The PDP-11 is here
bnl returned to Nottingham last week and brought my PDP-11 along with him. I ignored it for a few days, given that I was very busy, but on Friday decided to power it up.
Oops - the circuit tripped. Probed about a bit, realised the EMI filter had gone short, which is apparently quite common with old EMI filters. I was pretty sure they could be removed but emailed
classiccmp anyway to check - apparently it would be fine, so out it went.
It worked! The PDP-11 powered up in ODT and I was able to type in some machine code and run it.
Today bnl and I decided to build a cable to see if we could get the RL02 drive
FAULT light to disappear. Apparently, it should not appear if the clock line is there, and it was, but we did not succeed.
Debugging had to stop when the M7195 SLU card decided to stop working. Hmm - looks like I will need to find a replacement, unless I can figure out what has gone wrong with the card.
<2017-10-01 Sun 19:07> Decided to check the voltages from the power supply last night, and surprise surprise, the +12V rail was floating at about +4.4V. Not very good... BUT, that means that the SLU is probably fine (I hope...), and the power supply is probably easier to fix than the SLU. Marc (bnl) came over this evening and we traced it back to a MJE15030 transistor which we think is faulty. I'll order some new ones and replace it. Hopefully that will fix it. There is about +80V going into it, so it is as if it isn't responding to the feedback from the primary winding of the small transformer after it. Here's a snippet of that part of the circuitry:
I don't really know anything about the failure mode of transistors... I thought they only failed open or closed. I've tested the hFE and it came back as ~86. According to the spec, the minimum this value should be is 40 but it doesn't say what the maximum is, and most transistors I've tested have been around 8. I think this transistor is the problem, and as I've said, I've ordered a replacement, but this hFE stuff is complicated and making me sceptical, perhaps with no good reason.