27 Jul
2011
27 Jul
'11
1:25 p.m.
Hi.
I try to get my "new" PDP-11/34a back to live. I did a quick PSU test
today with dummy loads. All voltages, LTCL, DCLO and ACLO where OK so I
pluged in the logic boards. But the machine does not respond to the
knobs on the programmers console. Pressing HLT + CNTRL brings the
machine not to halt, but the "Bus Error" LED of the console lights up.
Grant continuity cards and NPR jumpers are in place.
It looks like I have the wrong bus terminator: There is a M9301-YF at
the CPU side and a M930 at the other bus end. It seems I need a M9302
in place of the M930? This would explain the bus error.
Err no. If you have an N930 terminator, a missing grant won't cause a
bus error. The amchine should run properly with such a terminator
(actuially, on a short Unibus, you can get away with no terminator at the
'far end' for testing. I don't recoemnd running a machine that way though.
I took my scope to have a look at the ripple of the voltages.
Here is what I got on the second +5 V PSU:
X = 0.2 V / div ; Y = 2 ms / div
http://www.unixag-kl.fh-kl.de/~jkunz/tmp/5_Volt_Ripple__y=0.2V__x=2ms.jpg
X = 0.2 V / div ; Y = 10 us / div
http://www.unixag-kl.fh-kl.de/~jkunz/tmp/5_Volt_Ripple__y=0.2V__x=10us.jpg
I get similar pictures on all voltages. Both +5 V power bricks (H7441)
show similar ripple as well as +15 V and -15 V.
There are spikes with up to 1.2 Vpp and it is clearly related to the
50 Hz lines frequency. This doesn't look that healthy. I had a look at
the BA11-K power distribution schematic. The two +5 V PSUs are
independent and run of different windings of the mains transformer.
I am a bit surprised to see about the same fault on all PSUs. I am no
SM-PSU expert, so I am now at the point where I have to ask for advice.
First quesiton. What is the ground lead of the 'scome conncted to ? Is it
connected to the system 0V rail?
These supplies do suffer from dried-up electrolytics. Mains-frequency
ripple is normally provblems with the capacitor at the output of the
bridge rectifier (the very large one mounted horizontally on the PCB),
spikes at switching freqeuncy are often due to the capactiro at the
output of the 'brick'. If you have an ESR meter, it's worth checking them.
-tony
