7

I have a NEC PC-8001mkII computer (sometimes incorrectly called a "PC-8001mk2") which has a built-in disk controller for PC-8000-series double-density 5.25" drives. (The English and Japanese Wikipedia pages contain a bit of information about this model.)

The floppy disk connector on the back is a female 36-pin micro-ribbon connector, also known as a "Centronics" connector, similar to printer and SCSI drive connectors from the 1980s. That Wikipedia link shows the male and female connectors, and a picture of the female connector on the computer is below.

I'd like to know what the pinout for this connector is. Bonus points given if you can describe how to hook it up to an HxC 2001 or similar floppy emulator.

Given the number of pins, I'm guessing that it's moderately likely that NEC is directly running Shugart's 34-pin SA-400 interface through that connector. I have an oscilloscope; if you have suggestions for me about how to reverse-engineer the pinout for interface I'll happily take them.

NEC PC-8001mkII floppy disk connector

Improve this question
2
  • @hoshikawa Yes, you are absolutely correct. I'd upvote your answer, except that a mod for some reason felt that it was best to delete the correct and best answer you posted.
    – cjs
    Commented Feb 15, 2020 at 0:25
  • 1
    @hoshikawa I'd also like to apologize for the reception you got here; the community should not be immediately deleting correct answers. I've raised this issue in meta; I'm hoping that we can fix the problem we have that immediately drives away new users with good information. This is entirely our failure as a community, and no reflection on you at all.
    – cjs
    Commented Feb 15, 2020 at 0:59

3 Answers 3

Reset to default
9

bad news, the connector is simply a data port for the "PC-8031-2W" external disk drive unit. all of the 'smarts' are in this external box. this 36 pin connector is literally just 3 8-bit data lanes and ground.

Improve this answer
4
  • 1
    I'm not sure this is very useful. Can you not give the pinout? Also, the numbers do not seem to add up.
    – Tomas By
    Commented Feb 13, 2020 at 21:20
  • 4
    @TomasBy Even without the pinout, this answer is incredibly useful: by explaining that the assumptions in both the question and the other answer are drastically wrong, it's saved me potentially hours of wasted debugging. And the pinout is now not the difficult problem; the difficult problem is the protocol. (Knowing that, the pinout would be relatively easy to determine in comparison. Without the protocol, the pinout is helpful, but not nearly as useful.)
    – cjs
    Commented Feb 17, 2020 at 23:36
  • 1
    @hoshikawa I hope you've not abandoned the community after a mod erroniously deleted your first answer. If you could expand your answer with links to information (even in Japanese) about the pinouts and/or protocol, that would be great.
    – cjs
    Commented Feb 17, 2020 at 23:38
  • @cjs: ok, good.
    – Tomas By
    Commented Feb 17, 2020 at 23:42
5

Starting with the working assumption that it's a variant of the 34-pin Shugart interface is a good start for reverse-engineering the connector.

Every even-numbered pin in the Shugart pinout (and the PC variant) is grounded. You can use a multimeter to test for continuity between the pins of the Centronics connector, and see if every other pin along the connector are electrically connected.

If you are able to open the computer, it would be useful to be able to trace the other pins back to the ICs that are connected to them. The part number of the chip could give you some information on the function of those pins. As an example, if you find some pins are driven by a 7438 chip, the fact that this is a chip with open collector outputs would indicate that these are generating outputs to be sent to the floppy drive.

With the computer running, you could issue commands that should operate the floppy drive, and observe the waveforms on the non-ground pins that you believe are outputs. control signals for Shugart floppy drives are active-low, so you should see the voltage drop to zero when you start trying to access the non-existent drive, and return high if/when the computer times out. One would expect to see this behaviour on the Drive Select lines (e.g. /DS0, pin 10), and on the /MOTOR-ON line (pin 16). There may also be activity on the /STEP (pin 20) and /DIRECTION (pin 18) lines as the floppy controller tries to step the drive's read/write heads back to track 0. (The controller expects to be told when it's there by the floppy drive, by an input signal on pin 26, and will continue stepping towards it until it gets there or gives up.)

If these tests match up with the 34-pin Shugart pinout, I'd suggest that what "walks like a duck and quacks like a duck" probably is a Shugart floppy interface, and it's worth assembling a cable to try it with an actual floppy drive.

Improve this answer
4

from the service manual

@hoshikawa is correct of course, the floppy controller chip is located within the external drive unit, the connector is just a data bus.

Improve this answer
6
  • 2
    Thanks for this. The bookmark is a little disturbing, but is a scanned version of that manual available somewhere? Perhaps, uploaded to archive.org?
    – cjs
    Commented May 6, 2020 at 12:06
  • No the service manual is thick like a phone book, hence my 'paperweight'. It's mostly uninteresting info about the memory map, cpu bus etc. If there is something else specific you need i can take a pic.
    – Chris Hill
    Commented May 7, 2020 at 15:32
  • I'm basically interested in the whole manual; the memory map and other information would be very useful to me too. If you can post a pic of the title page so I can get the exact title and part number (or just post that as text, if you have Japanese input capability), I can do a web search and hopefully find a scan. Thanks!
    – cjs
    Commented May 7, 2020 at 18:36
  • imgur.com/a/O0BFzHT feel free to reach out to me on the 'Japanese Vintage Computer Users' group on facebook, it's much easier to upload photos there.
    – Chris Hill
    Commented May 7, 2020 at 23:47
  • I think the bookmark is great. The only thing wrong with it is that you don't say what cartridge that is. That surely needs to be part of the answer.
    – Wayne Conrad
    Commented May 10, 2020 at 22:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .