Assn 1 Y Huang Sony floppy-PCMCIA


This page documents my exploits with a Sony PCMCIA Floppy drive. Unfortunately I was not able to put the thing back together - the screws were too much to keep track of, my screwdriver broke during the dissection, and floppy drives are probably too much of a precision part anyways.

 

Disassembly Process

 

Here is a picture of the floppy, pre-disassembly:

 

 

I started by opening the cover of the actual floppy drive (they were just plastic latches), then unscrewing the screws on the cover. Those screws were TINY! Here's a comparison shot between my small-size flat-head (normal) screwdriver, the miniature screwdriver I was using for these screws, and the screw itself:

 

 

(You can also see the cap of the film canister I was using to hold the screws.) Unfortunately while taking off this screw, my screwdriver broke:

 

 

So obviously I wouldn't be able to put the thing back together again. The next smallest mini-bit I had didn't quite fit, so for the screw pictured above (fortunately the last such screw on that side of the case) I actually chiseled the channel in the screw wider with the wider bit. Eventually I was able to get it off and slid the cover off. The inside has two halves - the top half contains some mechanical linkages for the floppy eject mechanism, and the bottom half the motors and control electronics. This is the bottom half (I have unscrewed the worm-gear motor assembly):

 

 

And here is a series of pictures demonstrating how the top half "expands" to push the floppy against the drive motor when the switch is actuated by the insertion of a disk:

 

 

I then tried to power some of the various electronics components using some old batteries I had lying around. (Incidentally, these 1.5V nominal batteries were drained to the point where they put out something like 0.8V, but I have a 4xAA battery holder sitting around and all four produced something like 3 V!) I was able to light up the drive active LED light, but was unable to make either the drive motor or the worm-gear motor spin. Either I didn't have enough power in the batteries or the motors were not of a simple brushless variety. (Most likely the latter.)

 

That's about all that I could do with the floppy mechanism, so I tried to take apart the PC card connector. That proved much more difficult - there were no screws anywhere to be seen, trying to pry it apart by force only resulted in broken plastic bits. Finally I was able to find some sticky tape which peeled off to reveal... absolutely nothing. I then tried to pry off the metal with a screwdriver - this actually met with some success. I was able to pull the entire top piece off and reveal a plastic casing underneath. I tried the same with the bottom side, but didn't realize that there were two screws hiding underneath another similar piece of adhesive. Here's a shot of the top with those screws:

 

 

Here you see a picture with both top & bottom metal plates taken off, but with the screws still attached (I snapped the plastic bit the screws were attached too, whoops.):

 

 

Finally I got all the mechanical things off and this is the circuit board inside:

 

 

Unfortunately nothing decipherable from the ICs. The two larger ones are made by "Ye Data" and have no information on the Internet; the smaller one is made by BSI and seems to be a RAM chip of some sort. Finally here's a shot of the whole assembly:

 

 

High-level Overview

 

Here's some basic schematics of what I think is going on:

 

Not too much going on here. Only so much you can do with a floppy drive...

 

Pretty much already covered in my notes above. Floppy goes in, engages latch, pressure plate comes down (use eject button to eject, of course). Drive motor engages floppy, spins disk, motor attached to wormgear moves read/write head back & forth to find track.

 

This one is kind of blocky but it has all the electronic components I can easily recognize. After the PCMCIA interface (kind of a black box) there's a central control chip of some kind which is probably linked to the rest of the circuitry. It looks like two stepper motor controllers (labeled "MC") and at least one status LED and a read/write head with a coil of some kind.

 

 

What I have discovered:

 

Still unanswered: