I also looked at the datalogs, and yes it does appear that more soak time would be appropriate to help bring the dut package temperature down closer to -40°C.
My only concern would be the condensation build up on the backside of the board. It builds up to the point of becoming a liquid (water).
I have experienced this while try to test at -40°C here in NY. It causes issues with testing, parts will fail for continuity, leakage ..ect. It can also cause
damage to the setup. I’ve had to replace relays on one board to correct for some test failures I was seeing afterwards. Also, switching the thermal stream to hot does
not get rid of all the water build up on the back side of the board. I had to remove the loadboard from the test head and dry it out manually with a heat gun.
However, if we still want to test these part down to -40°C we should keep all the above mentioned items in mind. The back of the loadboard should probably
be check after a couple of parts are tested. I don’t know how practical this will be.
We are reading the correct reference diode for the temperature reference as well as the external diode for the dut.