• Please review our updated Terms and Rules here

26 Pin Hard Disk Connector

The service manual says the PC7200 takes a JVC JD3824R0-0D1, which is an MFM drive. Granted that the interface cable is 26 pins, but it's definitely not IDE, but RLL MFM. The hard disk interface board contains several LSI components as well as a NEC uPD70008 CPU (a Z80 clone).

When I asked about the HD I/F board, Haem said that he didn't have one in his unit. I saw no reason not to believe him.

I wonder if you could fit a similar small MFM drive, such as a Seagate ST138R in place of the JVC drive. But I suspect any other 3.5" ST412-interface hard drive would be just as hard to come by.
 
Sorry, when I read 'MFM' I just assumed two cables. Can't remember if there was a separate controller board in the PC-7200 I had. The Data General 2T has a small board that sits under the drive. It contains a ROM, Two Toshiba chips (T7518 / DC2090P431A) and some logic. GRiD has a similar board with some Western Digital chips (WD11C00-JT / WD1010AJM) Both boards connect to the motherboard via a 34pin connection.

One thing I have noticed with these drives is that they seem to move the head to a safe area of the disk after a short time of inactivity. I'm assuming this is all to do with the drive / interface being designed for laptops.
 
So to clear the confusion, what type of formatting or encoding do the other JVC hard drives use, if JD3824R0-0D1 is a MFM drive with a 26-pin connector, but not compatible with the JVC 26-pin interface we're discussing for other computers?

Indeed I also looked at his pics, and it seems one ribbon cable on the PC-7200 motherboard goes to a daughter board where the 5.25" floppy drive is attached, as well as a 26-pin white female connector which I presumed is where the HDD would plug in. Perhaps that was a premature assumption, since I haven't checked the service manual for this particular machine.
 
carlsson said:
So to clear the confusion, what type of formatting or encoding do the other JVC hard drives use, if JD3824R0-0D1 is a MFM drive with a 26-pin connector, but not compatible with the JVC 26-pin interface we're discussing for other computers?
Click to expand...

It could be that we're talking about the same thing. JVC also made 2.5" XTA drives (e.g. JD-E2042M) and some other bizarre models, e.g. JD-E3824TA.

The spec in the Sharp says that 34 sectors per track are used, but the "PC Engineer's Source Book" says 48.
 
fjkraan said:
... I made some images of an 26-pin hard disk from an Toshiba T-3100: http://fjkraan.home.xs4all.nl/comp/divcomp/26pinHD.html
Click to expand...
Note that the T3100e uses a standard IDE drive.



Aside from the EGA display capability the T3200 is very similar to the 3100; the hard disk controller seems to be the same:
T3100HDC.JPG

The hard disk in mine is a 42MB Fujitsu M2227DT (615/8/17):
DSCI0171.jpg

With a 28 pin interface, (pins 27 & 28 unused):
T3200HDif.JPG
 
Just revisiting the forum and saw this post. The link led me to the post by Shawn who used to host the yahoo Grid forum trying to use a JVC HDD to replace the notoriously faulty Grid Conner in the 15xx Grid models.. He later came up with a BIOS mod to get around the specific Conner reference by installing an eprom mod. If I live long enough I'll do it also. :^). A good find.
Hows it going. Got to phone you and Druid at some point.
 
I just picked up a laptop (Victor V86P) with one of these drives that appears to be particularly unhappy. At first glance, it sounds like it's trying to spin up, but the spindle is stuck (it's also putting such a load on the rails the system won't start with it connected). Pulling the controller board does the trick as far as the rest of the machine is concerned, but it'd be nice to dig a little further.

Has anyone been able to dig further into what the signalling looks like on a functional system? There seems to be some hypothesising that this is an MFM/RLL and might have similar signalling to ST-506 which might make it a candidate for replacement with an emulator like David Gesswein's (although that'd need a board respin to drop unnecessary connectors and fit it into a 3.5" bay).
 
I've been doing a bunch of digging into this recently - I've posted my findings to date up on my site here and here

Basically, it looks like RLL, possibly ST-506 signalling, but all TTL (rather than differential pairs for read/write), and for my system at least, a hard-sectored drive.

I still need to confirm what's being read-from/written-to disk - there's only one head select line, which would indicate two heads, but DOS sees four. I've confirmed there isn't another head select line, and it looks like 17 sectors per track, so at this point I'm wondering if it might be storing 1k sectors and splitting them. I'll post more as I do some more digging - it's mostly a matter of finding the time at the moment.
 
According to the size/thickness of the drive I guess it just has one plattter so supporting two heads max. Dos/bios can be fooled into using another geometry than the disk physically has to keep the number of cylinders/sectors/heads low enough.
 
I recently loaned one of my Epson Equity LTs to the manufacturer of the drem (www.drem.info) and he was able to build an optional card for his emulator that will allow the emulator to replace the JVC 26pin 20MB RLL drives found in several Epson and Toshiba laptops. I got 2 of them from him and so far they work great. His drem documentation has been updated to include specs on the controller as well.
 
You must log in or register to reply here.
Back
Top