• Please review our updated Terms and Rules here

Adding an 8-bit MFM Card...

R

Raven

Veteran Member
Joined
Mar 7, 2009
Messages
2,752
Location
DE, USA..
I have a WD1002A-WX1 MFM controller that I got from a deceased 8086.

I want to put this in my Panasonic Sr. Partner, so that I can add a hard drive. I have an (afaik) functional MFM drive by Seagate, an ST-238R.

I tried hooking it all up, apparently backwards, and got nothing in the second slot of the Panasonic. Terry Y. advised me that the second 8-bit ISA slot may actually be some kind of proprietary slot. I tried the main slot, with the cords properly connected, and the computer boots with a 1701 error, bringing me to the diagnostics screen (this is with no boot disk). My DOS disks for this system don't have copies of FDISK - the DOS 2.11, 3.2, and 3.25 copies I have.

If the controller is hooked up properly should I only need to get a copy of fdisk and poof it will work? Or is there more to it?

BTW: What does MFM stand for?
 
ST-238R is actually a 30mb RLL drive. It will work on your MFM controller, but only as a 20mb drive. You will have to boot the computer from floppy, run debug, and issue the correct command to access the built in low level format software on your WD controller.
 
The "R" in ST-238R indicates that it is a RLL (run length limited) drive. You should still be able to use it with an MFM (modulated frequency method) controller though. In the case of the ST-238R it would format to 20 megs MFM or 30 megs RLL. RLL drives typically used better material for the platters and were capable of storing data in a more dense format.

If you haven't tried already, get a copy of debug and at the "-" prompt type --> G=C800:5. This should bring up the Western Digital low level formating app. Debug is included with MS-DOS but is sometimes located on additional utility disks and not necesarilly on the main boot floppy.

Then just follow the prompts on the low level formating utility. I usually just accept the defaults and do not dynamically configure the drive. I just recently did this to a ST-238 (not the RLL type) and found that an interleave of 4:1 seemed best for my drives performance. You can also use the error sticker on the drive, if there is one, and map any of factory printed defects track map. If all is good, then proceed with fdisk to partition the drive and then follow that up with a regular high level DOS format. Hopefully all work well from there on in.

---

One other thought. On the drive I just prepared, running the low level format several times did seem to decrease the errors reported by chkdsk a small amount after each attempt. Up to a point at least...

After first format - chkdsk reported --> 210944 in bad sectors.
After second format - chkdsk reported --> 202752 in bad sectors. Slight improvement.
After third format - chkdsk reported --> 161792 in bad sectors. Another slight improvement.
After fourth format - chkdsk reported --> 161792 in bad sectors. No change from previous.

Not really sure what was going in here, as I never tried multiple low level formattings before, but it appeared to work.
 
Last edited:
patscc said:
Actually, MFM stands for Modulated Frequency Modulation
I know, sounds a bit dorky.
patscc
Click to expand...

Uhh, Modulated??? You're slippin' Patrick... I always thought it was MODERATED Frequency Modulation. (Yeah, there's this tiny lil' nerd inside the drive, deleting all the spam and censoring the foul language).

--T
 
Anyways, lemme veer back on topic (mark this on your calenders, folks).

Raven, when you say 'diagnostic screen', what exactly are you referring to? Is this something that's built into the Sr. Partner? (Forgive me, it's over 10 years since I've handled one in the flesh). What does the screen say? Is it offering to format the drive, or what?

--T
 
M2fm

M2fm

Terry Yager said...You're slippin' Patrick
Click to expand...
Not yet, I think you just forgot about M2FM, which would then be the
MODERATED Modulated Frequency Modulation.

Oh dear, I think I took the wrong pill.

I think it's actually

Modified Frequency Modulation for MFM

and

Modified Modified Frequency Modulation for M2FM


mfmmfm
 
Terry Yager said:
Anyways, lemme veer back on topic (mark this on your calenders, folks).

Raven, when you say 'diagnostic screen', what exactly are you referring to? Is this something that's built into the Sr. Partner? (Forgive me, it's over 10 years since I've handled one in the flesh). What does the screen say? Is it offering to format the drive, or what?

--T
Click to expand...

It's a fairly useless (unless you're suffering major hardware problems) screen that offers to test your RAM, video memory, floppy drives, etc..

Nothing to do with fixed disks.
 
TomFCS said:
The "R" in ST-238R indicates that it is a RLL (run length limited) drive. You should still be able to use it with an MFM (modulated frequency method) controller though. In the case of the ST-238R it would format to 20 megs MFM or 30 megs RLL. RLL drives typically used better material for the platters and were capable of storing data in a more dense format.

If you haven't tried already, get a copy of debug and at the "-" prompt type --> G=C800:5. This should bring up the Western Digital low level formating app. Debug is included with MS-DOS but is sometimes located on additional utility disks and not necesarilly on the main boot floppy.

Then just follow the prompts on the low level formating utility. I usually just accept the defaults and do not dynamically configure the drive. I just recently did this to a ST-238 (not the RLL type) and found that an interleave of 4:1 seemed best for my drives performance. You can also use the error sticker on the drive, if there is one, and map any of factory printed defects track map. If all is good, then proceed with fdisk to partition the drive and then follow that up with a regular high level DOS format. Hopefully all work well from there on in.

---

One other thought. On the drive I just prepared, running the low level format several times did seem to decrease the errors reported by chkdsk a small amount after each attempt. Up to a point at least...

After first format - chkdsk reported --> 210944 in bad sectors.
After second format - chkdsk reported --> 202752 in bad sectors. Slight improvement.
After third format - chkdsk reported --> 161792 in bad sectors. Another slight improvement.
After fourth format - chkdsk reported --> 161792 in bad sectors. No change from previous.

Not really sure what was going in here, as I never tried multiple low level formattings before, but it appeared to work.
Click to expand...

Doing that line with debug did nothing but hang on both my 8-bit WD MFM controller (the aforementioned one) and my 16-bit WD MFM controller (unknown model, no blatant markings) in my testbed AT machine under DOS 6.22...
 
Wow.
Modified Frequency Modulation. Run Length Limited.(I checked on WIKIPEDIA)

Debug g=c800:5 is the way to start the low level format most of the time. Most 8 bit cards have the first instruction for the program located at that address. I have encountered some that can be started without the :5 and at least one instance where I had to use :eee instead of :5.

I agree to use the default values and see if everything works. Interleave of 3:1 or 4:1 should work. Maybe you can find one of the programs that do the low level format like "Disk Manager" or such. Those programs sometimes tell you the best interleave to use.

If you have access to another "better known" system try set up the controler card and drive on that system to make sure they both work.

If I remember correctly 1701 is a hardware failure such that the system can tell there is a Hard drive controler card but can't see the actual drive. If the card or drive failed or if it is set up incorrectly this would be the case. Or if the mobo sees but can't deal with the controler card.

Cables and jumpers need to be set correctly for it to work.
 
Raven said:
Doing that line with debug did nothing but hang on both my 8-bit WD MFM controller (the aforementioned one) and my 16-bit WD MFM controller (unknown model, no blatant markings) in my testbed AT machine under DOS 6.22...
Click to expand...

Go into debug and do

-dc800:0
-q

And tell us what it says. Not all HD BIOS extensions have a format entry point at c800:5. I've seen them with c800:3 and even c800:8.

(This assumes that your HD extension BIOS is jumpered for c800, of course).
 
Raven said:
... I want to put this in my Panasonic Sr. Partner, so that I can add a hard drive. ... and the computer boots with a 1701 error,
Click to expand...
The Panasonic Sr. Partner is a PC clone and therefore is highly unlikely to have anything in it's BIOS that relates to hard drives.
Therefore, the "1701" is sure to be generated by the BIOS in the WD1002A-WX1. If you boot the computer with the WD1002A-WX1 removed, and you don't see the "1701", that will pretty much prove it.

davd_bob said:
Debug g=c800:5 is the way to start the low level format most of the time. Most 8 bit cards have the first instruction for the program located at that address. I have encountered some that can be started without the :5 and at least one instance where I had to use :eee instead of :5.
Click to expand...
The WD1002A-WX1 is detailed at: http://stason.org/TULARC/pc/hard-di...L-CORPORATION-Two-MFM-ST506-412-driv-221.html

From the text at the end, if there's a low-level formatter entry point, it is C800:5
Also, it's not clear whether or not the low-level formatter referred to is only in the 'Super BIOS' revision of the product, but it seems to read that way.

I found a WD1002A-WX1 in my collection. Even though I can't find 'Super BIOS' printed anywhere on it, I've obviously got that version because when I run the code at C800:5, the first bit of text that gets displayed is "Super bios formatter".

I'm very, very positive that it's your WD1002A-WX1 that is generating the "1701", which means that the BIOS on the card is active.
There are no jumpers to change the address at which the BIOS appears, and so if there is a low-level formatter in the WD1002A-WX1's BIOS, we know from the quoted web site that it will only be at C800:5
So maybe you have an earlier revision of the card (not the 'Super BIOS' one), and maybe the earlier revisions don't have the low-level format code.

WD1002A-WX1 - BIOS revision

Now whilst the low-level formatting could be done via a program such as SpeedStor, the quoted web site indicates that the setting of the switches on the WD1002A-WX1 that you will need to make for the ST-238R varies dependent on the revision of the BIOS on the WD1002A-WX1. Is there a sticker on your WD1002A-WX1 that indicates the BIOS revision (expecting a letter from "A" to "H")?
 
Thought I'd wrap this up, as I tinkered with the machine for a solid night and ended up getting it booting to HDD.

Anyway, I ended up using (can't recall the whole model no.) a WD WX2 card, which functioned properly when the debug line was called. The other two MFM controllers by WD I have failed to do so. I have been using a dual edge-connector floppy cable for the control cable of the MFM/RLL disks, and while I'm still not certain if this is the cable to use, it did function when I changed it to the connector in the middle of the cable. The RLL disk (ST-235R iirc) worked great, while two ST-225s failed miserably (one screamed and one didn't power on), and another just didn't want to work porperly. The RLL disk came out of a largely untouched 8086 that had been sitting in a closet, so my bet is that the MFM disks I have were worn down quite a bit more. It's working great now, though the disk isn't 100% mounted perfectly, I had to use some tape and whatnot as the screws in the Sr. Partner for the floppy drive weren't wide enough to secure the HDD, and wider ones weren't thin enough to fit in the mounting brackets for the Sr. Partner.

To get the RLL drive to work I did have to change some jumpers. On the card (from research) it seems pins 3 and 4 correspond to disk 0, while 1 and 2 are for disk 1. I don't recall the exact details, but I had to close the 3 and 4 sets of pins to get the drive to be recognized properly.

Anywho, thanks for your help - I feel I've learned quite a bit about MFM/RLL and it's configuration through this ordeal, and my Sr. Partner is quite a bit more useful. :D
 
Raven said:
I have been using a dual edge-connector floppy cable for the control cable of the MFM/RLL disks, and while I'm still not certain if this is the cable to use,
Click to expand...
Chuck(G) said:
The "twist" in a floppy cable is very different from the twist in an ST412-interface HD control cable.
Click to expand...
Raven,
Unlike the floppy cable, where both the motor and 'drive select' signals are swapped, the hard drive cable twist only swaps the 'drive select' signals.
And so for a hard drive cable, one also has the option of using a straight-through cable (no twist) if desired, adjusting the drive selection jumpers/switches on both drives as appropriate.
In case you ever need it, here's a visual of the twisted cables:

floppy_vs_hard.jpg
 
Thanks, that clears that up and the visual will certainly help when I end up digging through my box of cables to find another ST cable.
 
If the card BIOS is configured to the address C800 (which is usually default), then G=C800:5 will work. However, if the BIOS address was changed by the previous user, then you'll have to change the debug command accordingly.
 
You must log in or register to reply here.
Back
Top