9 Mar
2005
9 Mar
'05
5:43 p.m.
I added support to SPCWR3.PA for the VC8/I.
http://www.chd.dyndns.org/pdp8/VC8/SPCWR3_vc8i.PA
This requires an EAE and now I find out that my untested EAE does not work.
The first sign of problems was that the processor paused while running
and the only way to stop it was to turn off power. The HALT switch did
nothing.
I got a copy of MAINDEC-8E-D0LB-D which is EAE test 1 and it locked up
pretty much the same way. I removed the EAE and the diagnostic halted at
a reasonable place insteading of locking up.
The EAE stops the main processor sequencer and then does it's thing and
when it is finished, restarts the processor sequencer. I seems to me
that this handoff is not occuring properly. The front panel shows that
the processor is stuck in Execute (E) State with a 7 in the IR and the
PC within one of an EAE instruction.
This not gonna be easy to fix. The over-the-top connectors prevent the
use of an extender card, so I can't probe anything easily.
Since the MAINDEC tests won't run, I will try to figure out which if any
of the EAE instructions function and which to not. Since, so far,
everything seems to hang, it might be something fundamental, and that
might be easier to find.
Suggestions, comments please.
-chuck
10 Mar
10 Mar
3:17 p.m.
This not gonna be easy to fix. The over-the-top
connectors prevent the
use of an extender card, so I can't probe anything easily.
A general comment. When I don't have the right extender, or in cases like
this when you can't use one, what I generally do is solder some wires to
useful points on the board, put the board back in, and probe the free
ends of the wires. Don't try this on fast stuff, but for the 8/e it
should be fine.
-tony
23 Mar
23 Mar
4:40 p.m.
Tony Duell wrote:
I thought of that... Just don't relish the idea... You do what you gotta
do though.
This not gonna
be easy to fix. The over-the-top connectors prevent the
use of an extender card, so I can't probe anything easily.
A general comment. When I don't have the right extender, or in cases like
this when you can't use one, what I generally do is solder some wires to
useful points on the board, put the board back in, and probe the free
ends of the wires. Don't try this on fast stuff, but for the 8/e it
should be fine.
-tony
I did some tests with the various EAE instructions and it seems that
when the main timing generator is stopped, either the EAE timing
generator does not start or it does not restart the main timing
generator. Any instruction that hands off control seems to hang the
processor.
-chuck
10 Mar
10 Mar
5:01 p.m.
On Wed, 09 Mar 2005 20:43:12 -0500, Charles H. Dickman
<chd_1 at nktelco.net> wrote:
I added support to SPCWR3.PA for the VC8/I.
http://www.chd.dyndns.org/pdp8/VC8/SPCWR3_vc8i.PA
Nice! Thanks!
This requires an EAE and now I find out that my
untested EAE does not work.
Doh! I didn't even think of that. :-( I don't have a single EAE for
_any_ machine.
The closest I could think about getting would be to bum some M-series
boards from a working machine/disk controller and stuff them in my -8/i,
but ISTR there are a few wire changes to make as well, so I think I'd
rather not do that if I thought I'd have to remove it. In any case, I don't
have a genuine VC8/I in the first place, or it might be attractive to try.
Ah, well... maybe _someday_ I'll run across the parts I need to pull this off.
I got a copy of MAINDEC-8E-D0LB-D which is EAE test 1
and it locked up
pretty much the same way. I removed the EAE and the diagnostic halted at
a reasonable place insteading of locking up.
Sounds ominous.
This not gonna be easy to fix. The over-the-top
connectors prevent the
use of an extender card, so I can't probe anything easily.
I don't know if you have access to any DEC module blocks, but it might be
possible to rig up a flexible over-the-top connector if you have a couple of
single-slot wire-wrap connectors and some wire-wrap wire.
Alternately, if you have some single-height proto cards, one could manufacture
a couple of over-the-top extension cables, space the CPU out two more spaces,
put the EAE board under test in the middle of the three blank spaces on a quad
extender, and hang one paddle card upside down under each over-the-top connector
to route the signals a few inches in the air over some ribbon cable...
Hopefully, you can envision this arrangement. Write me offline if it's not
clear enough and I'll try to describe it better.
I don't think any of these signals are critical enough that a couple of ns of
wire is going to cause any problems.
Unfortunately, I can provide no insight on the EAE fault itself; I
have no experience
with it.
-ethan
23 Mar
23 Mar
5:01 p.m.
Ethan Dicks wrote:
This version does use the EAE. I have another that is MUCH simpler that
does not seem to need EAE.
It seem like there is something wrong with the timing generator in the
EAE. Since NOTHING works when the EAE timing generator is needed, maybe
it is something simple. :-)
Lately what I need most is time...
On Wed, 09 Mar 2005 20:43:12 -0500, Charles H. Dickman
<chd_1 at nktelco.net> wrote:
It was nothing really... :-)
I added support to SPCWR3.PA for the VC8/I.
http://www.chd.dyndns.org/pdp8/VC8/SPCWR3_vc8i.PA
Nice! Thanks!
This requires
an EAE and now I find out that my untested EAE does not work.
Doh! I didn't even think of that. :-( I don't have a single EAE for
_any_ machine.
The closest I could think about getting would be to bum
some M-series
boards from a working machine/disk controller and stuff them in my -8/i,
but ISTR there are a few wire changes to make as well, so I think I'd
rather not do that if I thought I'd have to remove it. In any case, I don't
have a genuine VC8/I in the first place, or it might be attractive to try.
Ah, well... maybe _someday_ I'll run across the parts I need to pull this off.
From what I saw in the 8/I printset, you need 3 cards: two A607, and
one M701 or A701 (one place its and A and another place its an M). The
A607 might be generic, but the M701 is very specifically the VC8/I
display controller.
I got a copy of
MAINDEC-8E-D0LB-D which is EAE test 1 and it locked up
pretty much the same way. I removed the EAE and the diagnostic halted at
a reasonable place insteading of locking up.
Sounds ominous.
This not gonna
be easy to fix. The over-the-top connectors prevent the
use of an extender card, so I can't probe anything easily.
-ethan
-chuck
7214
days inactive
7228
days old
4 comments
3 participants
participants (3)
-
ard＠p850ug1.demon.co.uk -
chd_1＠nktelco.net -
ethan.dicks＠gmail.com