eGolf and EVSE pilot sequencing - problem

[nsayer]

Well, I've tested this one quite a few times, and I'm fairly confident it's confirmed. If an eGolf is plugged into a "paused" EVSE that wakes up some time later, the eGolf won't notice and won't start charging.

This one is a little esoteric, but here's the gist of it:

The pilot is the signal from the EVSE that indicates the available current that the vehicle is allowed to draw. It's in the form of a +/- 12 volt square wave, whose duty cycle indicates the ampacity. But when that signal isn't oscillating - that is, when it's pinned to +12 volts - it's an indication to the vehicle that the EVSE is not ready. The EVSE is allowed to maintain this state indefinitely. When the EVSE becomes ready - for whatever reason - it may make an offer to the vehicle by beginning the 1 kHz oscillations.

If the eGolf is "asleep," it won't notice this happening.

I plugged my eGolf into a Hydra at work this morning, but it was configured for "sequential" mode. What this means is that it allows one vehicle to charge at full power until it is finished, holding the other vehicle at bay. When the first vehicle is finished, it will transfer the oscillating pilot to the second vehicle, allowing it to charge. The benefit of this mode of operation is that it allows the transfer to happen automatically and unattended. Unless the second vehicle is an eGolf, it would seem.

I clicked the "lock" button on my key fob and the car woke up and immediately began charging, but it had been sitting there ignoring the EVSE for an indeterminate period of time before I checked it.

It's possible that the eGolf "polls" the EVSE and that when I checked on it it had not yet gotten a chance to do so. But that seems less likely to me than the alternative that it would have just sat there forever.

If others have noticed that EVSE charge timers malfunction in a similar way, then I believe this is the explanation.

 

[JoulesThief]

Well, I've tested this one quite a few times, and I'm fairly confident it's confirmed. If an eGolf is plugged into a "paused" EVSE that wakes up some time later, the eGolf won't notice and won't start charging.

This one is a little esoteric, but here's the gist of it:

The pilot is the signal from the EVSE that indicates the available current that the vehicle is allowed to draw. It's in the form of a +/- 12 volt square wave, whose duty cycle indicates the ampacity. But when that signal isn't oscillating - that is, when it's pinned to +12 volts - it's an indication to the vehicle that the EVSE is not ready. The EVSE is allowed to maintain this state indefinitely. When the EVSE becomes ready - for whatever reason - it may make an offer to the vehicle by beginning the 1 kHz oscillations.

If the eGolf is "asleep," it won't notice this happening.

I plugged my eGolf into a Hydra at work this morning, but it was configured for "sequential" mode. What this means is that it allows one vehicle to charge at full power until it is finished, holding the other vehicle at bay. When the first vehicle is finished, it will transfer the oscillating pilot to the second vehicle, allowing it to charge. The benefit of this mode of operation is that it allows the transfer to happen automatically and unattended. Unless the second vehicle is an eGolf, it would seem.

I clicked the "lock" button on my key fob and the car woke up and immediately began charging, but it had been sitting there ignoring the EVSE for an indeterminate period of time before I checked it.

It's possible that the eGolf "polls" the EVSE and that when I checked on it it had not yet gotten a chance to do so. But that seems less likely to me than the alternative that it would have just sat there forever.

If others have noticed that EVSE charge timers malfunction in a similar way, then I believe this is the explanation.

While not as knowledgeable or observant as you, I notice that any time you lock or unlock, it "wakes" up the e-Golf, be it with key, or touching the indentation on the door handle, or reaching hand in door handle with keyfob in pocket.

I have not tried delayed charging feature with either VW Car-Net or with the JuiceBox 40 Pro application on my smart phone, to see if either one or the other functions, mainly because LADWP provides me with no incentive in price per kwh, and I'm not curious enough with other cars that need to be driven also, that compete for garage space.

 

[miimura]

Somehow, the ChargePoint Home EVSE gets around this issue and is able to successfully implement its own timers with the e-Golf. It might be interesting for a technical person like Nick to find out their implementation.

I have long speculated that if you interrupted the presence pin that it would also wake up the car and it would be able to charge unattended. The problem with this approach is that there is usually not a conductor between the J1772 handle and the EVSE for the presence pin.

 

[nsayer]

I'd be very curious indeed to learn how else an EVSE is allowed to do something like that.

In principle, I suppose, an EVSE could simply offer a high-impedance ("dead") pilot. That'd be the same as an EVSE that had no power at all (circuit breaker blown/off). I'd kind of be surprised if that was the way. The advantage to the EVSE of offering a +12 pilot is that it can know whether there's a vehicle there or not by sensing the 2.7k impedance that a vehicle will impose on the positive side (there's a diode to ground, so pinning the pilot to -12 denies the EVSE the opportunity to do so, as the diode will be reverse biased).

And pinning the pilot to +12 to indicate "not ready" is perfectly within-spec. It's how most of the "pay for play" EVSEs make the vehicle wait until you've paid/authenticated/genuflected.

Waking the car up is a workaround - however you do it - but in principle, it ought not to be necessary.

I've sent a detailed e-mail with my findings to my dealer. If they have some conduit into some level of Volkswagen that can eventually reach some sort of engineering support department, then there's a chance for an answer to come back.

 

[miimura]

I've sent a detailed e-mail with my findings to my dealer. If they have some conduit into some level of Volkswagen that can eventually reach some sort of engineering support department, then there's a chance for an answer to come back.

I wouldn't count on that path giving any result at all. You would be better off opening a case with VW Customer Service and asking for an e-Golf specialist to contact you. That would have a much better chance of having some kind of conduit to people that are in a position to address it. JMHO.

 

[REM]

That's interesting. Thank you for sharing. For some reason when my e-Golf goes to sleep it triggers the fault light in my Leviton 160 charge station. It reads fault if charging is complete or if I set the charge timer (which never wakes the car up to charge). It will briefly go back to ready if I wake the car up, but then quickly go back to sleep and read fault again. It doesn't behave this way with the EVSE that came with the car. I'm curious to hear updates on this subject.

 

[miimura]

That's interesting. Thank you for sharing. For some reason when my e-Golf goes to sleep it triggers the fault light in my Leviton 160 charge station. It reads fault if charging is complete or if I set the charge timer (which never wakes the car up to charge). It will briefly go back to ready if I wake the car up, but then quickly go back to sleep and read fault again. It doesn't behave this way with the EVSE that came with the car. I'm curious to hear updates on this subject.

That is not normal. My Leviton EVB40 and Jesla don't do that. They are perfectly happy to stand by and wait for the car to wake up and start charging. Even the Clipper Creek units that don't work for delayed charging don't show any fault.

 

[nsayer]

I wouldn't count on that path giving any result at all.

Surprisingly enough, I got an answer back the same day. They say that this is fixed by a firmware update. I got (was suckered into?) a fabric and paint sealant treatment as part of the deal that's going to be done next week and have asked for this to be done at the same time.

 

[f1geek]

Is this firmware update a TSB? Can you tell us exactly what the firmware update changes?

Thanks!

 

[JoulesThief]

Surprisingly enough, I got an answer back the same day. They say that this is fixed by a firmware update. I got (was suckered into?) a fabric and paint sealant treatment as part of the deal that's going to be done next week and have asked for this to be done at the same time.

Please get the number of the firmware update before having the work done and post it up here, to add to the knowledge base.

 

[nsayer]

Surprisingly enough, I got an answer back the same day. They say that this is fixed by a firmware update. I got (was suckered into?) a fabric and paint sealant treatment as part of the deal that's going to be done next week and have asked for this to be done at the same time.

Please get the number of the firmware update before having the work done and post it up here, to add to the knowledge base.

Will do.

 

[miimura]

I wouldn't count on that path giving any result at all.

Surprisingly enough, I got an answer back the same day. They say that this is fixed by a firmware update. I got (was suckered into?) a fabric and paint sealant treatment as part of the deal that's going to be done next week and have asked for this to be done at the same time.

I will be very interested to see if you get something different than the TSB discussed in this thread that has been available for about 7 months. You have a 2016, so it may be something completely different. Or not.

 

[nsayer]

Today's the day, and the service advisor says there are no firmware updates available. So, right hand, meet left hand. :?

I've asked him to talk with the salesman to see what it was he was told while the car is in. We'll see what happens.

 

[dparson]

I'm currently using an OpenEVSE and had to perform a code change in order for it to "wake up" the charging system on a 2015 eGolf. I describe the necessary code change in this thread http://www.myvwegolf.com/forum/viewtopic.php?f=12&t=148 I'm trying to reply to all the threads I see about this issue today to get the word out, but there are a lot!

 

[JoulesThief]

I'm currently using an OpenEVSE and had to perform a code change in order for it to "wake up" the charging system on a 2015 eGolf. I describe the necessary code change in this thread http://www.myvwegolf.com/forum/viewtopic.php?f=12&t=148 I'm trying to reply to all the threads I see about this issue today to get the word out, but there are a lot!

It would be a good idea to start a thread on "Open Source EVSE Programming of delayed charging for your 2015 e-Golf" That way, if someone does a search, the Open Source brand EVSE will come up, under that heading, as well as the programming how to.

Thank you for sharing this discovery!