• Please review our updated Terms and Rules here

IBM/ALPS half height floppy issue

offensive_Jerk

Veteran Member
Joined
Jul 13, 2009
Messages
1,226
Location
Wisconsin
I just got a ALPS floppy drive.
Here's a pic.
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=160382436619&ssPageName=STRK:MEWNX:IT

I have it plugged into my Core2Duo system running Win7.
The drive is configured for 360k in the bios. I have it hooked up to the old style floppy connectors with the card edge connectors and the 1.44 connector type.

I noticed the disk wasn't spinning, so I flipped it upside down and noticed the belt was gone. I put a rubber band in as a replacement for testing since I have no idea where to get those belts.

I cleaned the heads a little with a qtip and lubed the rails.
When formatting in command prompt, it says :

Insert new disk in drive a:
Press ENTER when Ready.
The type of file system is RAW
The new file system is FAT.
(it then appears to format, drive spins up and the head moves around)
Verifiying 360k
(then it sits there for a while)
Error in IOCTL call.

I also noticed there are jumpers on the drive, but cannot find what they do, are they of importance?
 
I formatted a floppy in my 5150 and then put it in the drive. Still got the same error.

I then changed the ribbon out, and now when formatting, the head does not appear to move, then after 'verifying 360k' it says 'cannot open volume for direct access.'
 
Can you boot DOS on your Core2Duo machine? The problem with floppies under 2K and later is that a lot of what's really going on is under a couple of layers of software, so you really don't get decent diagnostics.

However, I didn't think that XP and later had 360K support, except as read-only.
 
You should have no issue formatting/reading/writing 360K drive under Win7, even 64-bit - I've done it fine. If you're going to be writing disks the drive can be a bit sensitive to writing speed, so I'd recommend an old machine or a custom card like the catweasel (this is the route I'm going.. eventually). It will work but if the nuances of the drive aren't happy on an older machine it won't read the disk written on the newer machine - it's a gamble.

Anywho, a rubber band might not cut it as a drive belt. Different tension means different spin speed, different friction, etc. Without a proper belt anything could happen. Even if you got it to write it would make floppies incompatible with any other drive (unless compensated with, say, a catweasel), and it may not be able to even read disks.

It's really hard to say.

Edit: They don't support 360K in GUI format tool, nor via the old switches, but they do if you manually specify the format parameters with /f:9,80 or what-have-you (my syntax may be off, and it depends on what sort of format you want). As for read/write it works fine, but writing on a new machine can cause the floppy to become unreadable to an old machine.
 
Edit: They don't support 360K in GUI format tool, nor via the old switches, but they do if you manually specify the format parameters with /f:9,80 or what-have-you (my syntax may be off, and it depends on what sort of format you want). As for read/write it works fine, but writing on a new machine can cause the floppy to become unreadable to an old machine.

That's odd--my XP Pro doesn't support 360K on a 360K drive in any form, not even with /n:9 /t:40. On a 1.2M drive, sure. I suspect that's what you're alluding to when you say the floppy may become unreadable on an old machine (ostensibly with real 360K drives). Double-stepping while writing a 360K format on a 1.2M drive will certainly make it unreadable in a real 360K drive--even when you format under DOS.
 
360K support on XP has been discussed here before. I'm pretty sure that the result was that nobody could get it to fully work (read/write/format), even with BIOS support.

Raven was speaking of Windows 7. I heard a rumor of full 360K support in Vista but never tried it because I don't have Vista. Could Microsoft have reintroduced full 360K support in recognition of retro-computing, in Vista or Windows 7?
 
Raven was speaking of Windows 7. I heard a rumor of full 360K support in Vista but never tried it because I don't have Vista. Could Microsoft have reintroduced full 360K support in recognition of retro-computing, in Vista or Windows 7?

I can confirm that real 360K drive read/write support is nowhere to be seen in Windows 7 Ultimate 64. Don't know about the 32-bit, but a quick perusal of the DDK for Win7 doesn't show much different in the floppy department from XP or Vista.

I'm actually surprised that legacy floppy support is present at all, judging from the number of new motherboards without even a floppy connector.

As an aside, I picked up 50 brand new TDK 3.5" HD floppies from my local Freecycle list. Our local computer recycle/reuse store won't accept floppy disks in any form.
 
i was wondering about the layering issue, and about the belt.
I will try the freedos live cd and see what happens.
 
Let me tell you what I've been doing for the past 3 days. Trying to get a really old Alps RJG2ZA 360k drive to work. Mine also has a belt.

[EDIT: FOR THE PARAGRAPH BELOW I THINK I USED THE WRONG JUMPER SETTINGS ON THE DTK PII-151B. FRANKLY THE DIAGRAM ON THE CARD IS CONFUSING AND CONFLICTS WITH THE ONLINE DOCUMENATION HERE http://stason.org/TULARC/pc/io-cards/C-D/DTK-COMPUTER-INC-Floppy-drive-controller-PII-151B.html. AFTER RETESTING ON 3 DIFFERENT CONTROLLERS, INCLUDING THE DTK PII-151, WITH BIOS BOTH ENABLED AND DISABLED, WITH SETTING 360K IN A 486 BIOS WHEN THE BIOS IS DISABLED, THEY ALL FAILED. THE THING THAT MAKES THE PII-151B DIFFERENT FROM THE OTHER ONES IS IT HAS JUMPERS TO SET 360K/1.2M/720K/1.44M. ALSO IN ORDER FOR THE ALPS TO WORK I HAD TO SET THE BIOS ENABLED JP5 TO CLOSE. WHEN THE ALPS IS NOT WORKING IT EITHER GIVES THE ERROR "DRIVE NOT READY" OR WILL SEE B: AS A CLONE OF A:]

I was trying to get it to work with an original IBM XT 8mhz. First of all I did not know that this drive does not get auto detected by the floppy controller! I have never seen this in my life. The motherboard detected it as a 720K. It wanted to format disks as 720K! Chkdsk.exe reports it as a 720K after a successful format. When it reached track 41 with "firm.com" it would stop with an error. Then I went and plugged it into a 486 with a VLB controller. The 486 thought that drive b: (alps) was drive a: (1.44) and would only access a:. I also tried everything with the XT motherboard, cables, different controller, swap a: for b:. Nothing.

You know what I had to do to get the damn thing to work? I had to force it with an 8-bit DTK controller that had manual override feature.

It is dtk pII-151B http://stason.org/TULARC/pc/io-cards/C-D/DTK-COMPUTER-INC-Floppy-drive-controller-PII-151B.html

[EDIT: BELOW I SAY NO MIXING DRIVES ON THIS CONTROLLER. I AM NOT SURE DUE TO STRANGE JUMPER DIAGRAMS.]

I had to set the DTK to 360K with jumpers. Oh and guess what? No mixing drives with this controller. You can't use 1.44M and 360K it has to be both 360K drives. Oh and don't forget to set the jumper to enable the BIOS on the card of course. Set the controller for single speed if possible. I also set data precompensation mode to the slowest transfer rate setting (JP2 = 1-2). On the XT motherboard set dip7 and dip8 to see no drives at all (since the DTK BIOS does it). Also don't think it will read your floppies formatted with a 1.2M drive nooooooo. Then you have track widths that are too thin. You will need to image the floppy with "firm.com" http://www.glennmcc.org/download/free_software/disk1.htm#floppyutil from another drive and then reimage it with firm.com on any 360K or the alps to get a nice fat track width. I mean this is really old equipment man!

Oh and that belt? It reads "405" on it. But that probabily doesn't help.

This Alps drive I have looks positively trippy! The front door looks just like commodore 1541 drive exactly! The noise it makes sounds like static electricity. A great find! I hope you get it working. Maybe your Alps has autodetect and has newer features than mine? Somehow because it has a belt I doubt it. I could be wrong.

One more thing. The jumpers on the drive itself. The only jumper you should need is DS2 *I THINK*. Here is a guide to what the jumpers mean for old 360k drives http://www.classiccmp.org/pipermail/cctalk/1999-March/122818.html

I just remembered one last thing to help you, I know this is getting crazy. When I tried to get the DTK to work in the 486 board (newer model) it did not work. I might have been able to get it working by setting the 8 bit I/O recovery setting in the BIOS higher but I didn't bother since I planned to use this drive in an XT. They have this setting in fast machines that still had ISA bus. Therefore if you still have a Pentium or Pentium 2 with ISA bus and you bought a special controller like this DTK it is *possible* it could work. If you really must be able to read 360k drives in your latest computer you might need a newer one like panasonic/matsushita and even then you might need to boot into dos, freedos or windows 98 lite, etc, etc in order to have 100% support.

Also you might be able to use a catweasel like that guy said.

Thank you and have an Alps-y nice day.
 
Last edited:
Auto detected? Every controller and motherboard I've ever seen requires you to select the drive type manually. Where have you found auto-detect motherboards?
 
You're right there is no auto-detect on motherboards but there is auto-detect on some DTK floppy controllers that have a BIOS enabled. I have another DTK card that does that. It was not able to detect the drive. Instead it thinks b: is a: just like a regular floppy controller. It only works by forcing it to 360k with jumpers on an old DTK controller.
 
You're right there is no auto-detect on motherboards but there is auto-detect on some DTK floppy controllers that have a BIOS enabled. I have another DTK card that does that. It was not able to detect the drive. Instead it thinks b: is a: just like a regular floppy controller. It only works by forcing it to 360k with jumpers on an old DTK controller.


Any auto-detect scheme without a formatted diskette in the drive can only guess based on the number of cylinders on the drive. This is done by stepping more than 40 cylinders out and then stepping the same number back. If the track zero sensor pops before the seek is finished, then it's an 80-cylinder drive. Beyond that, the interfaces look pretty much the same. Anything else, such as using the "disk changed" status depends on the specific manufacturer's implementation of the drive logic. It's not airtight by any means.
 
Any auto-detect scheme without a formatted diskette in the drive can only guess based on the number of cylinders on the drive. This is done by stepping more than 40 cylinders out and then stepping the same number back. If the track zero sensor pops before the seek is finished, then it's an 80-cylinder drive. Beyond that, the interfaces look pretty much the same. Anything else, such as using the "disk changed" status depends on the specific manufacturer's implementation of the drive logic. It's not airtight by any means.

So basically what you are saying is the DTK with the BIOS autodetect "feature" is just a bunch of circus tricks. I know it works with 1.44 so how does it know 360/720/1.2/1.44 I wonder? Pretty fancy BIOS when you think about it.
 
Many 360K floppies have DRIVE READY on pin 34, or leave it open and 360K normally have not more than 42 cylinders before they hit the stop. So you can separate 360K from everything else by the number of cylinders. After that, you can try to separate 720K from 1.44M by the "disk changed" line. If there were a disk in the 5.25" drive, you could also time the rotational speed and separate 1.2M from 1.44M, but that's not foolproof.

Fanyc? Not really--especially if it's not foolproof.
 
So basically what you are saying is the DTK with the BIOS autodetect "feature" is just a bunch of circus tricks. I know it works with 1.44 so how does it know 360/720/1.2/1.44 I wonder? Pretty fancy BIOS when you think about it.
Sounds like the clock that's right twice a day if it only 'autodetects' one type of drive...
And what do you mean "it thinks b: is a: just like a regular floppy controller"?
 
The DTK Mini Micro (I assume that's the card you have with a NS8473 FDC on it) is actually a pretty decent card. Not a Sysgen Ominibridge or CompatiCard, but not bad. I've got a couple here--they're useful in that they're one of the few controllers that can successfully read and write 128-byte sectored double-density media.
 
Is it possible that because it successfully "autodetects" one variety of drive, perhaps the default configuration of the card is that particular variety of drive and there is no detection of any sort going on?
 
Sounds like the clock that's right twice a day if it only 'autodetects' one type of drive...
And what do you mean "it thinks b: is a: just like a regular floppy controller"?

It means that if the Alps is the b: drive and you say "dir b:" it does different things depending on the controller. It will either show the directory content of a: or on another controller it will say "please insert disk into drive a:" and after pressing enter show drive a: instead of b:. Don't tell me that's normal.
 
The DTK Mini Micro (I assume that's the card you have with a NS8473 FDC on it) is actually a pretty decent card. Not a Sysgen Ominibridge or CompatiCard, but not bad. I've got a couple here--they're useful in that they're one of the few controllers that can successfully read and write 128-byte sectored double-density media.

It is the DTK PII-151 mini/micro (not to be confused with the PII-151b that works manually). Why the hell did you have to tell me these are special? I had 10 and sent 9 to the recyclers. Oh well there can't be that much demand for 128-byte sectored media can there? *GULP*

The main chip says UM8272A which is UMC. Please tell me it's not as good as the NS8473.
 
Back
Top