6 Jan
2005
6 Jan
'05
1:51 p.m.
Wai-Sun Chia asked:
<snip>
Not yet. I have 2 spares for this board, each slightly different from the
one that was installed. None of them 'work' correctly, but I suspect
that this may be due to other problems the machine had (that are now
corrected).
I've not installed or re-tested the 8K display extensions since resolving
the
last issues that got the machine booting 4K Imlac programs.
Currently I'm working my way through the available software and seeing
whats what. Then I'll jump back onto the display procesor so I can run
the 8K debugger.
7805!? Thats funny.
No, the main 5 volt supply uses a 723 linear regulator controller that
drives
a stack of 2N3055 power transistors to deliver 15 amps at +5 volts.
The 'quick fix' for the regulator board was to swap it with a working spare.
I'll replace the regulator chip in the other board in time, and perhaps add
a
small cap to the feedback pin for HF bypass to ground.
This will prevent any noise from the logic from pushing the regulation
feedback
loop into oscillation.
I designed the serial board, but trust me, this is not rocket science.
Mainly
I copied a small bit of interface logic from the original schematics, then I
tacked on a modern microcontroller and wrote a little code.
This is now I'm now able to reach inside and flick a switch to boot the CPU
The Display
Processor was another story. Any attempts to draw short
vectors with a -y displacement
produced +y axis vectors. This was chased down to an original Imlac
option board that expands the
display processors address space from 4K to 8K. With that option pulled
(as the machine could only address 4K anyway) I finally got the machine
to reliably display the HELLO test pattern.
Did you manage to debug that option board also? But the basic
stability problem soon returned. One recent evening I got
a little lucky and found
that the linear regulator for the +5 volt supply was dropping out of
regulation due to a high frequency
oscillation on one of the pins of the regulator IC. This was repaired,
and I once again had a solid
machine to work on.
How did you repair this? Did you replace the 7805 (I presume) or did you
put a bypass cap at the input? <snip>
No, the 'stock' buffers (2 sections of a 7404 in series) work fine, but
there
was no display buffers for PC.3 and MA.3 (Imlacs have their bits numbered
backwards!).
These were needed, but omitted from the design of the SRAM / 8K extensions
added back in the early 1980's. Essentially the conversion and memory
extension
never worked exactly right, and the machine was stored that way for many
years.
I had to add a tiny board with a 7408 between the MA.3 and PC.3 signals
and the display console connector to buffer these bits.
I really can't beleive Imlac omitted the base current limiters for the
console
lamp dirver transistors.
Later model Imlac's had LED panels, so I'd guess that this extra level of
buffering is not needed on a PDS-1D.
<snip>
Imlac's hardware engineers also decided that
the transistors driving the
console lamps did not need
base current limiting resistors, so they pull a TTL output to ground when
lit. This nailed the new
address bit low as far as the rest of the machine was concerned. A set
of TTL buffers restored
the new address bit with an absolute minimum of modifications.
So the TTL buffers were borked? A plug-in
replacement for the stock Imlac serial I/O board (110 baud!)
was designed and slipped
into the correct slot, with the ability to load Imlac software
Wow. You design this board yourself or did you copy from the original
schematics? from a binary image stored right on the serial
interface board itself, then
flick the switch back for a 'normal' serial
interface, that now has modern
baud rates and crystal controlled timing.
All in all, very impressive debugging/detective work.
Congrats again!
/wai-sun
Thanks!
As many people on the list with HP's know, I'm always willing to help others
get their machines running.
If anyone is in the general Boston area and needs a hand, let me know what
I can do to help.