• Please review our updated Terms and Rules here

Tandy 1000SX + PII-151B + Hardware Mods = Almost working. Need input.

S

sorphin

Experienced Member
Joined
Feb 11, 2014
Messages
370
Location
Bay Area
So I have a T1000SX, and a PII-151B. Yes, I know other people have worked on this, but i'm willing to mod what it takes even if i have to patch the Tandy BIOS. I've worked a bit with 'lowen' and we've made some hardware mods, but i'm still at an impass, so if anyone has any ideas, I'll take them. here's where I am so far.

U22 Pin 131 which controls DMA/I (which on the SX and TL both) which in turn then controls FDCINTI, FDCDMRQ, and FDACKI is disconnected from U22 and that line is pulled LOW so that the PI-151B can control the lines. I've also built a mini circuit consisting of a 74LS02. The onboard FDC is removed. The circuit is comprised of taking the output of U51 Pin 37 (FDCCHPB (Chip Select)) Inverting it, and taking the output of U18 Pin 17 (BUFENB), putting them into the NOR gate, taking that output, inverting it, and applying that result to U40 Pin 19 (BUFENB/Chip Enable on the LS245) basically replacing the direct output from U18 Pin 17, so that Buffer enable is only active when the FDC select is NOT active and the buffer enable IS active. Basically, NOT(NOT (FDCCS) NOR BUFFENB).

The result? The Tandy BIOS seeks the drives during POST (albeit with a stutter), the PII-151B's bios seeks them, no problem... When trying to boot, it poll the drive (but ignores any disk in it).. and boots off the HDD. When in DOS. If I go to A or B (or use DRIVEPARM or driver.sys).. I wind up with General Failure, Not Ready on A usually, and B (the 3 1/2) gets a Sector Not Found or whatever.

Any ideas of what to check/do would be helpful. I have a couple good scopes, both svc manuals, logic probe, and an EE background, but this is over my head.

Thanks,
-Daniel
 
I wish I had advice for this situation - you've probably gone further than the majority would in order to get a 1.44mb drive to work on an SX. A far better solution would be to find an SL or TL that allows bios disable of onboard FDC.

Is it possible the problem is an IRQ or address conflict?
 
mojorific said:
I wish I had advice for this situation - you've probably gone further than the majority would in order to get a 1.44mb drive to work on an SX. A far better solution would be to find an SL or TL that allows bios disable of onboard FDC.

Is it possible the problem is an IRQ or address conflict?
Click to expand...

The SL And TL have their own issues. The SX has a sentimental attachment, that's why I'm working with it (and I don't back down from a challenge). As for the IRQ/Address.. It cant' be an IRQ issue, because IRQ 6 goes to the slot and to the FDC, and the SX has a dip switch to disable IRQ6 from being used by the FDC but still feeds it to the slot. The physical FDC controller chip has been removed from the board, so that should also eliminate any address contention.
 
sorphin said:
The SL And TL have their own issues. The SX has a sentimental attachment, that's why I'm working with it (and I don't back down from a challenge). As for the IRQ/Address.. It cant' be an IRQ issue, because IRQ 6 goes to the slot and to the FDC, and the SX has a dip switch to disable IRQ6 from being used by the FDC but still feeds it to the slot. The physical FDC controller chip has been removed from the board, so that should also eliminate any address contention.
Click to expand...

If you haven't already, run MSD and see where your IRQ's actually are with the PII-151B installed. Also, may have missed on this, but do you have a HD installed?
 
Agent Orange said:
If you haven't already, run MSD and see where your IRQ's actually are with the PII-151B installed. Also, may have missed on this, but do you have a HD installed?
Click to expand...

I've ran msd many times. The irqs line up with what I expect.. irq 5 is where the WD hdd controller is (dip switch moved video off of it), irq 6 is where the floppy controller is, irq 7 on parallel, irq 3 has the ne2k card..
 
Did the PII-151B work as a secondary floppy controller in the SX prior to the modifications, and is it addressed to work as a primary now?

It does seem like you might be into BIOS-patching territory, at this point.
 
sorphin said:
I've ran msd many times. The irqs line up with what I expect.. irq 5 is where the WD hdd controller is (dip switch moved video off of it), irq 6 is where the floppy controller is, irq 7 on parallel, irq 3 has the ne2k card..
Click to expand...

If IRQ5 presents a problem, I have a hardware hack for my 1000SX's WD controller which puts it on IRQ2. I've been able to run a ST-225 with that configuration since about 1987. Have since semi-retired the ST-225 in favor of a SCSI and/or the forum's own XT-IDE. If you would like a drawing of the mod I'd be more than glad to make available to you. Good luck with a very interesting project.
 
Cloudschatze said:
Did the PII-151B work as a secondary floppy controller in the SX prior to the modifications, and is it addressed to work as a primary now?

It does seem like you might be into BIOS-patching territory, at this point.
Click to expand...

The PII-151B isn't address changable to secondary.

Agent Orange said:
If IRQ5 presents a problem, I have a hardware hack for my 1000SX's WD controller which puts it on IRQ2. I've been able to run a ST-225 with that configuration since about 1987. Have since semi-retired the ST-225 in favor of a SCSI and/or the forum's own XT-IDE. If you would like a drawing of the mod I'd be more than glad to make available to you. Good luck with a very interesting project.
Click to expand...

Thanks.. No, IRQ 5 has been working fine. the HDD has had 0 issues. it's solely the floppy controller.

As I said, you can poll the drives from DOS, and it attempts to read the disks.. but fails (the drive lights come on, etc).. Oh, I forgot to mention, I also lifted the FDCACK line on suggestion from lowen before too, basically keeping the LightBlue and BigBlue away from the floppy controller, since they're not necessary for this..

With regards to the SL/TL btw.. yes, you can disable the onboard... via a bit in the planar... but you still wouldn't be able to _boot_ from the drives, which is my end goal. I'm wondering if maybe it's a DMA issue. I don't know.
 
If you really want to run this card inside your SX, you are going to have to disable the built-in floppy controller. Unfortunately, if this can be done at all, it will require cutting traces inside your machine.
 
Great Hierophant said:
If you really want to run this card inside your SX, you are going to have to disable the built-in floppy controller. Unfortunately, if this can be done at all, it will require cutting traces inside your machine.
Click to expand...

I already made hardware modifications, the FDC is Physically removed from it's socket, the pins referenced in the first post are _lifted_ and the going through that circuit appropriately. I have no problems cutting traces. The internal FDC is already 'disabled' (and removed). Here is what has been done.

"U22 Pin 131 which controls DMA/I (which on the SX and TL both) which in turn then controls FDCINTI, FDCDMRQ, and FDACKI is disconnected from U22 and that line is pulled LOW so that the PI-151B can control the lines. I've also built a mini circuit consisting of a 74LS02. The onboard FDC is removed. The circuit is comprised of taking the output of U51 Pin 37 (FDCCHPB (Chip Select)) Inverting it, and taking the output of U18 Pin 17 (BUFENB), putting them into the NOR gate, taking that output, inverting it, and applying that result to U40 Pin 19 (BUFENB/Chip Enable on the LS245) basically replacing the direct output from U18 Pin 17, so that Buffer enable is only active when the FDC select is NOT active and the buffer enable IS active. Basically, NOT(NOT (FDCCS) NOR BUFFENB)." FDCACK is intercepted. Because of the first mod, FDCINTI (IRQ6), FDCDMRQ are pulled low.

There are no built on FDC related lines accessing anything anymore. the PII-151B's lines are in play. If you'd like, I can draw a schematic showing what's what. If anyone has any _other_ things to check. Tell me and I'll check them. As I said, bios patching isn't out of the question either.
 
Let me add that the PII-151B that sorphin is attempting to run has been tested operational, albeit in a 486 system.

We've looked through the schematic several times, and the SX puts a lot of logic in the two 'blue' chips, including the FDC chip select signals.

The U40 he mentions is the motherboard device buffer, incidentally.
 
Well, of course the other thing you could do is modify the PII-151B to be on another port, use a different IRQ and DMA, and use the DOS driver to run it instead of the built-in BIOS. But you would have to derive the schematic for the 151B, and I'm not sure that would be very easy.

You might need to build/buy a bus extender card and clip in a logic analyzer to see what's going on. It's not going to be easy; the SX wasn't designed to allow the on-board FDC to be disabled, and it seems to be a hard thing to make that happen.
 
I guess then my options are to go it alone on the tandy or revert it back to stock and sell the card and the dual floppy (I have no other legacy systems). It's not an IRQ conflict btw. If I have to use the 'dos driver' (which would have to be the 2M thing, since there's no 'dos driver' for this card that i'm aware of otherwise), then that defeats the point, because i will have nothing to boot from if the HDD were to ever have issues. Oh well. Feel free to delete this thread, since I can't, since it no longer serves a purpose.
 
You must log in or register to reply here.
Back
Top