VOGONS


First post, by clb

User metadata
Rank Newbie
Rank
Newbie

Hi all,

I have this old multi-IO controller that has a floppy disk connector on it. Trying to use this with a 486 PC to connect a 5.25" floppy drive or a 3.5" floppy drive, but whichever I try (tried 4x 5.25" drives and 2x 3.5" drives, of which the 3.5" drives at least are tested on other systems to be known working), I get a DOS boot error

Floppy disk(s) fail (C0)

BIOS has a drive setting for both primary and secondary floppy disks:

360KB 5.25"
1.2MB 5.25"
720KB 3.5"
1.44MB 3.5"

but adjusting those settings does not work (e.g. the 3.5" drives are 1.44MB, tried them as either primary or secondary - also trying all combinations did not work)

I can't quite figure out if there are any other settings that would affect this in BIOS, so I'm thinking that either the IO controller card is dead, or it has some jumper configuration I am missing.

The IO controller has markings

UMC
UM82C862F
9203-AA
NB2166

Before I rule the card out as dead, any chance someone might have the manual or reference for the jumper settings? See photos in attachments.

Thanks!

Attachments

  • IMG_2313.jpg
    Filename
    IMG_2313.jpg
    File size
    112.74 KiB
    Views
    397 views
    File license
    Public domain
  • IMG_2312.jpg
    Filename
    IMG_2312.jpg
    File size
    56.69 KiB
    Views
    397 views
    File license
    Public domain
  • IMG_2311.jpg
    Filename
    IMG_2311.jpg
    File size
    132.86 KiB
    Views
    397 views
    File license
    Public domain

Reply 1 of 13, by weedeewee

User metadata
Rank Oldbie
Rank
Oldbie

Looking at the photo at the block of pins where the jumpers are located, my first impression is, that whole block should have more jumpers.
It seems like you're missing about 7 jumpers.
just add more the same way the others like the others are.

You should end up with something like this

||||||||||.||
|||||||||||||
..........|..

where
|
|
is a jumper.

and a quick googling reveals that to be true.

Right to repair is fundamental. You own it, you're allowed to fix it.
How To Ask Questions The Smart Way
Do not ask Why !

Reply 2 of 13, by Horun

User metadata
Rank l33t
Rank
l33t

Agree weedeewee ! The jumpers for COM1 and COM2 could be the ones on left and why empty but that could mean they are "floating" and not disabled.

Attachments

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor.

Reply 3 of 13, by clb

User metadata
Rank Newbie
Rank
Newbie

Thanks for the suggestion! That is some sharp thinking - indeed it looks like the controller was missing jumpers. I populated the remaining row with jumpers, and just as if by magic, all my floppy drives are now working with the device.

Reply 5 of 13, by clb

User metadata
Rank Newbie
Rank
Newbie

I have been looking closer into this card, to help debugging some instabilities with audio playback on this system. (see Laser Squad audio problems with Orpheus and SB 64 AWE Gold )

One thing that System Information tool shows is the following hardware IRQ assignments:

IRQ3 COM2 BIOS
IRQ4 MOUSE CTMOUSE
IRQ5 LPT2 BIOS
IRQ6 FLOPPY BIOS
IRQ7 LPT1 UNKNOWN

See the attached screenshot hardware_IRQs.jpg.

That had prompted me to question whether there could be a IRQ conflict going on with the audio card and this multi-IO card. However there is no LPT2 on the system though, and audio card is set to IRQ5, so that might be a false positive message from Systeminfo (this tool is from Phil's Benchmark Suite)

However, just to rule everything out, I ended up deciphering what the different jumper settings all do on this card.

Captured the information that I was able to figure out on a Google spreadsheet at https://docs.google.com/spreadsheets/d/1xknh_ … dit?usp=sharing

Attached a screenshot of that sheet also in this post.

I was not able to figure out all of the jumper settings. If someone has a good hint, that would be welcome.
In particular, I don't know what J11, J12, JP19 and JP20 do, and the two jumpers that both affect Floppy disk controller working (J1 and J3) feels off. It is likely that one of those settings is for something else that I am unable to observe on this system.

Hopefully this helps anyone else who might have this card and is looking for information!

Attachments

Reply 6 of 13, by weedeewee

User metadata
Rank Oldbie
Rank
Oldbie

JP20 hdd led (either 1-2 or 3-4 can't tell from the photos)
JP19 IOCHRDY en/disable

which leaves JP11 and JP12...
if you don't observe it, one of them could be for setting the interrupt for printer port...

edit: comparing to this https://www.elhvb.com/iocard/DLU-270.txt , it seems to be either IDE AT/XT and something else to do with hdd controller.

Right to repair is fundamental. You own it, you're allowed to fix it.
How To Ask Questions The Smart Way
Do not ask Why !

Reply 7 of 13, by clb

User metadata
Rank Newbie
Rank
Newbie
weedeewee wrote on 2022-03-05, 14:04:

JP20 hdd led (either 1-2 or 3-4 can't tell from the photos)
JP19 IOCHRDY en/disable

Thanks, these make sense! Updated the sheet.

Btw do you know why there is an option to enable/disable IOCHRDY? My understanding is that if IOCHRDY is enabled, the PC can detect when it can issue a read or a write operation on the device, but I wonder why that would be a feature that needs an enable vs disable jumper option. (I suppose there must be a common incompatibility or a feature discrepancy between devices that necessitated being able to configure this)

weedeewee wrote on 2022-03-05, 14:04:

which leaves JP11 and JP12...
if you don't observe it, one of them could be for setting the interrupt for printer port...

There is a COM port listing utility that comes with CTMOUSE driver. I used that to see the COM settings and the COM IRQs, so J11 and J12 did not change those. Then used the System Information tool in Phil's benchmark suite to see the hardware IRQs for LPT ports, but those did not change with J11 and J12 either. (those settings were the ones I was most interested to find, since I was doubting that as the cause of audio issues)

weedeewee wrote on 2022-03-05, 14:04:

edit: comparing to this https://www.elhvb.com/iocard/DLU-270.txt , it seems to be either IDE AT/XT and something else to do with hdd controller.

I use a 4GB IDE Compact Flash as a hard disk on this computer. Toggling all combinations of J11 and J12 did not affect the hard drive at least, so if it is something to do with AT/XT IDE config, (e.g. a "XT fallback/emulation compatibility" related setting) then it is at least something that will at least not impair 4GB hard drives from working.

Reply 8 of 13, by weedeewee

User metadata
Rank Oldbie
Rank
Oldbie
clb wrote on 2022-03-05, 15:18:

There is a COM port listing utility that comes with CTMOUSE driver. I used that to see the COM settings and the COM IRQs, so J11 and J12 did not change those. Then used the System Information tool in Phil's benchmark suite to see the hardware IRQs for LPT ports, but those did not change with J11 and J12 either. (those settings were the ones I was most interested to find, since I was doubting that as the cause of audio issues)

from my own experience with testing programs...
NONE of them can truly determine the irqs in use unless there is a physical loopback possibility in the hardware,
either manually by a dongle on the com/lpt ports,
or automatically, as in some hardware parameter needs to be set to switch it to loopback and this requires the test program to be aware of the hardware it is testing, which most(all) are not .
They just display an assumption.

no idea on the why of having iochrdy on a jumper. never looked into it as far as i remember.

Right to repair is fundamental. You own it, you're allowed to fix it.
How To Ask Questions The Smart Way
Do not ask Why !

Reply 9 of 13, by clb

User metadata
Rank Newbie
Rank
Newbie

I also used Covox Speech Thing in Pinball Fantasies to test LPT, but I don't actually know if playing Covox audio requires using a hardware IRQ or not.

Not sure why that System Information tool would list IRQ 5 for LPT2. The DOS boot menu does not list LPT2 even to exist, so I am quite convinced that it is then a reporting fluke.

Reply 10 of 13, by weedeewee

User metadata
Rank Oldbie
Rank
Oldbie
clb wrote on 2022-03-05, 15:52:

I also used Covox Speech Thing in Pinball Fantasies to test LPT, but I don't actually know if playing Covox audio requires using a hardware IRQ or not.

Not sure why that System Information tool would list IRQ 5 for LPT2. The DOS boot menu does not list LPT2 even to exist, so I am quite convinced that it is then a reporting fluke.

covox speech thing using irq? I do not think so.

why sysinfo lists IRQ5 for lpt2... because it's a sort of kind of default. LPT1 irq7, com1 irq4, com2 irq3, com3 irq4, com4 irq3, ide1 irq14, ide2 irq15, etc...

Right to repair is fundamental. You own it, you're allowed to fix it.
How To Ask Questions The Smart Way
Do not ask Why !

Reply 11 of 13, by Horun

User metadata
Rank l33t
Rank
l33t
weedeewee wrote on 2022-03-05, 14:04:
JP20 hdd led (either 1-2 or 3-4 can't tell from the photos) JP19 IOCHRDY en/disable […]
Show full quote

JP20 hdd led (either 1-2 or 3-4 can't tell from the photos)
JP19 IOCHRDY en/disable

which leaves JP11 and JP12...
if you don't observe it, one of them could be for setting the interrupt for printer port...

edit: comparing to this https://www.elhvb.com/iocard/DLU-270.txt , it seems to be either IDE AT/XT and something else to do with hdd controller.

Yeah JP20 = appears 1 and 4 are both ground. 2 is + (counting from left looking straight at the card) and tied to R7 from what I can see... That allows 2 pin or 3 pin HDD LED connector to be used.
Usually on older ISA cards the IOCHRDY is default disabled, yes and I can see that JP19 has one trace going to pin 27 of the IDE connector to HDD IORDY signal.....so open is disabled ?? or is closed 🤣

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor.

Reply 12 of 13, by mkarcher

User metadata
Rank Oldbie
Rank
Oldbie
clb wrote on 2022-03-05, 15:18:
weedeewee wrote on 2022-03-05, 14:04:

JP20 hdd led (either 1-2 or 3-4 can't tell from the photos)
JP19 IOCHRDY en/disable

Btw do you know why there is an option to enable/disable IOCHRDY? My understanding is that if IOCHRDY is enabled, the PC can detect when it can issue a read or a write operation on the device, but I wonder why that would be a feature that needs an enable vs disable jumper option. (I suppose there must be a common incompatibility or a feature discrepancy between devices that necessitated being able to configure this)

It's from the wild west days of IDE before standardization: Some cards didn't forward the IOCHRDY signal from the disk to the ISA bus, and (nearly?) all drives were fast enough for PIO 0 access anyway. It seems some drives didn't properly leave IOCHRDY alone and disturbed bus operation (pull IOCHRDY low, and the ISA bus gets blocked completely; on classic PC/XT/AT computers, this will even block RAM refresh from happening). Most ISA IDE cards do not have a jumper, and installing a drive that uses the IOCHRDY pin improperly prevents the system from booting.

On this card, the IOCHRDY jumper seems to be in fact for the IDE interface. I only recently learned that modern I/O cards have a second device that can make use of IOCHRDY: The EPP parallel port. During EPP block transfers, the host can just repeatedly read or write the EPP data port register, and the card uses IOCHRDY to slow down the process to the speed of acknowledgements by the external device. If I understood it correctly, the major difference between EPP 1.7 and EPP 1.9 is that EPP 1.9 is specified to have a timeout for device readiness so the bus can not be blocked indefinitely. Moral of the story: If you see a jumper on IOCHRDY on a Super-I/O card that supports EPP, it might be for EPP, for IDE or for both. Generally: Jumper open = signal interrupted = IOCHRDY usage disabled.

Final sidenote: The reason that ISA NE2000 clones can freeze the system when you just poke their I/O ports (specifically: you access base+10h without setting up the 8390 chip for that access) is also caused by the card holding IORDY low.