Baffling High Density Floppy/Controller Issue

[Chuck(G)]

Well, you have two ways out on this one. I'm assuming that IC3 is socketed. You could change the ROM address to D800 by lifting pin 12 of the IC3 (you should tie it to pin 9 or Vcc, but for a trial, you can leave it floating).

Alternatively, you can burn the FDC BIOS into the XT-IDE EEPROM spare space (according to the documentation, only 4K is actually used).

 

[nc_mike]

Well, you have two ways out on this one. I'm assuming that IC3 is socketed. You could change the ROM address to D800 by lifting pin 12 of the IC3 (you should tie it to pin 9 or Vcc, but for a trial, you can leave it floating).

Alternatively, you can burn the FDC BIOS into the XT-IDE EEPROM spare space (according to the documentation, only 4K is actually used).

OK, I lifted pin 12 of IC3 and yes, the floppy disk controller BIOS loaded successfully first followed by the successful loading of the XT-IDE BIOS - and then I tried using a 1.44MB disk and it read fine. Thank you!

OK, so what happens if I don't tie pin 12 to pin 9 or Vcc (is the latter a power supply point?). I don't usually mess with soldering but if I have to tie pin 12, how do I do that - use a soldering iron and solder a wire on the back of the board from the base of pin 12 to the base of pin 9?

Getting close!

Regards,
Mike

 

[Chuck(G)]

OK, so what happens if I don't tie pin 12 to pin 9 or Vcc (is the latter a power supply point?). I don't usually mess with soldering but if I have to tie pin 12, how do I do that - use a soldering iron and solder a wire on the back of the board from the base of pin 12 to the base of pin 9?

Well, letting the pin "float" is lousy engineering practice and it might be susceptible to stray noise. But nothing really awful will happen under normal circumstances.

You needn't solder the stuff. For example, I will sometimes use wire wrap (AWG 30) wire wrapped around a a couple of pins.

 

[nc_mike]

Well, letting the pin "float" is lousy engineering practice and it might be susceptible to stray noise. But nothing really awful will happen under normal circumstances.

You needn't solder the stuff. For example, I will sometimes use wire wrap (AWG 30) wire wrapped around a a couple of pins.

Ah, that's better. I'll wire the two pins - just wanted to know if leaving it untied while I get the wire would cause any harm. I'll fix that today.

I did try the merged BIOS route and got pretty far until I tried to flash the XT-IDE BIOS as the final step. James Pearce gave me a handy ROMDUMP utility to easily gra b a copy the HD floppy disk controler's BIOS and write it to a .bin file which I trimmed with my hex editor down to a 4K chunk then merged with IDE_XT.BIN from the xtide-r566 zip resulting in a new BIOS bin file (NEWROM.BIN). When I went to flash the XT-IDE's BIOS (e.g. FLASH NEWROM.BIN C800) The erase operation appeared to be progressing for a few minutes then I got the following erase timeout failure:

---------------------------------------------------------------------
A:\>flash newrom.bin C800
Calibrating delay loops... calibration factor is 13
Opening ROM image... OK
Unable to determine flash chip type. Attempting JDEC programming.
Reading 8464 bytes read OK.
Erasing ................................... **TIMEOUT**
Expected FFh but found 0055h Erase operation FAILED

If BIOS images currently in use have been updated, it is HIGHLY
recommended that the computer is no restarted.

Press any key to return to DOS, or CTRL-ALT-DEL to restart.
--------------------------------------------------------------------

That looks like maybe the 8464 is not the full ROM memory range? These are lo-tech XT-IDE 8-bit cards I got off of eBay; I didn't built them myself, so I don't know if the problem is maybe that its a different ROM chip than what James supplies or some other cause. The ROM chip on the boards I have is an SST 27SF010 where Lo-Tech supplies an SST SST39SF010A. Maybe all I need to do if I want to flash the other XT-IDE card I have is to get a new BIOS chip directly from James?

Regards,
Mike

 

[Chuck(G)]

Well, the 27SF010 has a different pinout from the 39SF010A, as well as different programming times.

This is one you'll have to take up with James, I'm afraid.

 

[nc_mike]

Well, the 27SF010 has a different pinout from the 39SF010A, as well as different programming times.

This is one you'll have to take up with James, I'm afraid.

Chuck, thanks for all of your help. The pin solution is just fine; I'll work with James on the BIOS chip if only because now I'm curious and have two other XT-IDE boards to mess with. Thanks for hanging in there with me - it was truly baffling (to me anyhow) for a few weeks now and you've solved the mystery.

Thanks again,
Mike