htm

E6: BIOS ROM is OK
EA: Read in the bootstrap code

I have tried clearing the CMOS, lowering the CPU multipliers, and using an external graphics card in a PCI slot. I have a PS/2 keyboard connected and a CR2032 battery inserted. I also tried both a 2.8V and 3.3V 200Mhz (66 * 3) CPU.

If I remove one of the memory sticks, the system beeps a very unhappy beep. So I'm guessing that the memory needs to be installed in pairs. It also seems to be alright with the two sticks, I think. It doesn't beep at all with two.

I'll keep brainstorming, and would welcome any suggestions in the meantime.

The attachment 20221028_040758.jpg is no longer available
Last edited by Kahenraz on 2022-10-28, 08:35. Edited 2 times in total.

Reply 9 of 78, by TrashPanda

User metadata
Rank l33t
Rank
l33t

You might want to try flashing a new bios chips with the latest bios for the board and using that to see if the BIOS is actually ok, sometimes they can read as ok but have corruption later in the bios code that doesn't show till post.

****Scratch that .. Had a second closer look and I cant see a BIOS Chips on that board so I guess its like a few other silly Intel boards that has it integrated into the chipset, that does make it difficult to diagnose BIOS issues.

Possibly the little rectangular Intel branded IC near the first two ISA slots is the Sram module, might be worth looking at that if it is to make sure its ok and getting correct voltages.

Last edited by TrashPanda on 2022-10-28, 08:29. Edited 1 time in total.

Reply 10 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

Intel motherboards from this time use a really annoying surface mount flash chip that I can't remove to flash elsewhere. The board would have to be capable of blindly flashing from a floppy disk.

The attachment 20221028_042601.jpg is no longer available

Reply 11 of 78, by TrashPanda

User metadata
Rank l33t
Rank
l33t

The little pads alongside of it are for testing I assume .. perhaps they are also used to initially flash the IC.

Reply 12 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

Or for an alternative flash chip, in the event they need to source from a different supplier.

Reply 13 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

I've made some progress. I found two severed traces on the back side of the motherboard. They are extremely small and difficult to photograph. I scraped away some of the solder mask and wet the area with alcohol to try and increase the contrast a bit for the photo. I tested these with my multimeter; both are definitely cut.

The attachment 20221028_050159.jpg is no longer available
The attachment 20221028_050244.jpg is no longer available
The attachment 20221028_050832(0).jpg is no longer available
The attachment 20221028_051818.jpg is no longer available
Last edited by Kahenraz on 2023-03-02, 14:47. Edited 1 time in total.

Reply 14 of 78, by rasz_pl

User metadata
Rank l33t
Rank
l33t

>E6: BIOS ROM is OK

with any luck it read all the flash and computed checksum, but its possible it just checks the magic number somewhere.

as for flash chip, $3 "TSOP48 to DIP48 0.5mm pitch Interposer board pcb Board Adapter Plate" to make an adapter and flash it in something else. But I would wait with it seeing you are still finding damage 😀

https://github.com/raszpl/FIC-486-GAC-2-Cache-Module for AT&T Globalyst
https://github.com/raszpl/386RC-16 memory board
https://github.com/raszpl/440BX Reference Design adapted to Kicad
https://github.com/raszpl/Zenith_ZBIOS MFM-300 Monitor

Reply 15 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

I cleaned up the traces with the tip of an X-Acto knife, repositioned them carefully, then mashed them into place with the tip of my multimeter probe, pressing the surface flat, and sculpting the copper.

I wasn't confident with what material was left of the top trace, so I reinforced it with a piece of solder wick. I then covered coated the area in flux and set it all in place with a bit of solder.

Here are a few photos of the procedure, with a final one that emphasizes the scale. These kinds of repairs are very challenging to even focus on with my microscope.

The attachment 20221028_053256.jpg is no longer available
The attachment 20221028_053457.jpg is no longer available
The attachment 20221028_054052.jpg is no longer available
The attachment 20221028_055121.jpg is no longer available
The attachment 20221028_055458.jpg is no longer available

Reply 16 of 78, by Kahenraz

User metadata
Rank l33t
Rank
l33t

That was it! Another successful repair. 😀

Also, 🤣? It looks like this S3 chip suffers from the overbright bug. That is terribly anticlimactic.

S3 AGP Cards (and possibly others) Too Bright

The attachment 20221028_061137.jpg is no longer available

Reply 17 of 78, by Babasha

User metadata
Rank Oldbie
Rank
Oldbie

Welcome to the TC430HX club! 😀

Need help? Begin with photo and model of your hardware 😉

Reply 18 of 78, by TrashPanda

User metadata
Rank l33t
Rank
l33t
Kahenraz wrote on 2022-10-28, 10:15:
That was it! Another successful repair. :) […]
Show full quote

That was it! Another successful repair. 😀

Also, 🤣? It looks like this S3 chip suffers from the overbright bug. That is terribly anticlimactic.

S3 AGP Cards (and possibly others) Too Bright

20221028_061137.jpg

its a shame there isnt a way to update the S3 bios to one of the models that doesn't have the gamma bug.

Reply 19 of 78, by Babasha

User metadata
Rank Oldbie
Rank
Oldbie

There are Mr.BIOS'es for this motherboard (with their pros and cons)

Need help? Begin with photo and model of your hardware 😉