• Please review our updated Terms and Rules here

Looking for 8-bit MFM controller for a ST-4096 drive... suggestions?

Bmurching

Member
Joined
Jul 18, 2019
Messages
31
Location
Bay Area, CA
As per the title...

I have been looking and seem to keep finding,
- 8 bit controllers w/o stated support for this drive
- 16 bit controllers that do support it, but no clear evidence the controller will work in an XT class machine

I’ll take any leads anyone has...
 

Thanks Stone, are those controllers definitively supporting the ST-4096? I’ve had trouble confirming with certainty.

Does anyone know how to tell the difference between the ST11R and the ST11M? The latter might work however most options available online seem to be labeled with a BIOS chip with both models described.
 
the difference between the ST11R and the ST11M?
MFM and RLL (30% better density but requires hi-quality disk drive)

ST-4096 has 1024 tracks so virtually any controller with "user C/H/S" should work i.e. WD-XT-GEN2 or Longshine.
 
Often later model 8 bit cards (like the WD XT-GEN etc) will allow dynamic configuration during the low level formatting process - where you can enter in any C/H/S you desire.

16bit MFM cards will not work in an 8 bit slot.
 
You also can try the OMTI controllers, they can handle such drives, as they have their own interactive setup where you can input the drive geometry.
 
the difference between the ST11R and the ST11M?
MFM and RLL (30% better density but requires hi-quality disk drive)

ST-4096 has 1024 tracks so virtually any controller with "user C/H/S" should work i.e. WD-XT-GEN2 or Longshine.

I should restate my question more clearly: I understand the difference between MFM and RLL, and am trying to understand how to identify whether a card is the ST11R or ST11M. I searched through some pictures online and found a couple of examples that seem to screen print the model on the PCB against one edge so I might have answered my question. Unfortunately these boards carry a ROM chip labeled with both model numbers, and the ST11M seems to be less common and more expensive. If anyone has a lead on a reasonably priced one (e.g. not 2x what the drive would cost) I'd be interested.

Thanks all for the guidance on 16-bit cards being a no-no, the importance of paying attention to # of heads, and the referral to OMTI controllers. I took a look and don't see any reasonably priced options offhand but it's something I can be on the lookout for now.
 
Progress! I received a WD1002A-WX1 controller. After some work, the controller seems to work with my existing ST-225 that was formatted with an Everex EV-391.. I can see the logical drive and read/write without issue. When I connect the ST-4096 the computer throws a 1701 error and Spinrite does not see the disk. The connectors look the same and I have connected the drive identically. I’ve been running through jumper options though the ASCII manual I found is not terribly clear on meaning & location.

I am noticing that when running Spinrite against the st225 as a sanity check of the controller, Spinrite goes into an endless loop during drive fingerprinting, specifically the “drive write caches” step. I suppose I cannot rule out a controller issue until I do a successful Spinrite run.
 
Could also be format incompatibility, i.e., the Everex format is not completely compatible with the WD controller.

This turned out to be the right answer as to why the ST225 (formatted with the Everex controller) was throwing errors during fingerprinting with Spinrite. I have that second ST225 drive (has some errors but a good test article) and I did a fresh format using the controller ROM via debug, and Spinrite was happy doing its thorough analysis thereafter. (The only other difference I'd mention is that I downgraded my boot OS from DOS 6.22 to 5.0 though I imagine that would't have much of an impact. Unrelated to the ST4096 but I noticed that with the ST225, switching from the Everex controller to the WD, the system now recommends a 4:1 interleave (vs 5:1 with the Everex controller). I found myself smiling and thinking, "ah, is MFM controller performance another variable I need to study?!"

Once Spinrite is complete on the ST225 (both copies I have) I'll take a closer look at jumpers on the ST4096.
 
Continuing to work this...

I realized the wd1002a ROM was not the 62-000094-xxx version recommended so I managed to get another controller with the right BIOS. this drive’s debug LLF seems to work. I did a full check using the ST-225 and get no errors or issues. When I try debug low level format of the ST4096, configuring the drive dynamically and I think properly (1024 9 1025 1025 11 5) I get an Error 80. The only thing I’ve found thus far is that Error 80 may mean insufficient power. This is a 35 year old PSU powering a fairly beastly drive, so perhaps. I am going to see if I can power the drive alone off a separate PSU. While I prep to do so, I wanted to check in if anyone else has suggestions.

I disassembled and cleaned the PCB, connectors etc and even found a beer cap wedged between the board and drive (did someone use this as an opener?) but I have not yet designed to open up the drive itself to look for a stuck actuator or some such. The platters definitely spin and I can hear some thunks suggesting the heads move, but it’s hard to say with the cover on.

View attachment 60810

The controller as currently configured.
 
Last edited:
Back
Top