Charging and logging

Integrations OpenEVSE
1

With a few interruptions (I believe due to OCPP, now disabled) our OpenEVSE reliably charged our Smart ForTwo (451) ED3 for over a year. The EVSE’s timer is set for 00h30 to 05h30 and if plugged in and needing it, the car would charge. Same with a different EV.

Now struggling with a newer Smart ForTwo (453) EQ. Sometimes it will charge overnight, but more often than not, it doesn’t and I’m trying to figure the reason why.

Obviously, first thing to check is the EVSE’s log and that clearly shows it turning on at the appointed time (allowing for random variation) but the next event can be when it goes back to sleep with no charge having started.

I say ‘clearly’ as the log is actually far from clear. Each ‘event’ shows the status, but is the event actually created due to a change in status, or some other reason why it decides to create the log entry and the status is just what it logs? Log entries need more detail as to the reason for the entry and what the status means, e.g. has it just changed? From what to what? To be truly useful logging needs improvement.

Anyway, on the occasions it does charge, e.g. like last night, the EVSE awoke at 00h22. At least I believe the ‘car connected’ symbol means that is when it woke up. Did I say the log needs more detail. :grinning_face:

Assuming I’m right and it awoke at 00h22, the next entry is the lightning bolt that I think means the start of charging as at that stage it’s still 0 kWh. The next entry some time later is another ‘car connected’ which I assume means end of charge and it reports a number of kWhs that I again assume means how much charge was taken. Then of course the ‘no entry’ symbol when the EVSE goes back to sleep (although many such entries throughout the day with no indication of why).

Ok, as long as my assumptions are correct, all looks good, except…

The EVSE awoke at 00h22 but charging did not start until 01h24, so over an hour later. Huh?

Looking back over the log I can see typically the same behaviour. When charging does actually start after the EVSE wakes up, it is always quite some time after (sometimes up to 4 hours), if it starts at all, but sometimes it never gets going before the EVSE goes back to sleep. Which makes me think the problem is related. This delay before charging starts always occurs and sometimes it’s just too long.

So is this the car or the EVSE?

I have been thinking that the problem is the car which is going to sleep and not noticing the charge becoming available when the EVSE wakes up. This is supported by the fact that charging seems to work when the car is connected with less time before the EVSE wakes up, but if it’s been sitting connected for too long, then the car seems to have gone into its own sleep state and doesn’t wake up. I’m currently trying to ascertain how long it takes to go to sleep and currently my testing indicates more than 12 hours. So less than that allows charging to work, but some more time than that and it does not. I cannot be more precise as I’m still testing to confirm my findings and extending the time interval to determine the actual time it takes the car to go to sleep. But…

Maybe the problem is not the car going to sleep at all. Maybe it is all down to this charge delay I have just noticed. Why is there any delay at all after the EVSE wakes up. Obviously will take a few seconds for the car to see the availability of charge and to say “yes please”, but that cannot take several minutes, let alone hours. So what is going on?

Is the EVSE for some reason not supplying charge until such a long time after the car requests it, or is the car taking a long time to accept the charge being offered and why would that be dependent on how long the car has been waiting since being connected.

So could do with some support here from those with direct involvement in OpenEVSE software and experience of its behaviour with different EVs.

2

It is of concern to me that there is apparently no support whatsoever available from OeM for their customers who have purchased an OpenEVSE from them. :unamused_face:

3

That is disappointing.

I don’t experience your problems here but there is an option to change the pause status at the bottom of the “EVSE” configuration page which might help.

4

Thanks, yes I had seen that, but changing from ‘sleep’ to ‘disable’ apparently means the EVSE can no longer detect if a car is connected while it is so paused (i.e. disabled).

I have just changed it though so I can at least then see if it makes any difference.

Another oddity I see is that when the EVSE is enabled by its timer, if the car does charge, it sometimes takes hours before it starts charging, but I was wondering if the issues are related. Is this apparently random delay after the EVSE wakes up before the car starts charging sometimes just so long that by then the EVSE has gone back to sleep?

I have no idea what is causing this delay. All other cars I have used with this EVSE start charging almost immediately (very soon after) the EVSE wakes up. Why does this Smart not do that? If it is indeed because it has gone to sleep, then charging would not then start some hours later, without some event in the meantime that woke up the car, but there never is any such event.

I cannot help thinking though that this random (often large) delay before charging starts and the fact that it sometimes never does actually charge are somehow related.

Be good if OeM could chime in and offer their thoughts.

5

Without knowledge of the specific car I can only speculate, but maybe it wakes up for some unrelated reason, and if that happens to fall within an EVSE charging window then it begins to take a charge.

6

It’s a Smart ForTwo 453 EQ having the problem. The previous 451 ED had no such issue, but the 2 models use entirely different systems.

It had occurred to me that maybe sometimes it is coincidentally waking itself up for other reasons within the charging window, but hard to know for sure and I can determine no pattern to its behaviour.

I have an OVMS to install that might allow me to check such things, but also should allow me to wake it up at the appropriate time to ensure it does charge. In the meantime though I prefer to actually know what the problem is, even if I do manage to overcome it.