PDA

View Full Version : Vintage Hard Disks



famicomaster2
February 2nd, 2017, 04:02 PM
Hello everyone! I'm new here, so excuse me if this post looks badly formatted and let me know.

Anyways, if you saw my profile on here, I collect old hard disks (<200MB usually). I've recently come into possession of a Cogito CG912 10MB disk (Pictures below).

It's quite the interesting drive (see pictures I took of the inside - I've never seen a disk with a linear stepper motor at an angle like that!), I must say.

However, when I try to format it with anything at all, the software usually gives me 1701(04) or 1701(20).
1701 is a general controller failure
1701(04) means the controller couldn't find the sector
1701(20) means the drive stopped responding
I'm sure you already knew above, but I felt I should mention it for clarity.

It seems the disk is either made from bad sectors or I'm doing something wrong here.

Here's a recreation of what I did:

A>debug
-g=c800:5

* Format Utility *

Drive no.(1-2): 1

Interleave(2-9): 3

Drive Table:
0. ST-225
1. ST-4038
2. Miniscribe-3425
3. ST-212
4. Priam-V150
5. Priam-V170
6. Miniscribe-8425,Kyocera-20A/20B
7. ST-138
8. ST-4051
9. ST-251/251-1
10. HH-725
11. ST-4052,Miniscribe-3053/6053
12. Maxtor-1085,Newbury Data-1085,Miniscribe-6085
13. ST-4096
14. Maxtor-1140,Newbury Data-1140
15. Free Format

Table no.: 15

Cylinders(1-2048): 306

Heads(1-16): 4

Step Rate(us): 70
(5,10,20,30,40,50,60,70)

Reduced Write Current: 0

Write Precompensation: 128

Split into 2 logical units?(Y/N) N

Is above all correct?(Y/N) Y

Ready to ERASE entire disk?(Y/N) Y

Enter drive defect table?(Y/N) N

Proceed with format?(Y/N) Y

Formatting disk no. 1

Cyl 0 Head 0
ERROR during formatting, code: 20
A>

I'm using a DTC 5150XL controller card with SpeedStor 6.0.3 under a DOS 3.31 floppy diskette. The computer (an ACS Pentium-S 133MHz with 16MB of RAM and an S3 ViRGE video card) has formatted dozens of drives successfully in the past (Managed to get 4MB out of a Kalok Octagon KL320 that had not once been parked since it was new), despite having some issues (Many games crash the system. Not sure why.), such as the diskette drive indicator being perpetually on.

The drive is mechanically sound, since it spins up and has passed every seek test I threw at it. The drive seems unable to read / write any information, however, and some programs see the disk wrong (SpeedStor always sees it as 153C 4H 17S and HDAT sees it as 615C 4H 17S). It still seems to work, since I can park the heads and when the disk is powered on again it will draw them back to track 0.

Should I give up and simply put the disk in the stack that don't work? Should I try a different computer? Is the controller to blame? I have a few other controllers but not many other computers I can use. If anyone can find the manuals for the controllers (I'll give information as requested), that'd be a great help.

Thank's for reading my absurdly long post, and hopefully someone out there can help me.
Note: These pictures were taken about an hour before posting this, and the internal pictures were taken a few seconds after disassembly. The drive spent maybe 45 seconds unsealed in total. The problem occurred before the drive was disassembled.

luckybob
February 2nd, 2017, 06:46 PM
I've recently come to find out, some things from this era are incredibly speed sensitive. Do you have anything older to test it out on? I'm talking 286 or 8088, at most 386.

famicomaster2
February 2nd, 2017, 07:21 PM
I've recently come to find out, some things from this era are incredibly speed sensitive. Do you have anything older to test it out on? I'm talking 286 or 8088, at most 386.

I don't own or have access to a 286 / 386. The oldest machine I have is a 486SX33 with a bad onboard video card. It's all ISA slots (It's an IBM PS/1 I think?), and the only ISA video cards I have are EGA.

I've disabled the caches and run the system at 75MHz, but the disk refuses to format. I'm thinking it's a derivative of servofade or just out of alignment. Usually 1701(04) means the drive can't detect the sector marks. I'm thinking they've either faded or there's some timing-sensitive critical component that has drifted a little too much over the years.

I'm running it with an 8-bit controller card (See: http://www.minuszerodegrees.net/manuals/DTC%205150XL%20-%20Installation%20Guide.pdf) and set the stepping to 70us, which is the slowest I could set it to.

I've uploaded some quick video of testing the drive after it came in the mail, the connectors were oxidized and didn't make very good connection until I cleaned them with rubbing alcohol and a cotton swab:

https://www.youtube.com/watch?v=Z1RffjftGc0&t=29s

You can see in the video that it does seek tests just fine, but formatting will not happen. There's another video earlier on the channel for the "before I cleaned it" phase.

This leads me to believe that it could very well also just be dirty contacts on the card edge connectors, though I'll have to clean them again and see.

If you look around my channel, you can see a few other drives working fine with this controller card, including an ST-251 MLC-2 (If you go really far back), a Kalok Octagon KL320, and an IBM Type 0665-53 disk as well. It also nearly got an ST-151 with a missing head to work. Nearly.

Questions / comments / concerns? I don't particularly need this drive to work, let alone be usable, since I just collect these things for the heck of it.

On a side note, if anyone has any CMI or Miniscribe drives they want to get rid of, let me know

NeXT
February 2nd, 2017, 08:27 PM
I've legit never tried handling MFM drives on anything newer than a 486 myself.

famicomaster2
February 2nd, 2017, 08:35 PM
I've legit never tried handling MFM drives on anything newer than a 486 myself.
So long as you disable the ATA controller and have a free ISA slot, you're all good to go. It's the same as using a SCSI card for a boot device - you just have to set up the card and computer correctly and it'll work.

I've had one of these cards running on a Pentium 4 before (cheap no-name socket 478 motherboard with a single shared PCI / ISA slot at the very bottom), and while there was no way I'd ever have gotten something like Windows running on it, it was still kinda neat that it even worked. Too bad it has bloated caps now and the local radioshack has closed down, otherwise I'd be using it for this instead. The 533MHz FSB is much faster.

I don't believe I ever did a video on it, though.

Chuck(G)
February 2nd, 2017, 08:48 PM
Well, it depends on your diagnostic capabilities. Can you put a 'scope on the head amps and see if you're getting anything? Check to see that write current is getting to the heads. If the heads are bad (open), they're toast and you're done.

Early "winchester" drives are pretty interesting. My favorite oddity was the "shoebox" IMI 4MB drive, which used an early voice-coil+servo system. If the surface it was resting on was more than about 15 degrees off level, the drive would start encountering seek failures.

famicomaster2
February 2nd, 2017, 09:17 PM
Well, it depends on your diagnostic capabilities. Can you put a 'scope on the head amps and see if you're getting anything? Check to see that write current is getting to the heads. If the heads are bad (open), they're toast and you're done.
I'll pull out my scope in a bit and check that. What should I look for? About 1v maybe? I'll get my multimeter out and start checking components, if that's what you mean. I have no way to test things like the microcontroller, though. It doesn't seem like there's anything wrong with it right now, since the drive is detected and responds to seek and access commands. Stupid question, if I wave a big neodymium magnet around it, would that affect the disk enough to force it to reformat correctly?


Early "winchester" drives are pretty interesting.
That's why I enjoy them!


My favorite oddity was the "shoebox" IMI 4MB drive, which used an early voice-coil+servo system. If the surface it was resting on was more than about 15 degrees off level, the drive would start encountering seek failures.
I've never owned any IMI disks, or any less than 8MB total for that matter (Once had a dying ST-412 in my possession, poor thing was missing an entire head but kept on working. The spindle transistors eventually died and since the disks were worth next to nothing at the time I didn't bother to replace them. Oh well.). Do you mean the tilt of the drive or the actuator mechanism or what? I remember early Samsung drives had issues with that, if you had them off level by next to anything or wobbled them much you'd have a head crash on your hands. Horrible old things.

The Kalok Octagon KL3100 (A drive I wish I still had, once I bought about 15-20 used and they all worked to varying degrees until I got rid of them. This is long before I started collecting these things, mind you) was a 105MB IDE disk... With a stepper actuator. They were odd little things and quite cheap for the time, but now they're $40+ and I'm not really willing to pay that much for a moderately unreliable drive at the moment, especially an ATA disk like that. Same goes for the Tandon TM262.

compu_85
February 3rd, 2017, 07:40 AM
The head setup reminds me of the dead MicroScience 60mb unit I've got:

https://www.youtube.com/watch?v=H2P43FwE2MA

-J

famicomaster2
February 4th, 2017, 04:10 PM
The head setup reminds me of the dead MicroScience 60mb unit I've got:

https://www.youtube.com/watch?v=H2P43FwE2MA

Hey hey, I commented on that video! Sent you an email since YouTube was being stupid about notifications. Funny how things come full circle, no?

That disk is voice coil based, though, and this is a stepper motor.