IBM/Teac external floppy - modding possible?

[mark_k]

Originally Posted by prowler

The first picture shows the opened enclosure. The top is on the right. There is one lug in the centre at the back and three on either side, which need to be pressed inwards through the slots running around the back and sides before the case can be pulled apart. The eject button just lifts away with the top and can easily be reinserted when the drive is reassembled.

In the middle pic you posted, it looks like there's a yellow wire near the drive connector. Does that look like it was added separately? I'm wondering if that could have been done by Amitek. Does the top/bottom of the drive casing look like it would come off easily? If you could look at the circuit board to see if there's any sign of it being modified that would be great.

[mark_k]

Originally Posted by prowler

Okay, I now have the circuit board assembly and cable completely out of the shroud.

There are no broken wires, as far as I can tell, but there are three wires (white, green/black and blue/black) which are not connected.

Do you have a multimeter? You could use that to check which pins of the D23 connector those go to. Maybe they're the drive 2/3 select lines or extra grounds?

The GAL chip hopefully isn't too complicated, presumably used for the drive ID (the Amiga can determine the drive type by reading the ID). Maybe it would be possible to implement the same circuit with a few standard TTL chips, assuming someone is able to read out/dump the GAL to figure out exactly what it's doing.

Oh, check this page (in French). Nice pics of the drive with its casing off, it looks like Amitek did modify it slightly.

From the pics on that page, one wire goes from a 4MHz crystal to a (probably unused) pin of the drive connector. Another wire goes from another (unused?) drive connector pin to what looks like one of the contacts where a crystal used to be on the spindle motor PCB.

My guess is, the GAL chip divides the 4MHz clock down by a different factor depending on whether there's a DD or HD disk inserted. And that divided clock is sent to the spindle motor PCB to get it to spin at ~150rpm with HD disks. So the drive mod could be as simple as removing crystal from motor PCB and soldering two wires. I'd like to see a pic of an unmodified FD-05HF motor PCB to check the crystal frequency.

Using unused pins of the drive connector could explain why Amitek didn't modify the original drive cable; the original cable probably didn't connect to those pins.

[ElectroBlaster]

Im pretty sure the 3 loose wires in the cable are the same on mine so must be normal!

I am currently trying to find my cable, it was not where I thought it would be so have to go in again and move a couple layers of stuff to get at it.

[prowler]

Originally Posted by mark_k

Do you have a multimeter? You could use that to check which pins of the D23 connector those go to. Maybe they're the drive 2/3 select lines or extra grounds?

Yes, I have a multimeter, and when I have got rid of all the hot glue I will check which pins of the D23 connector ALL the wires go to.

Originally Posted by mark_k

Oh, check this page (in French). Nice pics of the drive with its casing off, it looks like Amitek did modify it slightly.

From the pics on that page, one wire goes from a 4MHz crystal to a (probably unused) pin of the drive connector. Another wire goes from another (unused?) drive connector pin to what looks like one of the contacts where a crystal used to be on the spindle motor PCB.

My guess is, the GAL chip divides the 4MHz clock down by a different factor depending on whether there's a DD or HD disk inserted. And that divided clock is sent to the spindle motor PCB to get it to spin at ~150rpm with HD disks. So the drive mod could be as simple as removing crystal from motor PCB and soldering two wires. I'd like to see a pic of an unmodified FD-05HF motor PCB to check the crystal frequency.

Using unused pins of the drive connector could explain why Amitek didn't modify the original drive cable; the original cable probably didn't connect to those pins.

Yes, as you have now seen, that yellow wire visible near the drive connector is, in fact, two wires. I have removed the bottom of the drive casing and it looks exactly like the one in the picture on that French webpage.

At this point, without removing the spindle motor PCB, I think it's safe to assume that my drive has been modified in the same way as the one shown in those pictures.

Now that I have my drive stripped down this far, I'm going to hook it up to one of my A1200s and satisfy my curiosity about whether it works or not before I go any further.

PS. I've just noticed that the serial number on my drive is only 5 below the French one!

[Stedy]

Hi,

The Teac FD-05HF drive has a density output pin. This feeds to the GAL device and indicates if double density (logic 0) or high density (logic 1). Most floppy devices have a density select input but no output. With the output, to know what disk is inserted, you either use the 4 MHz clock for DD disks or divide by 2 for HD disks and feed the clock back to it's destination.

You also need some logic to provide some device ID back to the Amiga.

I had one of these Amitek drives, it broke many years ago and I threw it out. Never got round to replacing it.

Ian

[K.C.Lee]

I made my hack way back using old Sony MPF420-D drive. Used a 74HC74 and a GAL16V8.

It was probably around 1995 (date code for drive is 1994) ... So working off my memory.

Basically I pulled off the Resonator (cheap subst for crystal), made an oscillator inside the GAL and feed the output or a divide by 2 back to the drive circuit depending on the HD sense. Went out of gates, so used the 74HC74 to latch the motor on signal. My circuit has a pass through to additional drives, but I didn't bother as the mod wasn't reliable for HD floppies.

I have attached the GAL equation I used. Hope someone might find some use for it.

Looking at the drive right now, I would probably check out pin 1 from the big chip to the motor driver. But this is now, not 20 years ago. My old mod slowed down everything (including step rate) instead of just the drive motor.

[amigoun]

I also have one of these external Amitek slim drives. I dont know why, but at some time it stopped working as a HD drive and now works only as a DD drive.

[mech]

Originally Posted by prowler

Yes, I have a multimeter, and when I have got rid of all the hot glue I will check which pins of the D23 connector ALL the wires go to.



Yes, as you have now seen, that yellow wire visible near the drive connector is, in fact, two wires. I have removed the bottom of the drive casing and it looks exactly like the one in the picture on that French webpage.

At this point, without removing the spindle motor PCB, I think it's safe to assume that my drive has been modified in the same way as the one shown in those pictures.

Now that I have my drive stripped down this far, I'm going to hook it up to one of my A1200s and satisfy my curiosity about whether it works or not before I go any further.

PS. I've just noticed that the serial number on my drive is only 5 below the French one!

Great work, we sure could use a pinout of the cable and it would also be nice to see if the Pal/Gal is protected or not. Connection to the drive would also be useful.

Its amazing how many people own these drives yet no one will pin it out

[prowler]

Originally Posted by mech

Its amazing how many people own these drives yet no one will pin it out

I will, now that you've reminded me. I just needed a gentle nudge.

[AwoLStill]

I had a different external slimline IBM drive (purchased from an Amiga dealer, I forget who)

Anyway, it's different from the one in this thread, my description is as follows:

An External slimline DSHD (high density) Floppy Disk drive, the drive itself was an IBM laptop external floppy disk drive, and I cannot remember which model it was.

The eject button was painfully small (and red, if memory serves) but the drive itself was metal cased (black) and quite sharp fronted (the top of the case sat forward of the drive)

Anyone remember which IBM drive it was, or have one lying around?

- I'd like to know the "FRU" number that should be on one of the labels on the drive itself ideally, see if I can source one...

[ElectroBlaster]

Gentle push/bump/nudge!

I would love for this flaming things connector to be pinned out, the original pictures are fuzzy and do not show enough detail. Hot glue is all over the board aswell