• Please review our updated Terms and Rules here

PDP-8/I at the RICM

We were able to substitute an off-the-shelf automotive alternator diode for a failed DM2 on the CMA 8/I until we could get the correct part. Doesn't help with connector/wiring harness losses but at least addresses the diode issue. DEC didn't seem to provide too much of a safety margin on some of these devices and death from overheating was not unusual.
 
djg said:
Is that at the regulator or backplane? I've improved it some at the backplane by tracking down the drops and improving the connections to reduce losses. My regulators were 5V but losses a problem.
Click to expand...
The 4.7V was measured at the faston connections at the flexprint cable adapter on the power supply. The manual says 4.75V is OK, but I imagine that should be measured at the chip pins.
 
We found that most of the shift signals to the M220 boards are enabled when you press LOAD ADDRESS. That would make a mess of the Register Bus data, and probably explains why the front panel doesn't work. We will start tracing the shift signals on Wednesday.
 
m_thompson said:
We will start tracing the shift signals on Wednesday.
Click to expand...
Most of the shift signals are generated on sheet D-BS-8i-0-5 Shift and Carry Gate Control, and are OP1 ANDed (in a couple of stages) with MB bits to determine the number and direction of the shift. The second stage of gating allows the EAE and the TT shifts to be injected.
 
When LOAD ADDRESS is pressed, DOUBLE RIGHT ROTATE is inactive, DOUBLE LEFT ROTATE is active, RIGHT SHIFT is active, LEFT SHIFT is active, NO SHIFT is active, and AND ENABLE is is active. This 8/I doesn't have EAE so all of those signals should be inactive, and the MB contains all zeros. Looking at the schematic the only condition that I can think of that would result in all of those signals being active is lots of 7440s failed on the M617 Flipchip in slot E31. I can test that Flipchip on Wednesday.
 
That feels to me as if +3v(40) has been lost. Inputs float high, though, so it would need to be pulled low somehow.
 
vrs42 said:
That feels to me as if +3v(40) has been lost. Inputs float high, though, so it would need to be pulled low somehow.
Click to expand...
Vince, I think you might be right. My guess is that +3V(30) and +3V(40) come from the two 3V sources on the M617 Flipchip in slot E31. If one of the 330 Ohm pull-up resistors in the voltage divider failed or there was an open solder joint, then the remaining 750 Ohm resistor would be a pull-down and make lots of inputs to the NAND gates active. It should be easy to check for voltage on the E31U1 and E31V1 3V output pins, and check the state of the inputs to the NAND gates.
 
If the lines are stuck, I'd look for a cracked resistor or solder joint. If they cycle with the active low KEY_LA_, I'd look to see if the backplane wires overlap.
 
We tested the M617 from slot E31 in a FlipChip tester. It said that the F2 input was shorted to ground which made DOUBLE LEFT ROTATE true. The short also pulled the 5V(40) source to ground which pulled inputs to RIGHT SHIFT and LEFT SHIFT and made those signals true. NO SHIFT was the only one of these signals that should have been true. Having the others true turned four inputs to the data multiplexor on and made a mess of things. Many of our spare 7400 chips got lost during the move to our new lab space. I found a 1970 vintage Motorola DEC7440 in one of our DEC field service kits. The M617 passed testing, and now the PDP-8/I boots the 4k monitor from DECtape and runs FOCAL and FORTRAN.

Thanks for all of the help with debugging.
 
You must log in or register to reply here.
Back
Top