The warmed up DMM's are in the thermal pic's below. These were an Agilent U1233A and an Agilent U1241A. I used a U1252A to keep an eye on the voltage.
During the soak-testing the current reading on the DMM's used to measure current steadily increased as they warmed up.
The power supply current reading did not change much at all. The voltage reading crept up marginally.
As most of you are probably aware, a DMM measures current by measuring the voltage over an internal shunt.
So what we were seeing with the slowly, but steadily, increasing reading on the DMM's was the internals of the DMM's getting warm, the internal resistance increasing with this temperature rise and the power supply giving a bit more voltage to keep the current at max. The test leads will also play a part in this resistance path. We can see that the resistance of the leads and the DMM input is a combined 0.1 Ohms.
The normal picture below shows the DMM after it had been running for a while and the current reading had drifted up a bit. This was while we were testing at a setting of 1V, 5.1A. The power supply went to constant current mode to deliver its max current and the compliance voltage necessary for this (0.5465V) is displayed on the U1252A.
It would have been more interesting if it hadn't all behaved as expected, but all ended up as hoped for.
We still have a working power supply and I got to play with a thermal camera.......