23 Feb
2008
23 Feb
'08
12:58 p.m.
Tim Riker wrote:
Yes, I used that in part as I RE'd the C power supply.
I was in a similar situation to you, I didn't have the manuals specifically
for the C but (eventually) found the B manuals on bitsavers.
(I'll claim my schematic is layed out more sensibly, but that's
just my opinion .. depends on one's preferences.)
Brent Hilpert wrote:
The resistor to +12 is provided on the 12531 board, I just made the connections
to force CTS/DSR/DCD high in case one's terminal needs such.
No, there's nothing in the way of flow-control going on.
--
I had a timing problem with the HST initially, it
was resolved by sending 2
stop bits and now runs at 19200 bps. There's some explanation on:
http://www.cs.ubc.ca/~hilpert/e/HP21xx/io12531BCD.html
if you haven't seen it already. I take it you're using a 12531C (HST) if
you're
running at 2400. If you're already sending 2 stop bits though .. wonder if the
oscillator could have drifted far enough out to be a problem.
Yes, I saw this before. I don't have my cable wired the same as yours. I
don't think my CSR/DSR/DCD go anyplace useful. Does your cable actually
get you valid CSR? or are you just forcing it? ie: you connect to "U -
+12V via 1k". I'll check the schematics, just have not yet. :) An
'A'! .. even better. Found your photos again, see you've effectively fixed
the frame .. thank goodness for symmetry. I made schematics for 2116C power supply,
should be largely applicable to the A, at least to get an idea of the organization,
if of any help.
There are schematics for the B power supply on page 5-91 of:
http://www.bitsavers.org/pdf/hp/2116/02116-9153_2116B_Vol2_Oct70.pdf
You've seen that I presume? That's what I'm working off. Al was trying
to track down an A version of Vol2 that he though he had, I've not heard
back on it. I could really use this if someone out there has it. :)
I was lead to believe that the C (perhaps only later C's?) used a
switching power supply. I assumed, perhaps incorrectly, that there are
different voltage outputs required for the A, B and C due to the
different core memory designs. From what I understand:
The A has 1 or 2 vertical 4kw stacks
The B has 1 or 2 horizontal 8kw stacks
The C has no stacks, but uses HP-2100[AS] compatible core cards
Both the A and B can use a 2nd chassis to add 1 or 2 more stacks. The C
can fit enough cards to take it to 32kw internally.
The C still used linear but - as you suggest - with some voltage mods for the
changed core memory. The 2100's were the first to use switchers, TMK.
The B used anti-coincident addressing with the core, and I believe this is how
it achieved the doubling of memory capacity over the A. There may have
been little or no change to the power supply from the A to the B, as
anti-coincident addressing just requires (in principle) a little twiddling with
the address decoding.
If you're working on your A again sometime, I would be interested in some
information regarding board part numbers to fill out some of the question marks on
http://www.cs.ubc.ca/~hilpert/e/HP21xx/HP2116.html
regarding the 'A' model, for the sake of comprehensiveness.
Correlating some of these part numbers from your A with those of the B might
help resolve whether or not there are any differences in the power supplies
between the two. (The above page has a list of board part numbers.) Specifically:
Power Supply board A301
Power Supply board A302
Sense Amplifier
Inhibit Driver
Address Driver
Too bad about the missing blinkenlights cover .. need to find a real junker to
get one from, or, in the realm of plastics and silk-screening, it
might not be too difficult to recreate one.