Research on Micronik HD 1.76 MB Power M-disk

[caver99]

Well i have a broken one here(drive that is), if anyone wants to try to read the chip and make a drive they are welcome to the chip only one side affect is that could i have a working one please.

[salax54]

Sorry for digging this old thread, but i do have a problem with my drive..
My A2000 psu seems to have gone kaput, and burnt the Micronik drive. Well, dunno yet if the whole drive has gone to Amiga heaven yet, but the CPLD certainly has! I will attach some pictures for you to see, she's in charcoal mode.
So, has anyone managed to dump the contents of the CPLD yet?

[salax54]

I also made a rough schematic of the cabling that was done by Micronik to modify the drive for Amiga use. However, i couldn't help noticing that this particular drive has been a subject of discussion with other Amigans, as a mod to use like a normal Amiga drive in PC default state, plus, there's been talk among the Atarians on how to use it as an HD drive, by modifying the ST motherboard itself..

[chrisfi]

Originally Posted by salax54

Sorry for digging this old thread, but i do have a problem with my drive..
My A2000 psu seems to have gone kaput, and burnt the Micronik drive. Well, dunno yet if the whole drive has gone to Amiga heaven yet, but the CPLD certainly has! I will attach some pictures for you to see, she's in charcoal mode.
So, has anyone managed to dump the contents of the CPLD yet?

Thanks for posting the pictures . Very interesting. I think I can guess how the Micronik mod works from the pictures you posted, well at least part of it.

I’ve been researching Amiga FDDs since completing my A4000TX. A 880k FDD just seems wrong in an A4000. It needs a HD FDD!

Time to get out my collection of Sony FDDs and borrow an oscilloscope from work. If I can confirm my guess on how the logic works, we can homebrew our own GALs to mod the Sony FDDs and perhaps repair your FDD too. I think I’ve even got a few spare 16V8 GALs somewhere.

It would help a great deal if you could take some pictures of the track cuts made on the PCB or put an X on the pictures you have already taken.

I can see the solder blob at SH1 has been removed and the track to pin 34 on CN1 has been cut. Pretty standard stuff.

I’m guessing the track to pin 16 of IC2 has been cut and possibly the track to pin 8 on CN1.

[chrisfi]

Well, I’ve got it working….in the main.



My hack varies from the Micronik method in that it doesn’t use the index pulse. It can be a bit temperamental so it will need some tweaks to refine it. To that end I’m looking at migrating from the rats nest logic to a PLD. .

I have a couple of spare GAL16v8 which need WINCUPL to create the JED file. Having downloaded this I can see its very old (Win 98/ NT) and although there is guidance online much of it is pretty basic or very high level.

I would be grateful if anyone can direct me to some resources for getting started to intermediate use of WINCUPL or similar software for 16V8s

Thanks in anticipation.

[chrisfi]

I’ve got this working quite well.

The previous FDD was faulty. Everything works nicely after swapping to an alternate FDD. It even reads HD PC formatted disks.

I’ve climbed the steep learning curve of learning and programming GALs and got all the logic onto a 16V8. It’s currently on a DIL-20 but I will migrate to a PLCC-20 so I can fit it in the FDD’s case and get it fitted in my A4000TX. Happy days

[orphan264]

Wow! Nice work!

[jbenam]

Originally Posted by chrisfi

It’s currently on a DIL-20 but I will migrate to a PLCC-20 so I can fit it in the FDD’s case and get it fitted in my A4000TX. Happy days

Great job! Do you intend to release it as open-source?

[chrisfi]

Originally Posted by jbenam

Great job! Do you intend to release it as open-source?

Thanks and yes, I do.

I want to modify another one of my FDDs to first check this wasn't a lucky fluke and second to take some pictures and document the procedure.

Github would be an ideal place to share this, but I don't think I have the time to work out how to use that, so will probably post on Hardware mods for the interim

[chrisfi]

The good - I’ve successfully modified another drive which has passed all my tests in both my B2000 and A4000TX.

The bad - I hit a small snag testing this with a PLCC-20 variant. Again everything works fine in my B2000, but in my A4000 when reading a PC HD disk I get occasional read errors. I suspect it's a timing issue because the PLCC-20 variant is over twice as fast at 10ns compared to 25ns for the DIP-20 variant. I have some ideas to try, but first I will dust off my logic analyser to see if I can narrow down where the issue is.

[jbenam]

I have a 10ns PLCC GAL waiting for this Especially since my original drive stopped reading floppies correctly.

Will this work on any PC drive (the moddable ones, at least) or just a specific model?

[chrisfi]

Originally Posted by jbenam

I have a 10ns PLCC GAL waiting for this Especially since my original drive stopped reading floppies correctly.

Ah, Only the 25ns version is working right now. I'm working on a fix for the 10ns PLCC, but not sure how long because I don't have much spare time right now.

Originally Posted by jbenam

Will this work on any PC drive (the moddable ones, at least) or just a specific model?

I've tested this on Sony MPF920 drives with the D9848N motor controller (IC2). I'm sure it will work on Sony drives with the BA6986FS IC2 because this is what is in the picture of the Micronik drive.

I would think this hack would work with some other PC drives, but not all. It's a case of finding the 1Mhz reference (easy with an oscilloscope) and possibly changing a resistor or capacitor to get a steady 300/150rpm (not so easy unless you can find the datasheet for the IC).

[lesta_smsc]

Are you able to sell a preprogrammed GAL chip so we can purchase and mod our own drives?

[chrisfi]

I would be happy to provide some from my spare GALs. If there is sufficient demand, I could buy a batch of GALs - much cheaper than small quantity orders.

However, after some use my drives have developed an issue writing DD disks using WB. Everything else works - full R/W HD disks including PC HD format and everything works at DD save writing from WB - formatting, tracksaver, X-Copy and reading all the floppies I can throw at it.

Same issue on both drives, so I have some logic debugging to do before this is ready to share.

[chrisfi]

I have made some changes and everything works well …… so far …..

It even works with the PLCC variant save one minor issue - the drive/disk ID can be a little bit erratic. Its good enough for the Amiga to correctly identify when HD or DD disks are inserted and reads/writes just fine, but ATK reports the ID is ‘unstable’. Probing with my oscilloscope I can see there is some noise pickup which sometimes shifts the ID by one bit so careful placement of the wires and keeping them short is key to minimising this.

There are some possible refinements I could make to this, but so far it works well in both my A4000 and B2000 so I would like to carry on testing for a bit longer before declaring a full success.

It would help if others could try the hack and report their experience be it good or bad.

This should work with any Sony MPF920. If you’ve already modified one to work as an Amiga DD drive, you have a good starting point and its easy to undo to revert back to a DD drive.

If there are any early adopters out there who wouldn’t mind trying it out and reporting back to me, please send a DM.

[salteadorneo]

Originally Posted by chrisfi

However, after some use my drives have developed an issue writing DD disks using WB. Everything else works - full R/W HD disks including PC HD format and everything works at DD save writing from WB - formatting, tracksaver, X-Copy and reading all the floppies I can throw at it.

I'm a similar problem they give me two SONY units, adapted as DD for AMIGA, copying data from the WB (AmigaOs 3.2.1). I have a couple of brand new GALs (disappeared to see if I can find them) and two Sony floppy drives from a different year.

[hooverphonique]

Originally Posted by chrisfi

I would think this hack would work with some other PC drives, but not all. It's a case of finding the 1Mhz reference (easy with an oscilloscope) and possibly changing a resistor or capacitor to get a steady 300/150rpm (not so easy unless you can find the datasheet for the IC).

I tried making a random HD drive run at 150rpm at one point by halving the 1MHz reference clock for the motor driver. It semi-worked, although the speed did not end up at exactly half for some reason. Never got any further with it.

[salteadorneo]

One question, what would have to be done to make this modified drive work as an external drive? I have a second Chinon 357A functional, but have not been able to get it to work in a box as a hi unit, it is only recognized as a low drive. I assume that the high density disk signal is not sent to the amiga through the external connector.

[Jope]

You will need to bypass the ready pulse circuit from the drive PCB.

[Mad-Matt]

At the risk of going off topic, did anyone ever try the HiDensity driver? http://aminet.net/package/disk/misc/HiDensity

Its been 20 odd years since I last tried. In my case it did kind of work, but was very unreliable in that it world appear work fine until you reset the computer and then the disk would not read any more and would need to start over.

Maybe there is something to this driver which just needs some more work to make it more reliable and useful.