Second Hard Drive for UNIX-pc; HwNote10

[John B. Milton]

In article <413 at limbic.UUCP> gil at limbic.UUCP (Gil Kloepfer Jr.) writes:
[describes merciless feet stomping and how he got the 2nd hard drive going]

I'm glad to see the schematic you got worked ok.

Yeah, yeah, I've been slackin' lately. My design is now complete except for
some board layout details. Now for the grubby details...

Unlike the P5.1 upgrade, this upgrade CAN NOT be passive.
Some wires just HAVE TO BE CUT FOR IT TO WORK!

For older mother boards:
  13N-1 must be clipped loose and connected to grounded. 13N-2 must be clipped
  loose and connected to 14M-3. This is actually an upgrade done by Convergent
  on later mother boards.
For all mother boards:
  13K-3 must be clipped loose from the board.

Parts list so far:
 1 double sided PC board
 1 26LS31 differential line driver
 1 26LS32 differential line receiver
 1 16R8 PAL
 2 16 pin sockets
 1 20 pin socket
 4 20 pin headers
 2 34 pin headers
 3 bypass caps
 4 100 ohm resistors

Other parts needed:
 1 long 34 pin daisy chain cable with one header connector and at least two
   drive connectors
 1 an internal 34 pin cable to go from the mother board header to the header on
   my board
 ? 20 pin data cables for each hard drive
 1 external box and power supply

Connections inside the UNIXpc:
 1. Data bus line D1
 2. Data bus line D2
 3. Data bus line D3
 4. The register select line MCR2SEL*
 5. volts.
 6. The hard disk read data line to the hard disk data separator PAL, 14M-3
 7. The hard disk write data line, 16M-6
 8. Maybe the RST* line.

It will:
 1. Add select lines for 3 more drives to the 34 pin cable. (total of 4 drives)
 2. Provide a 34 pin header.
 3. Provide active data connectors (20 pin) for 4 drives.
 4. Work no matter how you mount it. (within reason)
 5. Make you the envy of all your friends.

It will not:
 1. Provide P5.1 board version feedback.
 2. Provide additional 4th head select bit.
 3. Provide floppy tape support.
 4. Provide a watch dog circuit.
 5. Force you to take out your fan if you don't want to.
 6. Cost too much. A price of $50.00 is what I'm shooting for now.
 7. Affect the price of tea in China.

Since the 3 drive selects have to be added to the 34 pin cable, the ground and
DDRIVE0* line will be picked off of the 34 pin cable from the mother board.
This will save two patch wires.

I am assuming that your machine will have the "P5.1 FIELD UPGRADE" already
installed. I will make that available as a SEPARATE kit. As I have mentioned
in the HwNotes, you do not HAVE to have the P5.1 to use two drives. You DO have
to have the P5.1 to format the second drive without swapping cables around.
I plan to have a patch for the diag disk (3.5 and 3.51) by the time I start
selling. The reason I am not providing the P5.1 on my board is that a lot of
people already have the P5.1 installed. I also wanted to keep the number of
wires going to my board down. The P5.1 part can be installed and tested
separately, thus easing the upgrade process.

I will be providing: the board, schematics, parts list, board layout,
background, theory of operation, patch wiring instructions, formatting
instructions, and probably a disk with HwNotes and other stuff. The text part
of the instructions will be posted for review.

Work is progressing on modifying the hard disk driver to USE all four hard
drives. It will not be available with the initial release of the board, but
will be posted in source form if and when I get it done. Until that is
available, you will be restricted to only two hard drives.

As far as the fan wars go, I have designed the board so that the cable headers
will match up with the fan grates. I will recommend in my instructions that
the fan be removed, the board be placed in that space, and that the existing
internal hard drive be mounted in the external enclosure with the second drive.
There will, of course be a melt down disclamer :)

As always, it's never too late to get your opinions and advice in on the
design of this board. If there are bugs found later, I would expect at least
50% of them fixable with a PAL change. Any of what I have described here may
change between now and when I get evrything working.

John
-- 
John Bly Milton IV, jbm at uncle.UUCP, n8emr!uncle!jbm at osu-cis.cis.ohio-state.edu
(614) h:294-4823, w:764-2933;  Got any good 74LS503 circuits?