-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
DanielFriederich
Member
Nope.. it has nothing to do with the charging, maybe higher temperatures in the PEM might be a factor eventhough shortcuts have been reported at low temps...
A lower charging rate is still better for the battery
A lower charging rate is still better for the battery
Kerios
Member
I suspect a variety of factors including moisture ingress - not a definite factor but a possible one was that I had washed my car the day before my failure - the PEM sealing doesn't look super tight. Another future action may be to ensure that all rubber or equivalent seals meet IP68 standards.
Elevated temperatures, moisture plus possibly poor material section as the original gasket all probably lead to premature or even anticipated failure.
Changing the charging profile won't have an impact - the PEM failure is a matter of time I believe for all cars. The actual mode of failure may vary - I don't think there is a current list of those modes ?
Elevated temperatures, moisture plus possibly poor material section as the original gasket all probably lead to premature or even anticipated failure.
Changing the charging profile won't have an impact - the PEM failure is a matter of time I believe for all cars. The actual mode of failure may vary - I don't think there is a current list of those modes ?
Are the materials the same in the 1.5 PEM and the 2.x PEM? That is, would the probability of failure be similar?
Kerios
Member
Hard to know, but raises the question about whether Tesla or a third party is rebuilding the PEM and if so what materials are being used - its possible that the original Bill of Materials, is being regenerated i.e. rebuild it as it was originally or possible rebuild it with similar specification material or worse case just rebuild it!
I am trying to understand from my SC, the failure mode - at least as a pointer to determine mode and means. Once I have my Roadster back... I intend to take a look at the PEM boards etc, and additionally subject to warranty, install an environmental monitoring package i.e. temperature, humidity etc and start collecting the respective metrics.
I am trying to understand from my SC, the failure mode - at least as a pointer to determine mode and means. Once I have my Roadster back
... I intend to take a look at the PEM boards etc, and additionally subject to warranty, install an environmental monitoring package i.e. temperature, humidity etc and start collecting the respective metrics.
DanielFriederich
Member
There's a humidity sensor which stops the fans before reaching the dew point and every cooling plate contains a heat resistor.
When I opended up the PEM (and there was a lof debris inside the cooling fins on the outside!) it looked like new (no dust, debris, signs of moisture whatsoever).
To determine if the insulation has been changed needs a bit of work since the driver boards are mounted above the performance boards and the IGBTs are mounted on the backside (downside) of the circuit boards and they are mounted to the cooling plates. Maybe its possible to look it up with a little spy cam without dissasembling the circuit boards but won't be easy and still you won't see if there has been already a isolation fault causing a dsiruptive discharge from the backplane of a transistor to the cooling plate with metall to be evaporated so it is no constant shortage or a defect IGBT... this can only be detected by completely dissasembling the power boards unmounting the IGBTs from the cooling plates. And if done like this insulation and mounting insulation of the IGBT have to be replaced anyway...
Since it has been reported that there are no new PEMs available anymore Tesla has to refurbish used ones hence changing the insulation too when an IGBT on the circuit board has to be replaced.
Also IGBTs themselves have developed a lot (e.g. saturation voltage has gone done which means more efficiency and less heat development).
As far as I know by now the 1.5 PEM had different IGBTs with a much higher ampere and driver frequency (seems like Tesla wanted to built it bullet proof and later on with the 2.0 and 2.5 looked a bit more into saving some money since the 1.5er IGBTs are a lot more expensive than the ones used later).
GS posted some pictures in the german forum showing the old insulation and some disruptive discharges causing damage to the cooling plates....
When I opended up the PEM (and there was a lof debris inside the cooling fins on the outside!) it looked like new (no dust, debris, signs of moisture whatsoever).
To determine if the insulation has been changed needs a bit of work since the driver boards are mounted above the performance boards and the IGBTs are mounted on the backside (downside) of the circuit boards and they are mounted to the cooling plates. Maybe its possible to look it up with a little spy cam without dissasembling the circuit boards but won't be easy and still you won't see if there has been already a isolation fault causing a dsiruptive discharge from the backplane of a transistor to the cooling plate with metall to be evaporated so it is no constant shortage or a defect IGBT... this can only be detected by completely dissasembling the power boards unmounting the IGBTs from the cooling plates. And if done like this insulation and mounting insulation of the IGBT have to be replaced anyway...
Since it has been reported that there are no new PEMs available anymore Tesla has to refurbish used ones hence changing the insulation too when an IGBT on the circuit board has to be replaced.
Also IGBTs themselves have developed a lot (e.g. saturation voltage has gone done which means more efficiency and less heat development).
As far as I know by now the 1.5 PEM had different IGBTs with a much higher ampere and driver frequency (seems like Tesla wanted to built it bullet proof and later on with the 2.0 and 2.5 looked a bit more into saving some money since the 1.5er IGBTs are a lot more expensive than the ones used later).
GS posted some pictures in the german forum showing the old insulation and some disruptive discharges causing damage to the cooling plates....
Kerios
Member
To conclude on my Roadster, after 3 weeks with the Gatwick SC, it had another PEM flown over from Fremont and replaced due to Megapole failure in the first (I'm not entirely sure what that means). Additionally timing would have it, that the battery was also upgraded to the 3.0 version and as far as I can tell the car is driving really well now. The Gatwick SC did a really good job of getting this resolved quickly and its great to have a good level of service like that.
DanielFriederich
Member
The IGBT modules are called "megapole"
This means that you most likely had a defective transistor due to a short cut to ground (insulation fault)
there are 3 megapole cards in the PEM (one for each phase of the current for the motor) with 28 IGBTs on it
This means that you most likely had a defective transistor due to a short cut to ground (insulation fault)
there are 3 megapole cards in the PEM (one for each phase of the current for the motor) with 28 IGBTs on it
The alarm came up "DMC HW: PhaseA Low Side Desat Fault." out of debug mode it comes up as "Power Train Problem Restart When Safe". Car would not move - tried rebooting everything I could think of - 6 open and closes of the charge port, everything seems to boot up ok but the car just will not move - as soon as I touch the accelerator I get the error messages above...
At least got the car back to my garage. Sounds ominously like PEM problem but does anybody have anything more specific?
Thanks,
Jonathan
Last edited:
DaveD
EVs Kick Gas!
ID 906 - It's on this list: Open-Vehicle-Monitoring-System/ovms_server_tr.vece at master · openvehicles/Open-Vehicle-Monitoring-System · GitHub
Desat fault == Desaturation fault?
Desaturation Fault Detection | EEWeb Community
Desat fault == Desaturation fault?
Desaturation Fault Detection | EEWeb Community
dhrivnak
Active Member
If anyone is interested here is list of the errors from the log files from last night as the fun and games happened :-( ...
05/26/2017 19:03:17 | 1495818197 | ERR | error code 905 6 bytes DMC HW: PhaseA Low Side Desat Fault
05/26/2017 19:03:27 | 1495818207 | ERR | error code 1141 6 bytes (Can't find this error anywhere....)
05/26/2017 19:03:27 | 1495818207 | ERR | error code 1055 6 bytes (As above)
05/26/2017 19:03:27 | 1495818207 | ERR | error code 1056 6 bytes (As above)
05/26/2017 19:03:27 | 1495818207 | ERR | error code 1057 6 bytes (As above)
05/26/2017 19:03:41 | 1495818221 | ERR | error code 905 6 bytes DMC HW: PhaseA Low Side Desat Fault
and the same errors then repeated themselves in the same order every time I tried a reset or all the things I tried to get going...
05/26/2017 19:03:17 | 1495818197 | ERR | error code 905 6 bytes DMC HW: PhaseA Low Side Desat Fault
05/26/2017 19:03:27 | 1495818207 | ERR | error code 1141 6 bytes (Can't find this error anywhere....)
05/26/2017 19:03:27 | 1495818207 | ERR | error code 1055 6 bytes (As above)
05/26/2017 19:03:27 | 1495818207 | ERR | error code 1056 6 bytes (As above)
05/26/2017 19:03:27 | 1495818207 | ERR | error code 1057 6 bytes (As above)
05/26/2017 19:03:41 | 1495818221 | ERR | error code 905 6 bytes DMC HW: PhaseA Low Side Desat Fault
and the same errors then repeated themselves in the same order every time I tried a reset or all the things I tried to get going...
BTW I tried to charge the car a little earlier - it went through blue plug to yellow and then started making a hell of racket....maybe the contactors were doing some strange things. Anyway it is stopped and the APS is now disabled. Tomorrow I will pull the "Service Disconnect". I really don't know how long the car will be off the road for....my guess is quite some months...
Kerios
Member
Almost identical symptoms to my Roadster failure start of May, see PEM failure thread for more details. Turned out to be Megapole failure which resulted in a new (remanufactured) PEM being ordered from Fremont and car off road for almost a month.
Suspicion increasing is the insulation material breaking down when under the IGBTs, repairable for sure but question whether you get Tesla to do it or a third party.
Suspicion increasing is the insulation material breaking down when under the IGBTs, repairable for sure but question whether you get Tesla to do it or a third party.
DanielFriederich
Member
sometimes the error messages might lead you in a wrong direction... for example the BSM Isolation Fault can be a sign of ongoing bad isolation of an IGBT as also some moisture (for example after car wash) in the DC/DC converter under the front hood...
#905 is DMC HW: PhaseA Low Side Desat Fault meaning (as far as I know by now) that on Megapole Phase 1 Input Voltage to power ampflification has gone wrong
#906 is DMC HW: PhaseA High Side Desat Fault meaning behind IGBT (transistors) the power is lost
#909 #910 for second phase Megapole
#913 #914 for third phase
btw: it started with BSM Isolation fault and months later with #905 etc. too
#1055 #1056 are not listed but should correspond with the 905ers
As soon as my PEM has been repaired (the electronic guy promised to start soldering this week) I will make a repair thread with detailed information about the electronic parts replaced and the work been done
901 DMC HW: PhaseA OverCurrent Fault
902 DMC HW: PhaseB OverCurrent Fault
903 DMC HW: PhaseC OverCurrent Fault
904 DMC HW: Battery OverVoltage Fault
905 DMC HW: PhaseA Low Side Desat Fault
906 DMC HW: PhaseA High Side Desat Fault
907 DMC HW: PhaseA Bias Under Voltage Fault
908 DMC HW: PhaseA Bus Over voltage fault
909 DMC HW: PhaseB Low Side Desat Fault
910 DMC HW: PhaseB High Side Desat Fault
911 DMC HW: PhaseB Biad UnderVoltage fault
912 DMC HW: PhaseB Bus OverVoltage fault
913 DMC HW: PhaseC Low Side Desat fault
914 DMC HW: PhaseC High Side Desat fault
915 DMC HW: PhaseC Bias UnderVoltage fault
916 DMC HW: PhaseC Bus OverVoltage fault
#905 is DMC HW: PhaseA Low Side Desat Fault meaning (as far as I know by now) that on Megapole Phase 1 Input Voltage to power ampflification has gone wrong
#906 is DMC HW: PhaseA High Side Desat Fault meaning behind IGBT (transistors) the power is lost
#909 #910 for second phase Megapole
#913 #914 for third phase
btw: it started with BSM Isolation fault and months later with #905 etc. too
#1055 #1056 are not listed but should correspond with the 905ers
As soon as my PEM has been repaired (the electronic guy promised to start soldering this week) I will make a repair thread with detailed information about the electronic parts replaced and the work been done
901 DMC HW: PhaseA OverCurrent Fault
902 DMC HW: PhaseB OverCurrent Fault
903 DMC HW: PhaseC OverCurrent Fault
904 DMC HW: Battery OverVoltage Fault
905 DMC HW: PhaseA Low Side Desat Fault
906 DMC HW: PhaseA High Side Desat Fault
907 DMC HW: PhaseA Bias Under Voltage Fault
908 DMC HW: PhaseA Bus Over voltage fault
909 DMC HW: PhaseB Low Side Desat Fault
910 DMC HW: PhaseB High Side Desat Fault
911 DMC HW: PhaseB Biad UnderVoltage fault
912 DMC HW: PhaseB Bus OverVoltage fault
913 DMC HW: PhaseC Low Side Desat fault
914 DMC HW: PhaseC High Side Desat fault
915 DMC HW: PhaseC Bias UnderVoltage fault
916 DMC HW: PhaseC Bus OverVoltage fault
DanielFriederich
Member
Thanks Dave!
This corresponds exactly with a or more than one faulty IGBTs since they have to be "saturated" with a certain voltage for full operation...
We're narrowing it down more and more to faulty IGBGTs due to bad gone isolation.
I wonder how many of us have already been running into this problem and if we are many (and that's what I guess meanwhile) how Tesla would react to it!
What Daniel is doing here for us all is going to be a game changer...it means that we will be able to repair our PEM's for a much more modest price than if we went the Tesla route (or any other route for that matter...) - I did talk to them (Tesla France) by email - they just said my car was out of guarantee - yes technically correct but not really very catholic.
According to people much smarter than me it appears to be a breakdown in the insulation - my understanding is that they went a bit cheaper with the electrical components for the 2.0 & 2.5 cars...so choices made some years ago are biting us in the ass now...it would be very honorable for Tesla to at least come to the party in some capacity for our repair costs - but in any event I would much rather repair my PEM with electrical components of a higher standard than what we previously had. There is no reason to disbelieve that our PEM's will not be better than standard after this repair. Perhaps I am jumping the gun but I think this is going to be the case.
I agreed with Tesla (France) that I would inform them of how things worked out.
I'm standing by and hoping that everything will work well for Daniel (hopefully end of next week) then I intend to go down the same route.
A huge thank you Daniel!
According to people much smarter than me it appears to be a breakdown in the insulation - my understanding is that they went a bit cheaper with the electrical components for the 2.0 & 2.5 cars...so choices made some years ago are biting us in the ass now...it would be very honorable for Tesla to at least come to the party in some capacity for our repair costs - but in any event I would much rather repair my PEM with electrical components of a higher standard than what we previously had. There is no reason to disbelieve that our PEM's will not be better than standard after this repair. Perhaps I am jumping the gun but I think this is going to be the case.
I agreed with Tesla (France) that I would inform them of how things worked out.
I'm standing by and hoping that everything will work well for Daniel (hopefully end of next week) then I intend to go down the same route.
A huge thank you Daniel!
Similar threads
