EV Home RAV4 EV   Kubvan Electric
Kubvan Home
Created August 2017
Updated September 2018

Battery Charge failures as a result of a ground fault

Introduction:
It seems that charging failures mostly occur in rainy conditions or in California, after you wash your car.
It goes without saying that once I got my car to Vancouver Island it did not take long for me to get the dreaded "isolation error"
After a lot of trouble shooting and some reverse engineering I was led to the HWS circuit and a signal called LeakyV.
Now LeakyV which must mean leaky voltage is the output of a circuit which measures a voltage leak or potential between the chassis of the car and the battery pack.
This circuit is much more complex but not very different to the ground fault circuit in your bath room where you plug in your razor or hair dryer.
Water being the common factor here too. So the idea is that if there is a voltage potential to the chassis of your car and ground or your hair dryer and ground you want to trigger an alarm or shut the power off. In the case of your car it shuts down the charging system which is an interesting compromise that allows you at least to perhaps limp home.

Work around #1:
After a bit more trouble shooting I was able to determine that if you unplugged the high voltage connector from the HWS/LeakyV circuit the car computer would not detect the error and would charge normally. Good work around but not safe I said.
PLEASE TAKE NOTE. NOT SAFE.

Discovery #1 :
When I pulled the batteries from my car the first time as an exploration to figure out what was going in the battery pack, we learned  a lot of things but made one significant discovery, thanks Bill. I had built a fancy high voltage impedance tester as part of my initial trouble shooting and discovered then that the impedance of the pack high voltage to the chassis was around 60k ohms. A minimum industry standard from what I could determine was 120K ohms. No wonder it was failing. When we started to disconnect the pack it was not long before we had an impedance of over 900 meg ohms. Yea problem solved, well not really. Thanks to Bills tenacity he poked around with a ordinary ohm meter and found that where the paint was chipped inside the case mostly underneath the batteries you could measure a short to ground. Hint #1. Further with a battery placed in a slot with chipped paint and using the high voltage impedance meter we also found a low impedance from the battery post to chassis ground. Ah ha we said. using blue weather shield tape we were able to eliminate the short. So what did we do .... put the car back together of course. Bill was not convinced. The charging worked for a little while but soon I was back to unplugging the HWS/LeakyV circuit to charge again. Bill was right.

Discovery #2 :
It was not long before through some good fortune and the help of Darren that we sourced some used batteries and a plan was set to replace a bunch in my pack. This time armed with 2 part epoxy garage floor paint and another friend I dropped the pack and completely cleaned the case inside and out in preparation for painting. This was the next ah ha moment.
chipped chipped
 
Lots of chipped paint and paths to ground which led to the discovery that the entire case is coated inside and out with a conductive copper coating essentially a giant ground.
The inside coating is isolated from the batteries by a barrier coat, with lots of chips and scuffs from vibrating after 70,000 miles and careless removal and installation of batteries by me and the previous owners friend who had tried to fix it once before. Now the big questions are where is the path to chassis ground from the battery case and why are all the bolts in the case completely isolated from the battery case so that there is no possible path to the chassis. See pictures 3 & 4 attached. Picture 3 shows the rubber isolation mounts of the main bolts I thought they were for vibration damping. But if you look at picture 4 you will see that each of the bolt mounts have a black ring around them the conductive copper coating is purposely isolated. There is no connection to ground.

three four

 
Discovery # 3 :

There is a bonding strap that goes from the fire wall to the battery case where the high voltage cables go to the battery. This strap serves no obvious purpose except to trigger the ground fault circuit after 10's of thousands of miles.

bonding strap bonding starp


Was it designed specifically to do that or is that an unintended consequence of wanting to create an RF shield around the batteries and the battery ECU or create a path to ground in the case of an accident or serious mechanical failure. Somebody at Toyota may know, but I don't think they are letting on. If anyone else has any suggestions happy to hear. In the mean time this may be a work around but I am not recommending you use this information for anything other than quiet contemplation.

Final footnote:

The conductive path from the battery posts to the chassis ground is not completly as a result of moisture after all fresh water is mostly non-conductive. Whats really happening is that a soup is being created n the battery tray after many years of charging and driving. The soup is a mixture of moisture, off gassed electrolyte, road dirt and dust. This creates a lovely film over all the batteries and in the bottom of the tray which is conductive and creates a conductive path across the tops of the batteries, down the sides of the batteries and the restraining brackets to the tray. Any abrasian, cracks or chips then provide a direct path to the chassis ground through the bonding strap. Except for this bonding strap the battery case is purposefully completly isolated from the chassis and by the barrier inside the tray from the batteries.

Cleaning the batteries is your next step


Pictures 7 & 8 shows my nice epoxy coated tray that may be 100% isolated now but runs without a bonding strap.
battery case bottom battery case top