osiweb.org

I recently purchase a know working MPI drive of the same type that OSI used. It came from a TRS-80 machine.

So today I decided to try to get the MPI drive working on my OSI. I was so sure I was going to have 100% success....but alas, no.

So basically it didn't boot from the floppies disks that I have. These disks were recently sent to me to get my machine up and running.

This is what I have done:

- Tested the new MPI drive on my System 80...works great
- Soldered on the connector to the New MPI motherboard for the data splitter
- Swapped boards around between old and new MPI drive, and this is what I now know about my old drive:
•The old MPI mechanism is good with the new MPI motherboard (surprised!!! Especially considering how much work I did on the old drive to try to get it working, new motor, fixed multiple cut wires (some super tiny leading to the drive head), adjusted the alignment etc etc)
•The new MPI mechanism is no go with the old MPI board (Not surprising). S now I know there is still an issue with this board. I had already replaced one IC, but I guess there are more problems.
•The System 80 boots with the new MPI (or the old MPI mechanism with the new motherboard) even with the data splitter daughter board installed (really surprised. Is that even possible!)

So this tells me that the old MPI motherboard is bad, but also maybe the data splitter board is not working too. I would expect that the System 80 would not load anything with the data splitter card installed, but it booted up as if the data splitter wasn't even there! Something doesn't sound right.

The new motherboard does have a few post production wires added to the bottom, so I have no idea what these do.

So thats where I'm at. If you have any thoughts, I'm all ears.

Philip

Hi Philip,
Is this a problem with your OSI disk controller or the floppy media? What OSI and what controller are you using? Does it step the drive to trk 0 when attempting to boot ? More info needed ...

Sorry for the late reply, I've been busy concentrating on a few other projects (fixing a PET 2001 motherboard, and repairing a Keytronic keyboard). It seems the older I get the longer these projects take to complete...

Anyway, I'm done with these other projects now and I have 2 more projects to finish before I believe my collection is 100% up and running (for the first time ever). One thing to do is to get a disk setup running on my SOL-20, and the other is to get the MPI drive working on my OSI.

Well I had a little bit of time to play around with the drives a while back and briefly shared a few emails with Dave. Firstly I was able to confirm that the New MPI drive works. I was also able to confirm through swapping the new MPI motherboard onto the old MPI mechanism, that the old mechanism is good and the issue must lie in the old MPI drive's motherboard somewhere.

Dave mentioned that is an article in the Sept 1986 of peek(65) (vol 7 no 9) which indicates that you actually have to cut a trace (with the raw data signal) on the control board before installing the data separator.

Firstly I compared the boards from both my MPI drives boards. There are very small differences, but not much and there are some jumper wires placed on the back of both that are different between boards. So I went ahead and cut the trace that Dave mentioned. The board from my non working MPI drive had this trace cut. However plugging the new drive back into the OSI still resulted in a no boot.

Apart from the differences with the jumper wires between the two boards, another place where there was an obvious difference is a cut trace from pin 32 of the floppy connector. I have indicated it on the attached photo (this one of the original board). Both boards had this line cut, but the new board then had a jumper installed taking the line to another part of the board. I believe this maybe the side select line, so maybe it's not important.

Well, thats where I am at. Not sure how to proceed, Maybe it is easiest to try to find the issue with the old MPI motherboard, or try to get the new MPI drive working with the OSI, both directions are valid…but which has the best chance of success.

Thanks again

Philip

Finally got this drive working with my OSI. Basically I left all the new MPI board jumpers intact, except I matched the area around J1 to the old board with jumpers and a cut trace. The drive however doesn't seem quite as reliable as the original, but I think thats probably just a very slight alignment issue.

Thanks

Phil