DIY Electric Car Forums banner
Cancel

Chevy volt BECM IMD (insulation monitoring device) and battery insulation informations

3082 Views 50 Replies 5 Participants Last post by  remy_martian, 
P
2
Hello,

I would like to know if someone already deep studied or reverse engineered the insulation monitoring device integrated on the BECM module of the chevy volt Gen1 ?
I know that it's not the most appropriated forum category but I know that's is probably where I can potentially find this kind of informations.
At the same time I'm also looking for someone that already measured insulation values of a complete chevy volt without and with official GM coolant inside with a megohmeter ?

I'm currently fighting with capricious Chevy volt on which I've tested insulation of all the HV parts independently, with very good values on all of them including the battery pack, but the car's IMD still measuring a low insulation value.

So I started to reverse engineer the IMD to find any problem on it and I'm looking for references values of battery/modules insulation.
For the moment I get >2000MOhms @500Vdc on the total battery pack without coolant and >30MOhms when it's full of official GM coolant. I can't imagine that this value is considered too low by the IMD...

So i'm interrested by any reliable informations on this topic.

Thanks.

Attachments

See less See more
Reply
Save
Like
1 - 18 of 51 Posts
P
This is the IMD (Insulation Monitoring Device) which is integrated on the BECM board in the Chevrolet Volt Gen1 (2011-2014).
I've started to draft the measuring circuit which seems to be pretty common. Measurement principe is well explain is this document for example: http://journal.it.cas.cz/62(2017)-1B/Paper 03 Han Peng.pdf
I'm not fully confident on some references for the moment but that's the basis.
I get 6.2Mohms of insulation @500V with a megohmeter between ground and HV minus and ground and HV positive input which is more or less the resistor chain value. Is there a more reliable way to check these chip capacitors ?
Thanks


Handwriting Rectangle Font Material property Parallel
See less See more
Reply
Save
Like
P
remy_martian said:
Again, is this in a car or on a bench?

If in a car, which one?
Click to expand...
In a car. As I wrote before in the Chevrolet Volt Gen1 2011.
This IMD is integrated in the battery pack of the car.
Reply
Save
Like
P
remy_martian said:
So, it's in a Volt, then?

Take it to the dealer.
Click to expand...
Ahah good luck to find any dealer capable to diagnose and repair the insulation monitoring device in a battery pack.
In the best case scenario it will propose to replace the whole battery pack. In the worst to sell you a brand new car...

Thanks for your nice recommandation but that's not the plan.
Reply
Save
Like
  • Like
Reactions: 1
P
remy_martian said:
This is clearly something a dealer won't touch, and you appear to me to be weaseling the location of the IMD vs saying whether this is your own contraption (I image processed your scrap piece of paper, lol):

And being your own contraption, the Volt pack is not happy with it, so you're looking for a defeat. If that's the case, you should spell out what you are doing so you can get ideas from people here of how to remedy the fault.
Click to expand...
Hi Remy, indeed this paper comes from the left drawer of my desk I used this draft paper to scribble the circuit diagram of the IMD circuit... This document was an RFQ from a potential client in 2019, as you saw for agriculture applications. It could have been my last mobile phone invoice or wathever piece of paper guy ! Nothing confidential and nothing related to the subject we discuss here. I've a company in France, we design battery pack for small series vehicle, motorcycles, ebike, ... we have our own technology based on cylindrical cells and honestly we have a pretty low interest to use or distort a >10 years old battery pack design for any other application.

You're welcome to imagine whatever weasel story if you want but I personally own a 2011 Chevrolet Volt (or more precisely a French "Opel Ampera") since 2016 on which I've random insulation problems for 2 years now. Because of lack of time I didn't had time before to investigate on it until now. As you can see below this is the car and the battery pack where the IMD is integrated :
Tire Car Land vehicle Vehicle Wheel


I only try to bring my expertise and my technical ressources with the help of the EV community to move forward on a technical issue that concerns more and more Chevy Volt today.
See less See more
Reply
Save
Like
P
2
kennybobby said:
Are you sure that the 6M resistor strings are connected to chassis ground? Or is that tied to some other ground plane.
Click to expand...
Yes as you can see in red, the center of the resistor chain is connected to the ground through a 0ohms resistor. So yes to my great surprise on this car you I always have half of the HV battery pack voltage on the car frame.
Passive circuit component Circuit component Hardware programmer Electronic component Microcontroller


I'm looking for full pinout diagram of the BECM but this voltage reference created by resistor chain is connected at least to the overall plan and to the both mx34 black and gray connectors (pin circle in red)
Passive circuit component Circuit component Hardware programmer Capacitor Computer hardware
See less See more
Reply
Save
Like
P
kennybobby said:
It's nice that there are reference designators for all the components so the board could be easily traced out.

Any manufacture logo on the wide leg spacing devices, is marking code W216SC04 ?

Maybe some sort of isolation or opto coupler device?
Click to expand...
I have difficulties to find the W216SC04 reference. There is a "M" logo on it which seems to be Matsushita/Panasonic logo so the most probable reference is a custom/prototype/old version of AQW216 photoMOS component. But I'm not 100% sure at the moment. These solid state relay are probably used to connect bias resistor when insulation monitoring is activated.
Reply
Save
Like
P
remy_martian said:
That is not chassis ground. That's the board ground. Chassis ground would be connected through the missing device above the one your highlighted red run is showing.
Click to expand...
You're right Remy, that's board ground. And the surrounded plan is the car ground and the two references are not supposed to be at the same potential. But when I connect HV on IMD input I get the voltage divider reference (half of the battery pack voltage) on both plan (board ground and surround chassis ground).
And that's propably the point @kennybobby I think that your remark about ceramic cap in the the corner was very pertinent ! I only have 50kohms insulation @500Vdc between board ground and chassis ground !
Reply
Save
Like
P
remy_martian said:
Those resistors are locally grounded on the board - "ground" is a common return for the circuits on that board.

The four caps KB circled are the only connection to chassis on that board - the schematic is wrong.
Click to expand...
Yes my draft schematic is wrong, I've been distrubed by these likely shorted or partially shorted capacitor apparently. I was very surprised that these voltage reference created by the resistor chain on the IMD was permanently connected to the car frame !

Yes it can probably works well without this cap, I will replace those properly and verify again the value measured by the car. Do you have any idea or at least an approximate value on this kind of ceramic cap ?

I will keep you informed soon. 🙏
Reply
Save
Like
P
kennybobby said:
If you are reading a voltage or 6Mohms to chassis, then inspect/replace any shorted little ceramic caps marked in the photo

View attachment 134919
Click to expand...
I removed the fourth capacitor you mentioned, by the way they are 11nF capacitor. But I still have 0.05Mohms between board gnd and chassis gnd and around +/-180V between ground and both HV polarity when HV is connected. So I have to find and other shorted part on the board...
Reply
Save
Like
P
kennybobby said:
What is on the bottom side of the board--there may be other components and circuits?
Click to expand...
Not that much as you can see :
Circuit component Passive circuit component Font Hardware programmer Electronic component

I will try to identify component connected to frame ground reference.
See less See more
Reply
Save
Like
P
R0042 is the 51kOhms resistor that connect board ground to car ground.
So apparently this mid-point HV reference is connected permanently to the car ground reference !?
Passive circuit component Circuit component Hardware programmer Electronic engineering Electronic component
See less See more
Reply
Save
Like
P
kennybobby said:
That explains your measurements; makes you wonder how they use that circuit to determine that insulation has degraded.

All the circuits on the HV side will need to be traced and drawn up to get some idea of what is the purpose.

Need to know the pinout of that orange connector too, looks like 4 inputs with 2 blank pins between each one..?
Click to expand...
Yes it makes sense now. So my first draft of the circuit diagram is not totally wrong except that there is this 51kohms resistor and some 11nF cap between PCB ground and car ground.
I know that on cars equiped with DC charging interface, the insulation monitoring device must be disconnected at least during charging session because insulation is tested and monitored by the charging infrastructure. So apparently on Volt Gen1, which was not equiped with CCS or Chademo, IMD is always connected...

Yes on the orange connector you get x4 inputs separated by two NC pins as follow :
Circuit component Green Passive circuit component Hardware programmer Electronic component

I didn't detailed the "after HV relays" inputs in my diagram because I supposed these are only used to check a welded contactor or a similar feature. This part of the circuit is not used for isolation measurement. The three right AQW216 (supposed) photomos are used for this part of the circuit. (@piotrsko what is the year of BECM you have ? because this part of the circuit whas apparently removed after 2013)
The three left AQW216 (supposed) photomos are used with the components described in my draft schematic and linked with HV inputs directly coming from battery pack (before HV relays)

I suspect this pcb to be at least a x4 layers board, so it makes reverse engineering a bit more tricky.
See less See more
Reply
Save
Like
P
The point is that actually I didn't found any measurable degradation of isolation.
I've checked all internal parts of the battery pack with a megohmeter @500V (I get over 2000Mohms on all the HV interfaces of the contactor board including battery heater, >25MOhms on the whole battery pack filled with official GM coolant inside). I also checked all the external parts (inverter, cabin heater, AC, harness, charger, ...) I measure isolation values between 1000 and 2000MOhms or above on these parts.
Then, even when the car ignition is OFF (all HV relays open in the battery pack) the IMD measure an insulation value between 500 and 800kohms which is a way too low. It's supposed to be >3000kOhms when ignition is off.
So that's why after several days of insulation measurement i'm almost convinced the problem comes from the IMD itself...
Reply
Save
Like
P
remy_martian said:
You cannot get megs of isolation with a 51k resistor connecting to a board worth of circuitry.

I'm not buying that resistor via goes to chassis ground - did you verify that via connection by eye or with a meter?
Click to expand...
I've measured this value with ohmmeter.
Reply
Save
Like
P
remy_martian said:
One thought - did you replace any coolant lines or do any work on it at all?
Click to expand...
No, I didn't replaced any coolant line. I only replaced the battery heater with a new one which was the first suspicious part when I dropped the battery pack.
Reply
Save
Like
P
remy_martian said:
Yeah - you better look. There could be something else making that connection.
Click to expand...
But that's coherent I have 0ohms value in relation to board gnd on one side of this resistor and 0ohms value on the other side of the resistor in relation to the frame gnd. And the value measured accross is coherent as it's indicate "513" on this resistor (51x10^3 ohms).
So apparently insulation measurement is based on this design on this car...
Reply
Save
Like
P
I found an interesting document here : https://www.seattleautomotive.com/files/SAP0201SM.pdf from page 82 where they describe two different isolation circuit (one passive and one active).
This document is not focused exclusively on chevy volt gen1. But what we spoke about here is the passive isolation circuit measurement.
I can't say for the moment is there are two different circuits (passive + active) on chevy volt gen1 and if my suspicious error of measurement comes from the passive one or active.
Reply
Save
Like
1 - 18 of 51 Posts
DIY Electric Car Forums
A forum community dedicated to DIY electric car owners and enthusiasts. Come join the discussion about electric vehicle conversions, builds, performance, modifications, classifieds, troubleshooting, maintenance, and more!
Full Forum Listing
Explore Our Forums
All EV Conversions and Builds Batteries and Charging Technical Discussion New Member Introductions eBay EV-Related Auctions
Top