Curtis 1219-4860 PMC controller repair

[GriffinRU]

- deburr and apply heatsink grease (not sure why but OEM (CURTIS) didn't use any heatsinking compaund... maybe it wasn't available in late 90 or whatever was another reason... 40A can sink through screw no problem but poor heatsinking might cost you some - especially for diodes...

[BrianG]

Haven't seen you here for a long time. Nice repair job!

Originally Posted by GriffinRU

...not sure why but OEM (CURTIS) didn't use any heatsinking compaund...

IIRC, normal thermal compound is electrically insulating and if they are using the tab of the FET as one of the conductors, then it would be like adding resistance.

Could also be related to the environment that the controller is meant to operate. Maybe the TIM stuff back then had some issues with certain temps or humidity levels making it more of a thermal hindrance than help.

[GriffinRU]

Originally Posted by BrianG

Haven't seen you here for a long time. Nice repair job!



IIRC, normal thermal compound is electrically insulating and if they are using the tab of the FET as one of the conductors, then it would be like adding resistance.

Could also be related to the environment that the controller is meant to operate. Maybe the TIM stuff back then had some issues with certain temps or humidity levels making it more of a thermal hindrance than help.

Hey Brian,
You are correct about electrical contact, but they got it wrong from the beginning, given their heatsink configuration they could end up with FET's leaning away from heatsink if it was assembled not strait or due to heat expansion. They should use special clamps to keep FET's at contact and you can tell by the oxidation behind FET's and diodes tabs...
I think for 40A per FET (OEM) mounting screw was more than enough and use the rest for heatsinking or at least 1/3 by 2/3 - large area under die.

Thanks for feedback, and glad to be back,
-Artur