I spent some time on the project last night. Still no-go.
The SX and the software is behaving as expected by responding to commands, transferring the buffer, so and so forth. I’m getting garbage data from the floppy drive. I put a scope on the floppy, and the data indeed looks screwed up. Last time this happened, putting a pull-up resistor on the data line fixed the problem. I have 1K pullup in place now, which is correct according to the specs I’ve read. Small cable, only 2ft long, well within the 1 meter specs.
Changing the value of the pull-up really seems to affect the shape of the incoming signal. I tried several different values from about 700 ohms through 4.7k. While they did change the shape to what looks normal, the time between edges is way off. In some cases, the time is way too short. In other cases, its excessively long.
What’s odd is that if I pop out the floppy disk, I still get a stream of output data from the drive. Where’s this data coming from? What is this data?
I have connected the DATA READY lead, and wait for that signal to start accepting data. This drops as it should, a short time after the motor lead goes low.
Speaking of motor leads, my floppy drive spins constantly as long as there is 12v applied. The spinning of the motor is *not* attached to the MTRXD lead. As soon as there’s power it starts spinning, then, when MTRXD is dropped low, the red led activity light comes on, and the data starts flowing. What’s odd is that ejecting the disk produces not even a hiccup in the data stream.
I really wish that the drive would simply not send ANYTHING whenever the heads aren’t on the floppy disk surface. What’s the head picking up and transmitting when its not on the surface? I would expect NOTHING…..
I’m not even confident that the data I’m currently seeing from the drive is even real data from the disk itself!!
I took apart my floppy drive yesterday, and it looks like a regular PC floppy but with an add-on board to handle the BUS-nature of amiga floppy drives. Outside of that circuitry, there were OR-GATES, NAND-GATES, and a D-flop flop. I’m half tempted to say screw the amiga floppy and go straight a cheap (and much more easily available) generic PC floppy.
This may be a silly question, but does the floppy drive still work with the Amiga itself? Am I right in saying you’ve got a ‘tap’ on the floppy cable from the Amiga to the external floppy drive? I’m finding it hard to remember what I read earlier in your blog. Did you disconnect your Amiga from the floppy a long time ago? March 21st entry seems to indicate so:
https://techtravels.org/?p=87
I’m just wondering if somehow the floppy drive is being put in a bad state so that nothing can read it (not even the Amiga?). It seems a bit concerning what you say about needing to change the pull-up resistor, the motor constantly running and all the noisy data. Is the drive dying???
I haven’t tried it with the amiga. That did cross my mind.
Yeah, I disconnected the drive from the Amiga awhile ago. I was using the amiga to seek and read the disk, in essence, worrying only about the data from the SX side, instead of the CONTROL and the DATA from the SX side. I’ve since disconnected (and the post you found is the right date) the drive, and it is now hooked up directly to the SX.
I think the drive is OK. I think the drive’s motor has ALWAYS run, but I’ve never really noticed it before. I assumed the light corresponded with the motor, but in fact, the light just corresponds to the MTRXD signal being low.
I’m going to put some time in today and see what I can come up with.
I am working on a similar project (just starting)
that will use an unmodified standard pc drive.
The electronics look fairly similar, so I should be able to do most of this in the controller that I am using, I too am using the afr.c program that marco wrote, I think it would be beneficial to collaborate on the project, I have a few resources that might be useful in trying to market this, as well as accelerate the development.
I have procured a small quantity of usb floppy controllers that can be programmed with a different firmware to do this entirely in the chip and have the unit work as a real amiga floppy disk drive without needing to dump to adf. (2x speed might also be possible)
email me if interested.
The Catweasel thingy can read and write amiga disks using an ordinary pc floppy, so for sure it can be done. But I guess that’s no news.
But like you said, maybe you should try.
Could you miss terminating some floppy control signal, like floppy side or direction? Would’ve checked these with the scope, and have added pull-up resistors if necessary.