Notes on the IDE controller schematic...
updated 12/16/07 9/20/08 3/14/09 11/16/10 6/14/11

The entire circuit has been tested with the hpdiskc2.asm code on a
"Paul Rev5" development board and works with the level shifters connected
to an old "+16 BIT DUP REG" interface board. The proper jumper settings
for such a board should be...

W1 to +12, W2 to -12, W3 to -2
W4 to B, W5 to B, W6 to B, W7 to A
W9, W10, W11, W12 removed
W13 to A, W14 to A

The board it was tested with did not have W9-W14 and could not be configured
for "transparent inputs", causing a failure in the status read function as
implemented by the current code. To work around this the HP-IPL/OS <IDE word
which reads the disk status was modified so that it always returned 0.

The IDE drive must support LBA addressing, this code can't be used on very
large (over 120 gigs) or very small (only CHS addressing) drives, I used
an 80 gig IBM DeskStar "ATA/IDE" model IC35L040AVVN07-0 dated 21-APR-02.
Smaller 800meg drives I had would not work.

All resistors 1/4W 5%, 1N914 diodes used (1N4148 might be better), chips
were C and HC family mainly because that's what I had but the level shifter
inverters should definitely be CMOS due to the hi-Z nature of the signal.
A 74LS04 or similar inverter would probably be OK for the IDE interface
portion of the circuit, a '14 was used because I had one. All IC's should
be socketed. A mix of 2N5551 and MPSA42 transistors were used for the level
shifter outputs but any common NPN should work. The emitter follower for
the flag output was a generic NPN with an unknown part number.

The Paul Dev5 board was obtained from...
http://www.pjrc.com/tech/8051/board5/index.html

The LED's connected to PE.0 to PE.7 on the stock Paul board must be removed.
They can be reused for the status and drive activity LED's. The logic of the
status LED's are reversed, light off is true. All of the circuitry for the
TTL-compatible portion was wired on the prototype area, the level shifters
were wired on a separate board placed underneath the dev board and connected
via ribbon cables. The hpide_layout.gif file depicts the arrangement I used
on the proto area, arrived at after several iterations of sketches.
If building this circuit it's probably OK to wire the entire circuit on
a separate board for less risk, keep the interconnects short as possible.

The stock Paul board requires a CR from the terminal to set the baud rate,
even if the code is modified to auto-start. This fixes that...
http://www.pjrc.com/tech/8051/fix_baud.asm
The initial lines should be changed to something like...
.equ    baud_const, 244         ;9600 baud w/ 22.1182 MHz
.equ    locat, 0xF000     ;Location for this program
... and assemble with AS31. Modify the hpdiskc2.asm code to autostart
(past the .equ's) and assemble with AS31 and upload both hex files into
flash. It now autostarts. Erase the flash (by shorting a jumper on powerup)
to remove or whenever changing the code, programs can't overwrite themselves
because flash bits can only change in one direction.

The HP mini should be powered first to avoid sending a random command to
the HP-IDE interface, particularly if level-shifters used. There is an issue
I'm still pondering about how or if to solve... when the Paul board is off,
the HP will feed enough power to the board through the level shifter input
diodes to actually power up the board to a prompt - but not enough power
to activate the flash chip. Connecting a 220 ohm resistor between +5 and
ground should stop that from happening (haven't tested). The board still
resets when power is actually applied so probably not a real issue.

If the interface electronics and the disk drive are separately powered,
power up the interface first, then the drive. Otherwise the drive gets
confused and doesn't become ready.

[11/16/10] This circuit should NOT be connected directly to an HP without
some kind of interfacing or level shifters... 12V "+16B" boards have 10K
pullups on outputs and 1K pullups on inputs, 5V "microcircuit" boards have
1K pullups on outputs and 316 ohm pullups on inputs - this can result in
damaging currents if not properly interfaced. From what I can tell, the
level shifter circuits shown on the schematic and present firmware "should"
also work with a 5V ground-true microcircuit board, provided it is jumpered
so that the command and flag signals are also ground-true. Not tested.
Various HP interface boards (even of the same type) often behave differently
and may require hardware and/or firmware changes to work with this hardware.

. . .

9/20/08 [edited 3/14/09] - Added project notes...

The .obj files have been removed, of limited use considering the source
quite likely will need adapting to the particular application/board/etc.
The source files can be edited as needed then assembled using the
AS31 assembler to produce...

  FIX_9600.hex - the 9600 baud init program             (F000)
  HPDISKC2.hex - a non-autostarting version of hpdiskc2 (8000)
  HPDISKCA.hex - an autostarting version of hpdiskc2    (8000)

   (after ; Main Program uncomment: .db     253,255,0,0
                       and comment: ;.db     35,255,0,0 
    and reassemble to make the autostarting version)

Once the electronics are wired connect a serial-port terminal emulator and
press a key to get the PaulMon prompt. For initial testing use the terminal
ascii send function to send HPDISKC2.hex and connect the debug jumper. Use
PaulMon's Run function to execute, if everything is good a debug prompt
appears where basic functionality of the drive and interface can be tested.
For permanent use without having to connect a terminal, erase the flash and
load HPDISKCA.hex and FIX_9600.hex and remove the debug jumper.

Connecting a VDRIVE module [edited 6/14/11]
--------------------------

The hpideusb.asm firmware provides all the features of the previous
hpdisk firmware and permits connecting a VDRIVE2 module for using a
common USB thumbdrive for file storage and for backing up and restoring
the IDE disk. The VDRIVE2 connects to the Paul board as follows...

Pin 1 (ground) to Ground
Pin 2 (data out) to Port F bit 3
Pin 3 (power) to +5V
Pin 4 (data in) to Port F bit 4
Pin 5 (clock) to Port F bit 5
Pin 6 (select) to Port F bit 6

Port F is Port C of the 2nd 8255 chip.
To make use of the optional "Auto" program then a 10K pullup should be
connected from +5V to Port F bit 2, with a switch pulling the line low
to automatically start the interface code without having to make the
interface code itself autostarting.

. . .