Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Fix for Roadster 80A Charging Bug

This site may earn commission on affiliate links.
It's the same bug that was just fixed for the Roadster. If the station reports that it can supply 80A, the car refuses to charge. If the station reports 70A then it charges. This is particularly silly for the Leaf as it won't ever draw anywhere near that much current.
 
Sorry if this is OT, but could you explain what software bug prevents the Leafs from using J1772? Do you mean it does not charge on EVSEs above 24/30A?
Just like the Roadster before this fix, a LEAF can't cope with a J1772 station that sends out an 80A pilot signal, even though an 80A pilot signal is totally within the J1772 spec. We can perhaps almost excuse the Roadster because it shipped long before the J1772 spec was finalized, but the LEAF has no excuse. (I'm a Roadster and LEAF owner.)
 
Just like the Roadster before this fix, a LEAF can't cope with a J1772 station that sends out an 80A pilot signal, even though an 80A pilot signal is totally within the J1772 spec. We can perhaps almost excuse the Roadster because it shipped long before the J1772 spec was finalized, but the LEAF has no excuse. (I'm a Roadster and LEAF owner.)

I am a LEAF and Roadster owner too but my Leaf can't charge at anything over 16A anyway. Are you saying the leaf won't charge at any level if the station sends a max current pilot of 80?
 
Thanks for the explanation, and I agree that it's quite silly of them not to fix this. Leaf owners make up a fairly large portion of EV drivers so any public stations would have to keep this limitation in mind when installing chargers. I hope the Leaf can at least charge if the pilot sends a 70A signal.
 
I am a LEAF and Roadster owner too but my Leaf can't charge at anything over 16A anyway. Are you saying the leaf won't charge at any level if the station sends a max current pilot of 80?

Exactly. It the station presents a pilot signal of 80A the Leaf cancels the charge instead of asking for 16A. This is going to be a growing problem as more 80A stations are installed.
 
I am a LEAF and Roadster owner too but my Leaf can't charge at anything over 16A anyway. Are you saying the leaf won't charge at any level if the station sends a max current pilot of 80?

that's what I'm saying. current J1772 standard includes pilot up to/including 80A, the Leaf can only speak the language up to around 70A. I spent a couple of hours today jabbering my way up the corporate tree at Nissan. the position that I find astounding is that since AV is the only official L2 charging station manufacturer for Nissan, and AV doesn't produce an 80A HAL2 station, that Nissan doesn't care if the Leaf is compatible with a clipper creek 80A station. They literally told me that officially, Nissan does not recommend charging the Leaf on anything but an AV L2 charger, even outside the home, they simply don't support it. I spelled out how ludicrous this is, given that the Leaf's success is actually the main issue right now, that there are way too may Leafs on the road around here for how infantile the charging network is, that the Leaf community is very dependent on non-Nissan official public charging and there is simply no sense in limiting support to only AV charging station since they account for a very small portion of what is available. The guy kept talking in circles and I came close to really loosing it on him. It's going to take a lot more Leaf owners making a stink and being very clear about the request, that the Leaf needs to be fully J1772 pilot signal complient, like every other EV on the road, even the Tesla Roadster, which came out before the Leaf!
 
Can one set the amperage at a J1772 station? I've always just plugged my roadster in and I believe it has charged at 30A.

The Roadster will ask for up to 70A and the station will deliver what it's capable of. You can dial down the current in the car, but you can't exceed the max for the station.

Most US 1772 stations are 30A but in Canada we have a large network of high amp stations thanks to Sun Country Highway.
 
Can one set the amperage at a J1772 station? I've always just plugged my roadster in and I believe it has charged at 30A.

The charging amperage for L2 stations is set by the car's charger but the car needs to be able to recognize the signal from the station to even allow charging to initiate. for the time being we are moving forward with dual amperage modified station from clipper creek, the default is 70A and can work with the Leaf, but it's a work around that will surely be outmoded relatively soon and it adds unnecessary complexity to an otherwise beautifully simple "plug and play" charging station.
 
Can one set the amperage at a J1772 station? I've always just plugged my roadster in and I believe it has charged at 30A.
When a J1772 station is installed, it is programmed with the maximum amount of current it can supply. The limit is based on both the station's capabilities and what the circuit can provide. When a car connects, the station tells the car how much current it is allowed to draw. It is then the car's responsibility to draw no more than the stated current limit.

The J1772 spec released in January of 2010 defines a communication protocol that allows a station to convey a current limit as high as 80A. In the older (2000) version of the spec, the maximum current limit supported by the protocol was 40A. Tesla extended that communication specification to allow 70A charging for the Roadster and Tesla Motors founder Martin Eberhard participated in the committee to update the spec and lobbied them to adopt Tesla's extension. The extension mechanism Tesla created was adopted and defined up to 80A. So, it is a little sad that the Roadster firmware rejected an 80A pilot signal as invalid until this firmware update, but we're glad that problem is fixed now.

Most public stations have a 30A limit, but some are lower and some are higher. Blink and ChargePoint stations, among others, are all limited to delivering at most 30A regardless of circuit capacity. Clipper Creek makes J1772 stations that can deliver up to the full 80A allowed by the spec.

Similarly, when a LEAF sees a 80A pilot signal, perfectly complaint with the J1772 spec, it rejects it as invalid and will not charge at all from that station.
 
http://open-evse.googlecode.com/files/J1772.pdf

evse-spec.png


Apparently some methods of trying to read the duty cycle start to get error prone as it approaches 100%, particularly if the pilot signal is noisy from the EVSE.
Some charger implementations had trouble reading the 80A signal, and had to use a different method to read the signal to get realiable results.
There are different ways you can try to detect the edges of the square wave and calculate the duty cycle.
 
Just like the Roadster before this fix, a LEAF can't cope with a J1772 station that sends out an 80A pilot signal, even though an 80A pilot signal is totally within the J1772 spec. We can perhaps almost excuse the Roadster because it shipped long before the J1772 spec was finalized, but the LEAF has no excuse. (I'm a Roadster and LEAF owner.)
For those that know more of the history...

Setting aside whether the LEAF has an excuse or not, what is the reason? Was there some collaboration with Tesla (or, um, "unofficial borrowing of technology") or did they just make remarkably similar mistakes?
 
My guess would be that they both made a similar mistake / took the same approach that didn't work properly at the 80A level. I really doubt there was any copying involved. They both might point fingers back at EVSE makers saying that the pilot signal wasn't as good as they hoped... Anyways Tesla steps up to the plate with a creative fix... Is Nissan listening?
 
I was told this update might include lowering the minimum cruise control speed from 30 to 25. Can anyone with the new firmware check this?
The update does not change the cruise control engage speed still 30 mph. One change is that I don't seem to get the error when switching between reverse and drive as often. Seems different, but I could have just finally caught on.

Kent
 
One change is that I don't seem to get the error when switching between reverse and drive as often. Seems different, but I could have just finally caught on.
Yes, that was the only other change and only for 2.x Roadsters. It's good to have that confirmed. (I believe that's always been the behavior in 1.5 Roadsters; I frequently pop into D while slowly rolling backwards.) It's mentioned in my description of the changes in my update to the original post in this thread.
 
For people with OVMS, just a few reminders:

(a) Disconnect OVMS before you send your car in. Just unplug it. If you don't, Tesla will (because they need access to the DIAG port to do the firmware upgrade).
(b) Once you get the car back, tap the car boot (trunk) on the "Car" tab of the app, and choose "wakeup". You'll get your temperatures back updating ok after that.
(c) If you no longer have TPMS pressures & temperatures, just drive the car for a minute or so - it will come back.

No problems reported so far, with OVMS and this update.

Regards, Mark.
 
Disconnect OVMS before you send your car in. Just unplug it. If you don't, Tesla will (because they need access to the DIAG port to do the firmware upgrade).
Before turning our Roadster over to the Seattle service center for the update, I unplugged the OVMS. When I got it back and crawled down into the footwell to plug it back in, I found the Roadster tech had already done it. He's been around for years and is obviously used to dealing with OVMS and Tattler. Still, I agree it's a good idea to unplug before turning your car over for service.

If you keep a history of log files for your Roadster, I also recommend downloading the log file before taking it in for service. Although it wasn't an issue with our Roadster for this firmware update, sometimes part (or all) of the log gets reset while being serviced.