Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Tesla Roadster PEM Tech Tip Alert

This site may earn commission on affiliate links.
In every Tesla Roadster PEM we have serviced to date there is evidence of failing thermal insulating material between the IGBT Inverter Transistors and Aluminum Heat Sink. Corrective action is PEM removal and disassembly, and replacement of the thermal insulating material with an upgrade, higher quality material more suitable for automotive wide temperature application.

2016-08-26-Tech Tip Pem.jpg
 
  • Informative
Reactions: dhrivnak and markwj
Ok sounds interesting. Is this a preventive maintenance procedure done in your shop??

Yes it is Botbldr45.

Once we noticed a clear pattern of deterioration, we now disassemble all PEM inverter sections and replace this material as a standard operating procedure. We sent this Tech Tip to 64 Tesla Service Centers to share this information, and hope they also replace this failing material.
 
Curious... You noted Roadster versions 2.0/2.5 in the versions above, so the 1.5 does not exhibit this issue of the deteriorating thermal insulation?

The only 1.5 we have had in the shop Wiztecy, owned by one of our Managers at the time, preceded our realization this material is deteriorating. Unless Tesla switched insulating material during the production years, it is likely all PEMs will need this upgrade.
 
@racevpr
Did you also test ceramic insulating pads for the IGBTs?
From the net I found the hint, that IXGK120N60B IGBTs are used, which are obsolete today. These IGBTs have an Ucesat of approx. 2.1V. Have you tried other similiar types with lower Ucesat (found some types with approx. 1.2V), which should also reduce the power loss and produce less heat?
 
@racevpr
Did you also test ceramic insulating pads for the IGBTs?
From the net I found the hint, that IXGK120N60B IGBTs are used, which are obsolete today. These IGBTs have an Ucesat of approx. 2.1V. Have you tried other similiar types with lower Ucesat (found some types with approx. 1.2V), which should also reduce the power loss and produce less heat?

We are aware of the next generation IGBT's Austrianer and you are right - a better product is now available.

We will be extensively testing these newer, more efficient IGBT's in one of our test mules and report findings on TMC. Less heat dissipation is of course a plus.
 
@racevpr
Did you also test ceramic insulating pads for the IGBTs?
From the net I found the hint, that IXGK120N60B IGBTs are used, which are obsolete today. These IGBTs have an Ucesat of approx. 2.1V. Have you tried other similiar types with lower Ucesat (found some types with approx. 1.2V), which should also reduce the power loss and produce less heat?
IXGK120N60B3 is the recommended replacement. It has a U ce sat of 1.6v.
 
For folks no where near Phoenix, who is available to do similar work as you suggest? I think my PEM needs cleaning and probably things like mentioned here as the car is higher milage. Pretty sure the local Tesla service folks aren't gonna be helping us; I'm still a week out for a simple key recoding job to be performed...
 
I think my PEM needs cleaning and probably things like mentioned here as the car is higher milage..

The 1.5 PEMs stay way cleaner since they pull air from up high above the rear fender, the 2.x's get air that's blown or forced from the lower rear of the car which gets all the dirt, leaves, and other road debris lodged in the PEM which beings limiting the cooling. Its been 2.5 years since I pulled off my PEM and when I did pull it after 1 year from the annual service it was still totally clean. I may pull it off next April to inspect, however if I'm not seeing any elevated PEM temps there's no reason to so, and with that I may push it out another year.... Taking the PEM on/off less is better on the connectors which we've found evidence that worn fan connectors (at least on the 2.x's) start building up resistance, which creates heat that leads to burned out connectors and potential PEM failures.

Do you know when the last annual service/mileage of when it was performed at Tesla on your Roadster? If you don't you can have Tesla pull the service records.
 
In every Tesla Roadster PEM we have serviced to date there is evidence of failing thermal insulating material between the IGBT Inverter Transistors and Aluminum Heat Sink. Corrective action is PEM removal and disassembly, and replacement of the thermal insulating material with an upgrade, higher quality material more suitable for automotive wide temperature application.

View attachment 191882
The name Gruber makes my skin crawl. This based on previous threads with them. Would consider waiting for tesla to confirm
 
  • Disagree
Reactions: dhrivnak
My car seems to run with PEM temps in the right most blue but I'll get more data there; it is what made wonder on proper cooling

Sounds normal for the 1.5 Roadster and it really depends / sensitive on the outside ambient temps. The 2.x's with the fan/blower upgrade run cooler and when the temps get pushed higher, the recovery is very fast. But Tesla is doing that by pushing more amps to the PEM fan which on the forum some believe is one reason for the fan connector failures.
 
In every Tesla Roadster PEM we have serviced to date there is evidence of failing thermal insulating material between the IGBT Inverter Transistors and Aluminum Heat Sink. Corrective action is PEM removal and disassembly, and replacement of the thermal insulating material with an upgrade, higher quality material more suitable for automotive wide temperature application.

View attachment 191882

I figured it might be good to change the insulating paste in the PEM of my roadster. But I am located in Belgium, nowhere near a shop like GMC or Medlock.
Could you tell me what material I have to buy for the new insulating paste please?
I'll try the replacement by myself, with a friend that knows how to dismount everything.
Thank you
 
Can anyone give more details on the original material that Tesla Motors used for this insulation? Some people refer to it as "paste" but it looks to be much thicker than typical thermal paste - more like a sheet of rubber than thermal paste.

In addition, what sort(s) of replacement material(s) have been successful? ... and how long have the "new" materials been tested?