[DATA] - HVBJB Mega Data Aggregation Thread

[DevSecOps]

Preface:

I decided I would put together some of the data from HVBJB failure on our vehicles. The below data represents only a fraction of the vehicles that have had issues.

Key Points:

• We probably assisted with an additional 10-15 that are not part of this data set for various reasons.
• Red indicators are for HVBJB complete failure (SSN) with the recall applied
• This data only represents forum members who had failure and does not include people who never contacted me in various forum posts or the general public who isn't on the forum

Key Notable Data Points:

• Speed seems to be a good indicator of failure
• ECU Uptime is very, very strange. That has to be a bug or we can create a bunch of theories which might be more fun.
• Amps @ failure were always above ~400A
• No failures during regeneration
• Data is missing from SSN due to the DTC triggering on power on events
• Other data is missing from some fields because the car hasn't been scanned yet
• There's a healthy mix of MY20, 21 & 22 vehicles
• Engine Types
  • GTPE (Motor #7)
  • GT (Motor #6)
  • Non-GT AWD ER (Motor #5)
  • Non-GT RWD ER (Motor #4)
  • Non-GT AWD SR (Motor #3)
  • Non-GT RWD SR (Motor #2)

July:

August:

September (as of 09/22/22):

 

[SnBGC]

Preface:

I decided I would put together some of the data from this months SVS errors on our vehicles. The below data is just for this month and it represents a fraction of the vehicles that we assisted with.

Key Points:

• We probably assisted with an additional 10-15 that are not part of this data set for various reasons.
• This is only for post-recall failures (SVS)
• We assisted with a number of SSN failures, but those aren't included in this data set
• This data only represents forum members who had SVS in August 2022 and does not include people who never contacted me in various forum posts or the general public who isn't on the forum

Key Notable Data Points:

• Everyone was traveling more than 50mph when failure happened
• There are no RWD vehicles in the data set
• ECU Uptime is very very strange. That has to be a bug.
• Amps @ failure were always above 400A
• No failures during regeneration
• There's a healthy mix of MY20, 21 & 22 vehicles
• There's about an equal amount of GT (7) to Non-GT (8) failures

View attachment 77901

If anyone wants the CSV data:

VIN Last 4,Motor #,Build Date,Failure #,Mileage,DTC,MPH,SOC,HVB Temp (F),Coolant Temp (F),LVB V,HVB V,HVB Age (Months),Amps @ Failure,Leakage Resistance (Ohms),ECU Uptime (Minutes)
,,,,,,,,,,,,,,,
0526,7,07/19/21,1,"23,185",P0ADE,81,59.03,95,95,12.8,336,12.68,416.7,1239575,8.5332
2033,6,08/30/21,1,"15,883",P0ADA,102,62.3,95,93,14,340,11.43,392.1,1271800,8.5332
6679,5,12/08/21,2,"11,876",P0ADA,58,88.18,75,79,12.75,358,8.25,671.7,1154650,8.5332
9256,6,03/07/22,1,"5,063",P0ADE,53,48.22,89,102,14.26,327,5.65,588.1,1259750,8.5332
0243,6,02/28/22,1,"8,097",P0ADE,89,79.58,95,91,12.75,353,5.55,491.3,1210350,8.5332
2858,5,07/11/21,1,"16,536",P0ADE,72,52.63,80,80,12.73,326,11.14,645.3,1138550,8.5332
3150,6,09/14/21,1,"20,330",P0ADE,88,79.43,98,104,12.75,352,10.81,680,1246725,8.5332
0846,5,01/11/21,1,"11,432",P0ADA,53,49.92,80,80,14.5,325,18.4,608,1114200,8.5332
2649,5,11/09/20,1,"29,809",P0ADE,59,93.12,80,79,12.75,363,20.9,669.6,1176375,8.5332
7548,5,12/15/20,1,"34,213",P0ADE,69,75.9,86,93,12.75,345,16.23,733.2,1112025,8.5332
8125,4,12/23/20,1,"21,473",P0ADE,62,72.18,102,100,12.75,347,19.61,549.3,1638375,8.5332
2034,7,09/21/21,1,"7,447",P0ADE,98,82.22,75,84,12.84,355,10.58,532.8,1269325,8.5332
9660,6,07/20/21,1,"25,085",P0ADE,72,64.23,90,86,12.75,341,12.32,413,1253425,8.5332
6620,5,01/05/21,1,"10,303",P0ADE,82,54.38,68,75,12.99,324,19.06,623,1140625,8.5332

Under what conditions are you seeing 500A of regen? Max I have recorded so far is just over 300A. ???

 

[scoopman]

Looks like 8.5332 minutes is the most unlucky time!

 

[Mach-Lee]

Makes sense, the voltage drop across the contacts the BECM is looking for is proportional to the current, so it would take a high load event before it will notice a high-resistance contactor. I'm guessing there is just a specific voltage drop it's looking for as a threshold (something like 1.0V?). Not sure if you have contactor voltages you could compare to pack voltages to determine the voltage drop.

I wouldn't worry too much about the ECU uptime, 8.5332 min is exactly 512 seconds. That might be some kind of default time or placeholder number since it's a power of 2. I think most people were driving the car longer than 8 minutes when it happened.

 

[DevSecOps]

Not sure if you have contactor voltages you could compare to pack voltages to determine the voltage drop.

I wouldn't worry too much about the ECU uptime, 8.5332 min is exactly 512 seconds. That might be some kind of default time or placeholder number since it's a power of 2. I think most people were driving the car longer than 8 minutes when it happened.

I don't have contactor voltages unfortunately. They aren't logged.

The ECU Uptime is strange because it isn't that exact same time for other DTCs. It's random. For some reason on this specific DTC it's always 8.5332 minutes. If the DTC is a false positive then it could be that they are all happening at the same time. I'm not saying that's the case, but just as an example. I really don't know why it's like that. For me and my fault, it was about 8 minutes after I left Starbucks. I don't have anyone else chiming in on how long they were driving before the alert.

 

[benk016]

I actually just went back and looked at a text I sent as I started the car, and when I got the error. Mine was actually around 16 minutes into my drive when it happened.

 

[DevSecOps]

I actually just went back and looked at a text I sent as I started the car, and when I got the error. Mine was actually around 16 minutes into my drive when it happened.

As I said in my OP, I figure it's a bug, but it's just so strange. It's a timer and it just stops at the same time every time for that DTC.

 

[benk016]

As I said in my OP, I figure it's a bug, but it's just so strange. It's a timer and it just stops at the same time every time for that DTC.

Like @Mach-Lee alluded to, It could be the software takes 512 second samples and at the end of a sample it decides to throw the error or not. Kind of jives with mine erroring at around 16 minutes.

Or its just a bogus reading all together.


Or.....

 

[portlandg]

Preface:

• There are no RWD vehicles in the data set

This just confirms my theory that RWD cars are less likely to have an issue as there is less strain/demand due to only having one motor.

 

[Regulus7]

Preface:

I decided I would put together some of the data from this months SVS errors on our vehicles. The below data is just for this month and it represents a fraction of the vehicles that we assisted with.

Key Points:

• We probably assisted with an additional 10-15 that are not part of this data set for various reasons.
• This is only for post-recall failures (SVS)
• We assisted with a number of SSN failures, but those aren't included in this data set
• This data only represents forum members who had SVS in August 2022 and does not include people who never contacted me in various forum posts or the general public who isn't on the forum

Key Notable Data Points:

• Everyone was traveling more than 50mph when failure happened
• There are no RWD vehicles in the data set
• ECU Uptime is very, very strange. That has to be a bug or we can create a bunch of theories which might be more fun.
• Amps @ failure were always above ~400A
• No failures during regeneration
• There's a healthy mix of MY20, 21 & 22 vehicles
• There's about an equal amount of GT (7) to Non-GT (8) failures

View attachment 77901

If anyone wants the CSV data:

VIN Last 4,Motor #,Build Date,Failure #,Mileage,DTC,MPH,SOC,HVB Temp (F),Coolant Temp (F),LVB V,HVB V,HVB Age (Months),Amps @ Failure,Leakage Resistance (Ohms),ECU Uptime (Minutes)
,,,,,,,,,,,,,,,
0526,7,07/19/21,1,"23,185",P0ADE,81,59.03,95,95,12.8,336,12.68,416.7,1239575,8.5332
2033,6,08/30/21,1,"15,883",P0ADA,102,62.3,95,93,14,340,11.43,392.1,1271800,8.5332
6679,5,12/08/21,2,"11,876",P0ADA,58,88.18,75,79,12.75,358,8.25,671.7,1154650,8.5332
9256,6,03/07/22,1,"5,063",P0ADE,53,48.22,89,102,14.26,327,5.65,588.1,1259750,8.5332
0243,6,02/28/22,1,"8,097",P0ADE,89,79.58,95,91,12.75,353,5.55,491.3,1210350,8.5332
2858,5,07/11/21,1,"16,536",P0ADE,72,52.63,80,80,12.73,326,11.14,645.3,1138550,8.5332
3150,6,09/14/21,1,"20,330",P0ADE,88,79.43,98,104,12.75,352,10.81,680,1246725,8.5332
0846,5,01/11/21,1,"11,432",P0ADA,53,49.92,80,80,14.5,325,18.4,608,1114200,8.5332
2649,5,11/09/20,1,"29,809",P0ADE,59,93.12,80,79,12.75,363,20.9,669.6,1176375,8.5332
7548,5,12/15/20,1,"34,213",P0ADE,69,75.9,86,93,12.75,345,16.23,733.2,1112025,8.5332
8125,4,12/23/20,1,"21,473",P0ADE,62,72.18,102,100,12.75,347,19.61,549.3,1638375,8.5332
2034,7,09/21/21,1,"7,447",P0ADE,98,82.22,75,84,12.84,355,10.58,532.8,1269325,8.5332
9660,6,07/20/21,1,"25,085",P0ADE,72,64.23,90,86,12.75,341,12.32,413,1253425,8.5332
6620,5,01/05/21,1,"10,303",P0ADE,82,54.38,68,75,12.99,324,19.06,623,1140625,8.5332

I am assuming your data dump comes from looking up the codes when the failures are reported using FDRS? Just wondering if we were to use car scanner and keep an OBDE dongle record, what would you be monitoring as a prelude to failure? max amps?

 

[BillPitman]

Makes sense, the voltage drop across the contacts the BECM is looking for is proportional to the current, so it would take a high load event before it will notice a high-resistance contactor. I'm guessing there is just a specific voltage drop it's looking for as a threshold (something like 1.0V?). Not sure if you have contactor voltages you could compare to pack voltages to determine the voltage drop.

I wouldn't worry too much about the ECU uptime, 8.5332 min is exactly 512 seconds. That might be some kind of default time or placeholder number since it's a power of 2. I think most people were driving the car longer than 8 minutes when it happened.

I don’t know if this matters but,- I monitor the HvCVP parameter & the EV Battery system voltage parameter…. Typically the contactor voltage is .5-1 volt higher than the system voltage. I figure if the contactor voltage ever goes lower than the system voltage, this would mean a higher resistance and therefore a larger voltage drop?! Does this make any sense, or am I all wet on this? Bill P

 

[DevSecOps]

This just confirms my theory that RWD cars are less likely to have an issue as there is less strain/demand due to only having one motor.

That's never been a theory. It's always been a known fact. We've said over and over that more power draw has a higher likelihood of failure. I did however revise my OP because there was 1x RWD vehicle in the data set.

what would you be monitoring as a prelude to failure? max amps?

Typically the contactor voltage is .5-1 volt higher than the system voltage. I figure if the contactor voltage ever goes lower than the system voltage, this would mean a higher resistance and therefore a larger voltage drop?! Does this make any sense, or am I all wet on this?

I was monitoring all parameters when I had failure and so far I haven't been able to figure out how the car determined the failure. When I eventually had SSN it was obvious, looking at the data, why I had SSN, but for the initial SVS, I still don't know what the car saw that I couldn't see. So, I'm not sure you can monitor it and determine failure ahead of SVS.

 

[hawkeye3point1]

Can anyone provide a technical description of Leakage Resistance and what the BECM does with it, TIA.

 

[DevSecOps]

Can anyone provide a technical description of Leakage Resistance and what the BECM does with it, TIA.

If you want technical, here you go:

https://www.ti.com/lit/ug/tidudj6a/tidudj6a.pdf

Scroll down to page 8

 

[Mach-Lee]

I don’t know if this matters but,- I monitor the HvCVP parameter & the EV Battery system voltage parameter…. Typically the contactor voltage is .5-1 volt higher than the system voltage. I figure if the contactor voltage ever goes lower than the system voltage, this would mean a higher resistance and therefore a larger voltage drop?! Does this make any sense, or am I all wet on this? Bill P

Voltage drop is proportional to resistance, so a higher voltage drop is bad. However the positive contactor voltage should always be less than the pack voltage, and the negative contactor voltage should rise above 0V. I’ll have to look at that on my car. Might be rounding errors or something. A good contactor in spec should always have a voltage drop less than 0.5V, 1.0V would be double spec and where I would get concerned.

Can anyone provide a technical description of Leakage Resistance and what the BECM does with it, TIA.

The entire battery circuit is isolated from the chassis. The leakage resistance is the resistance between the battery circuit and the chassis. Normally it should be very high because nothing should be touching the chassis or it’s insulated. If there is a problem in the system somewhere such as a HV cable rubbing through, cell tabs touching ground, or moisture in the pack the resistance will go down, and the BECM will shut down the pack.

Todd had an isolation fault at the end and I theorize it may be because a contactor coil touched the contact plunger while energized. Coil insulation melting through could cause that.

Sponsored