PDP-11/45: LA30 repair II

Sun 26 November 2017 by Fritz Mueller

Okay, first thing to debug today is the ready light. This is lit by RDY LITE L, pin A16D2 (lower right of sheet M7721-0-1 in the LA30 engineering drawings). Logic probe showed this was correctly asserted low. Pulled the lamp and checked it with bench supply, and it checked fine. Verified +10 and ground at the lamp socket as well, so why isn't it lit? Turns out it is polarized, and the socket is soldered in backwards (?!). Corrected the socket, and ready light is working.

Noticed that ribbon is stalling occasionally. Ribbon tension seems to be overcoming the clutch on the take-up side. It seems tensioning drag on the inactive side is too high. Not sure what to do about this one yet; there is not much adjustable within the clutches, and the service manual only recommends replacement if they are out of spec (yeah, good luck finding one!)

Repeated the experiment with loopback jumper (A15R2 to A15C2). Turns out I had miscounted backplane pins the first time. With jumper placed correctly, I am now get a printing response to the keyboard. Not quite right, but definite progress:

Here I had typed :;L, and then some other letters. You can see evidence of correct pins firing for the first three characters, though either head movement or pin timing is off. Letter spacing appears more or less correct for 80 characters per line.

Okay, hooked up the logic analyzer, and started to take a look at the character generator clocking to see if all the the columns are being clocked out correctly. The logic analyzer shows a malfunction consistent with the print behavior: character column clocking resets after two columns rather than proceeding through all five:

This signaling is mediated by a ripple counter on the M7724 character generator card:

So it looks like one or more of these 7474 quad flops has failed. I note on my chargen board that these are early 70's Nat Semi parts; Mattis had a very similar issue (search "7474" on this page) on his LA30 chargen with the same parts.

All I have time for this weekend; next time I'll get the chip clip on these for a closer look, then pull and replace the baddies.