Although there is no ROT or HBK alarms, check fuse FU1 on the servo amp control board anyways.
Make sure all servo amp and e-stop board connections are seated.
Check to see if the discharge resistor is actually getting hot enough to cause the thermal switch inside it to open. I can't think of any motion programming of an ArcMate 120i that could regenerate enough energy to overheat the discharge resistor since it's the same size resistor as the one used in an S-430iF. If it is getting hot, change the servo amp.
If the resistor is not hot, then there is a possible intermittent problem in the thermal switch circuit. That circuit has the +24VF ('F' stands for fused) originating from the servo amp control board (FU1), going thru the e-stop board, to the DCR thermal switch, back thru the e-stop as DCRTH and back to a system input on the servo amp control board.
To check the DCR thermal switch, disconnect connector CRP10 on the servo amp and measure the resistance between CRP10-20 (24VF) & CRP10-8 (DCRTH) on the wire side - you should measure just a few ohms. If it checks good, try changing the servo amp control board.
After reading your reply, I suggest first making sure the rear compartment of the controller is clear of debris, the vents & louvers are clear, and the fans are working. The rear compartment can get hot if there is no air flow.
A laser/infrared thermometer might make checking the temperature of the DCR easier.
To answer your questions, the CRP10 connector (upper right of the servo amp control board) provides power and some signals to the servo amp. The controller disconnect needs to be off to make the resistance check of the thermal. As long as the batteries are good, you should not lose mastering either checking resistance or replacing the servo amp.
One last thing - some thermal switches can weaken with repeated activation. A laser thermometer can let you know if the thermal switch is now activating at a lower temperature than rated. I'm not sure on the R-J3, but the R-J2 DCR rating was 180degC.
