Then I checked the "modern" 0.1 ohm resistor on the lower side of the PCB fitted in place of a broken trace. One side is grounded. So I followed ground traces to nearby components suspecting a ground loop but this is what I saw:
Almost 5 mm of missing (ground) trace! And it is on the biasing network around the dual-slope integrator. The result was an impedance mismatch causing stray currents and disturbing the integrator.
A piece of resistor leg was promptly used to rebuild the exposed trace and I anxiously powered up the multimeter (still without the AC converter board). Numbers were shown, didn't flicker or runaway but did not make sense. OPS! I forgot to press one of the "mode" buttons. So I got a stable, not null reading.
I shorted V/ohm input and turned the zero calibration pot as described in the manual. I had moved it around when troubleshooting. Voilà! The Weston Schlumberger 1240 multimeter is with me again!
I have no idea how I managed to pull away that piece of track. Perhaps with time it has "glued" to a floating cable coming from the front panel, or .. well, it's fixed now!
