-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
tomsax
Member
Hi Steve,
In your log file from August, all the VINF records in the permanent section showing firmware version 3.7.1 have invalid time stamps (in 2033), so they get ignored in the summary produced by the -b flag.
How often are permanent VINF records written? Now I do have two with valid timestamps on 11/16/2016, perhaps as a consequence of actions taken during the annual service work that day.
No, I believe the issue is independent of the CAN Adapter, though I haven't actually tried the Tesla-original J1772 adapter, now that I think of it. I have one now, and if it matters, I can give it a try.
The "bug" (arguably) is in the Clipper Creek equipment, in how they sequence the GFI test after the charging is done, cycling the Pilot signal. That tricks the car to think it just got plugged in, starting the charge again. Based on the testing we did at their office, Clipper Creek is aware of what is going on, but I haven't heard of any firmware updates that resulted from it. By my archive, my last interaction with them was in November 2016, with their statement that they'd let me know what they would do. I'm guessing that the lack of feedback from EV owners has led to this just being dropped.
If you are experiencing this with a home charger, you might work around the problem by configuring the car to charge-at-time (vs the default charge-on-plugin). Or, use a different charger. I'm using an OpenEVSE, and it works perfectly.
The "bug" (arguably) is in the Clipper Creek equipment, in how they sequence the GFI test after the charging is done, cycling the Pilot signal. That tricks the car to think it just got plugged in, starting the charge again. Based on the testing we did at their office, Clipper Creek is aware of what is going on, but I haven't heard of any firmware updates that resulted from it. By my archive, my last interaction with them was in November 2016, with their statement that they'd let me know what they would do. I'm guessing that the lack of feedback from EV owners has led to this just being dropped.
If you are experiencing this with a home charger, you might work around the problem by configuring the car to charge-at-time (vs the default charge-on-plugin). Or, use a different charger. I'm using an OpenEVSE, and it works perfectly.
NigelM
Recovering Member
I am using the original Tesla J1772 adapter although I'd never noticed the issue until now.
spaceballs
Member
hjr
#1291
A few of us did extensive research on this awhile back and one owner visited Clipper Creek. The owner of Clipper. Reek wa snot aware of the issue in his own Roadster until it was shown to him! The Roadster was developed and built before the J1772 protocol was written. The handshake for the car to tell the charger “I Am done charging” on the car is a little off from the signal most J1772 chargers are expecting. Many J1772 chargers think this means “I am thirsty start a charge” . The Clipper Creek thus starts a charge, the car says its done, and the charger starts over. I think the only issue is the wear and tear on your contractors. There is no solution to this as Tesla will not ever do the simple software upgrade to fix it. We are forgotten. What I do is set the charge voltage low so it does not ever finish a charge before I get back to the car!
This post is pretty much correct although I don't recommend the technique in the last sentence because it prevents the battery from balancing. It also uses more energy to charge at rates below 30A.
It's hard to say who is responsible for the bug that causes this cycling behavior using Clipper Creek chargers. On the one hand the Roadster is not completely compliant with SAE J1772 and the problem would go away if it was. On the other hand Clipper Creek uses an unusual protocol to test for a stuck contactor, although it is technically J1772 compliant. No chargers made by other manufacturers have this issue. So who's fault is it? Either company (Clipper Creek or Tesla) could make a minor software change to eliminate the issue. However, I don't blame either company for not making the update. If CC were to make the fix they would then have to re-test their chargers for UL. In addition, they would have to perform extensive testing and validation with every EV that they sell chargers for. If Tesla made the change then they would also have to perform extensive validation and frankly this is a minor problem compared to other firmware priorities that we could all suggest.
hjr
#1291
Henry is right about the car not balancing if you basically stop the charge and drive off, although the car will balance on its own even unplugged. I meant to point out that to me, it made more sense not to risk damage to the contactors in the relays from excessive cycling for no reason.
I never tried to stop a charge using OVMS when it got near the top, to prevent the cycling. I am not sure it would work as the CC charger would still misinterpret the pilot signal from the Roadster.
Similar threads
