Huge amount of wasted energy caused by undesirable sleep patterns using smart chargers

[vanhtu1987]

I have a Model S 75D with MCU2 & AP3. The car's usual sleep pattern is 58 minutes asleep, 14 minutes awake.

It consumes ~30W whilst being awake. Extrapolating, over a year the car would be awake for 1703h, or 51kWh of energy. This is okay-ish as even at 40p/kWh we'll lose ~20 quid/year.

I've found that the car's sleep pattern changes drastically with what I'd consider a fairly normal usage (steps to reproduced detailed below) where it would be asleep for 8 minutes, then awake for 16 minutes.

What is a little odd is that with this sleep pattern, the car would consume ~100W whilst being awake. Over a year it would be awake for 5840h, or 584kWh of energy. At 40p/kWh we'll lose ~234 quid/year!. This seems far from acceptable.

The behaviours as well as the power consumption/range loss might be different depending on a car's configuration, but to reproduce it:

• Disable all in-car scheduled departure and scheduled charging.
• Plug the smart charger in, but don't let it charge e.g., if it's only scheduled to charge during off-peak hours.
• (Optional) The car's current state of charge is lower than the max SoC e.g., it's currently at 70% but the max charge percentage is 80%.

Once I've learnt of this behaviour, I had to remember to reduce the max SoC to a low value in the morning, and set it backup to my usual value (85%) most days in order to avoid this. Has anyone else observed this issue?.

An example of the undesirable vs. desirable sleep patterns is as follows. The car falls back to sleep for longer periods if:

• The max SoC is set to below the current SoC, or
• The car's scheduled charging time is set: I have Scheduled Charging Time set at 00:30. In which case it would sleep well until 00:30.

 

[GRiLLA]

I have a Model S 75D with MCU2 & AP3. The car's usual sleep pattern is 58 minutes asleep, 14 minutes awake.

It consumes ~30W whilst being awake. Extrapolating, over a year the car would be awake for 1703h, or 51kWh of energy. This is okay-ish as even at 40p/kWh we'll lose ~20 quid/year.

I've found that the car's sleep pattern changes drastically with what I'd consider a fairly normal usage (steps to reproduced detailed below) where it would be asleep for 8 minutes, then awake for 16 minutes.

What is a little odd is that with this sleep pattern, the car would consume ~100W whilst being awake. Over a year it would be awake for 5840h, or 584kWh of energy. At 40p/kWh we'll lose ~234 quid/year!. This seems far from acceptable.

The behaviours as well as the power consumption/range loss might be different depending on a car's configuration, but to reproduce it:

• Disable all in-car scheduled departure and scheduled charging.
• Plug the smart charger in, but don't let it charge e.g., if it's only scheduled to charge during off-peak hours.
• (Optional) The car's current state of charge is lower than the max SoC e.g., it's currently at 70% but the max charge percentage is 80%.

Once I've learnt of this behaviour, I had to remember to reduce the max SoC to a low value in the morning, and set it backup to my usual value (85%) most days in order to avoid this. Has anyone else observed this issue?.

An example of the undesirable vs. desirable sleep patterns is as follows. The car falls back to sleep for longer periods if:

• The max SoC is set to below the current SoC, or
• The car's scheduled charging time is set: I have Scheduled Charging Time set at 00:30. In which case it would sleep well until 00:30.

View attachment 786559

The waking is the similar on my Model 3 under those conditions

1. Not reached desired charge
2. Connected to chargers that isn't providing power
3. No timer or other set on the car

It will wake up every hour and be back asleep within 15 minutes.

However, this is where the data needs to be questioned. My TeslaFi is set to wait 10 minutes before going into it's mode to allow the car to sleep, so even if the car is waking for a few seconds it's possible that TeslaFi is keeping it awake a little longer. I suspect the same is true of your Teslamate.

In terms of loss, TeslaFi is telling me it's lost a couple of hundred watts over 3 hours before I unplugged. I guess I could extrapolate that and assume the same is happening all the time (I would get 200KWH) but that's simply not the case.

So my opinion, it's our monitoring tools that are exacerbating the time the car stays awake. Neither tool will wake the car, but both have built in timers to suspend polling the vehicle details as this keeps an awake car awake.

 

[rotor2k]

It sounds like you aren't using scheduled charging (in the car)? From what I gather, the car is below its desired State of Charge (SoC), and can detect a charger is plugged in, but the charger is not supplying current. So perhaps this causes it to wake up more frequently to see if the charger situation has changed? With scheduled charging disabled it effectively has been told "charge as soon as you can, please".

I would configure the smart charger into dumb mode, and use scheduled charging in the car. That way you plug the car in, it can see the charger is ready to go, but the car says "OK, everything is in order, I shall wake up at the scheduled time and start charging then", and proceeds to "relax" and go to sleep until then.

This is how I charge my Model 3 and it stays asleep with no problem all through the day (even when it is below desired SoC).

 

[vanhtu1987]

Thanks @GRiLLA , it is unclear from the graph above whether the car has reached its desired SoC, but from your description it sounds like your worst-case scenario is my best-case scenario?.

My car never sleeps for over an hour in any condition and I've always assumed that it has been doing so in order to react to adverse weather condition promptly. But some people said that their car could sleep for hours so it would be interesting to see if this is backed up by data.

The TeslaMate docs states that it'd stop polling the car after 3 minutes of going idle Frequently Asked Questions | TeslaMate . However, it's not supposed to wake up the car, which is the issue. Based on what I've seen, the frequent waking up was triggered by the car itself.

Regarding the reported drain during sleep period. I'm out at the minute so I'm unable to pull the data from TeslaMate, but my calculations in the first post was based on the sleep/awake ratio. The amount of expected energy consumed would be in line with reality, if we could trust the reported energy consumption.

 

[Kansalis]

It sounds like you aren't using scheduled charging (in the car)? From what I gather, the car is below its desired State of Charge (SoC), and can detect a charger is plugged in, but the charger is not supplying current. So perhaps this causes it to wake up more frequently to see if the charger situation has changed? With scheduled charging disabled it effectively has been told "charge as soon as you can, please".

I have scheduled charging set up on my PodPoint & my car sleeps fine when it's plugged in waiting for the charger to start. It then wakes up to charge & goes back to sleep 11 minutes after the charge stops, every time. The car sometimes wakes up randomly for 13 minutes, but not often. I never open the Tesla app on my phone unless I'm about to go somewhere, so it doesn't wake up the car. Software version 2022.8.2.

This is the chart from a couple of days ago when I last charged. The awake times around the driving on the 24th are sentry mode.

Edit for detail: The first charge you see on the left got the car to 65% & stopped at the end of my GoFaster window as it was after a long drive. The charge in the middle was finishing the topup the next night.

 

[WannabeOwner]

If your overnight parasitic drain is noticeable I would suggest turning off TeslaMate for one night and seeing if that makes any difference to SoC drop overnight. Just in case.

Actually, rather than that, change your Tesla Account password. If something else, which you have forgotten about and long since stopped using , is connecting to the car that might be the culprit. Changing the password will prevent access from everything, phone APP too.

If that cures the problem something is waking the car up. If not then its the issue of not having scheduled charging and the car checking to see if it there is mains available and to start charging. (Or some other 1st world problem like that!)

 

[vanhtu1987]

It sounds like you aren't using scheduled charging (in the car)? From what I gather, the car is below its desired State of Charge (SoC), and can detect a charger is plugged in, but the charger is not supplying current. So perhaps this causes it to wake up more frequently to see if the charger situation has changed? With scheduled charging disabled it effectively has been told "charge as soon as you can, please".

I would configure the smart charger into dumb mode, and use scheduled charging in the car. That way you plug the car in, it can see the charger is ready to go, but the car says "OK, everything is in order, I shall wake up at the scheduled time and start charging then", and proceeds to "relax" and go to sleep until then.

This is how I charge my Model 3 and it stays asleep with no problem all through the day (even when it is below desired SoC).

I had considered doing so but it has a few drawbacks compared to using a smart charger:

• The car is unable to charge based on solar panels output to minimise export.
• The car doesn't allow us to charge only during a set period (off-peak hours) without workarounds such as manually setting the max SoC every night. In most cases, a 77% SoC is as good as 80%.
• The car's charging pattern is not as good for the battery. With the Ohme that I've got, it adjusts the charging curve in a way as to minimise the time the car spent charged at high current. In contrast, the car only charges at a flat 32A.

I agree that the symptom was most likely caused by the car keep coming back to check for charger's power. It is just unfortunate that it does so, because I'm unable to think of a scenario where that'd actually help: the charger shall produce power when it wants to, and it will wake up the car at the right time too. There doesn't seem to be a good reason for the car to keep checking since it won't be able to remedy the situation.

 

[Kansalis]

I agree that the symptom was most likely caused by the car keep coming back to check for charger's power.

Not sure if it's a difference between the MS & the M3, but as you can see from my post above my car doesn't do that at all. It just stays asleep until the charger wakes it with power.

 

[vanhtu1987]

I have scheduled charging set up on my PodPoint & my car sleeps fine when it's plugged in waiting for the charger to start. It then wakes up to charge & goes back to sleep 11 minutes after the charge stops, every time. The car sometimes wakes up randomly for 13 minutes, but not often. I never open the Tesla app on my phone unless I'm about to go somewhere, so it doesn't wake up the car. Software version 2022.8.2.

This is the chart from a couple of days ago when I last charged. The awake times around the driving on the 24th are sentry mode.

Edit for detail: The first charge you see on the left got the car to 65% & stopped at the end of my GoFaster window as it was after a long drive. The charge in the middle was finishing the topup the next night.

View attachment 786581

That is a very good sleep pattern, one that I haven't seen with mine. Another with the Model 3 has confirmed similar behaviour though. Could you please clarify your car's model & year number?.

 

[Kansalis]

That is a very good sleep pattern, one that I haven't seen with mine. Another with the Model 3 has confirmed similar behaviour though. Could you please clarify your car's model & year number?.

2022 Model 3, I've had it a month. On the 2022.8.2 software now but it's been the same since I got it, through various versions.

 

[vanhtu1987]

If your overnight parasitic drain is noticeable I would suggest turning off TeslaMate for one night and seeing if that makes any difference to SoC drop overnight. Just in case.

Actually, rather than that, change your Tesla Account password. If something else, which you have forgotten about and long since stopped using , is connecting to the car that might be the culprit. Changing the password will prevent access from everything, phone APP too.

If that cures the problem something is waking the car up. If not then its the issue of not having scheduled charging and the car checking to see if it there is mains available and to start charging. (Or some other 1st world problem like that!)

Thanks @WannabeOwner , I've tried changing my passwords mutiple times, as well as not having any 3rd party services connected but still observed this issue.

I don't have any data from my car when it had the MCU1 so couldn't confirm. It'd be interesting if others with older Model S/3 could chime in.

@Kansalis has shown that his Model 3 has a good sleep pattern so it seems that it's not a software issue but something intentional from Tesla.

I've come to accept this limitation as a quirk of the platform, but would be very much interested in seeing if owners with the same car like mine (2017 MS 75D with MCU2, AP3) don't have this issue .

 

[vanhtu1987]

2022 Model 3, I've had it a month. On the 2022.8.2 software now but it's been the same since I got it, through various versions.

Thanks, it's something to look forward to whenever the 7 seater Model Y arrives at this side of the pond . One fewer thing to micro-manage.

 

[WannabeOwner]

I've come to accept this limitation as a quirk of the platform, but would be very much interested in seeing if owners ... don't have this issue

I'm in the process of commissioning a Zappi to use excess solar (in additional to the existing Tesla charger)

I may find that that gives me the same problem, if so I'll let you know.

I have yet to find out how I'm going to control this. My cars are set to start charging at the start of overnight cheap rate. This is based on them being "at home". The Zappi is only a couple of meters away, I doubt I can set up a different geofence! so I presume if a car is plugged into the Zappi its going to think that it should be trying to charge at start of off-peak ...

... maybe Zappi will be programable to be "disconnected" at that time, given that I won't be exporting from PV in the middle of the night. That would be a solution ...

... but maybe that is also going to mean the car will wake up every hour knowing that cable-is-connected but no-power-available to see if power is available

However, that scenario is like a power cut, surely? Restoration of power ought to be something that it could trigger immediately, rather than waiting potentially 59 minutes after power comes back on before resuming charging.

 

[vanhtu1987]

I'm in the process of commissioning a Zappi to use excess solar (in additional to the existing Tesla charger)

I may find that that gives me the same problem, if so I'll let you know.

I have yet to find out how I'm going to control this. My cars are set to start charging at the start of overnight cheap rate. This is based on them being "at home". The Zappi is only a couple of meters away, I doubt I can set up a different geofence! so I presume if a car is plugged into the Zappi its going to think that it should be trying to charge at start of off-peak ...

... maybe Zappi will be programable to be "disconnected" at that time, given that I won't be exporting from PV in the middle of the night. That would be a solution ...

... but maybe that is also going to mean the car will wake up every hour knowing that cable-is-connected but no-power-available to see if power is available

However, that scenario is like a power cut, surely? Restoration of power ought to be something that it could trigger immediately, rather than waiting potentially 59 minutes after power comes back on before resuming charging.

Thanks @WannabeOwner , there seems to be 2 separate issues at play.

1) Telling the car to charge outside of off-peak hours with Scheduled Departure/Charging Time enabled. We'd normally need to initiate the charging manually outside of the schedule.

This is something that likely needs support from the charge point. Not only does it need to give power, it also needs to tell the car to start charging in order to charge during the day. The Ohme doesn't do so, it'd be interesting to see if the Zappi/Hypervolt does.

2) Letting the car go to sleep when charger is connected (data and ground pins) but power pins aren't.

At the minute it looks like only latest-generation Teslas won't suffer from this issue.

That does give me an idea though. We might be able to solve these issues with:

• For charging off solar power: as long as it's possible to communicate with the Zappi to get the excess solar power, we could have a script that forces the car to charge/stop charging at the same time.
• For getting a good sleep: perhaps the same script could detect when the car stopped charging/charged stopped providing power and unlock the charge port?. This assumes that an unlocked charge port lets the car go back to sleep and it won't try to lock it again..

 

[WannabeOwner]

At the minute it looks like only latest-generation Teslas won't suffer from this issue.

Pro: Very good trade in prices at present making upgrade/refresh look cheap .

Con: No Model-S available

I expect that was no help at all!

we could have a script that forces the car to charge/stop charging at the same time.

Would dropping the LIMIT overnight be any good?

My overnight charging script, based on when I did have to get up early to commute, dropped limit to 80% 5 minutes before start of off-peak, Car's scheduler then started the charge. That finished whenever 80% was reached. My script then changed the limit to 90% an hour or so before expected departure [I left early, that was still within off-peak] and started charging (if it had finished / stopped) .. and then turned on climate etc. a bit before expected departure.

 

[GRiLLA]

Thanks @GRiLLA , it is unclear from the graph above whether the car has reached its desired SoC, but from your description it sounds like your worst-case scenario is my best-case scenario?.

My car never sleeps for over an hour in any condition and I've always assumed that it has been doing so in order to react to adverse weather condition promptly. But some people said that their car could sleep for hours so it would be interesting to see if this is backed up by data.

The TeslaMate docs states that it'd stop polling the car after 3 minutes of going idle Frequently Asked Questions | TeslaMate . However, it's not supposed to wake up the car, which is the issue. Based on what I've seen, the frequent waking up was triggered by the car itself.

Regarding the reported drain during sleep period. I'm out at the minute so I'm unable to pull the data from TeslaMate, but my calculations in the first post was based on the sleep/awake ratio. The amount of expected energy consumed would be in line with reality, if we could trust the reported energy consumption.

Teslamate isn't waking the car, the car is waking itself to see if power is being provided by the charger, however TeslaMate is therefore detecting that and then triggering it's strategy to allow the car to go back to sleep. For the 3 minutes TeslaMate is preventing the car going back to sleep. After the three minutes TeslaMate stops querying the Vehicle Data API for 15 mins, during which time it doesn't know if the car is awake or asleep.

My point being that TeslaMate and TeslaFi aren't able to handle this well, and assume the cars are awake more than they really are.

If I unplug or allow the car to reach it's target then it sleeps entirely normally, typically waking once a day at most.

 

[WannabeOwner]

My point being that TeslaMate and TeslaFi aren't able to handle this well, and assume the cars are awake more than they really are

Good point.

There is something in TeslaFi (TeslaMate too maybe) where, when it knows the last status was "asleep", that it asks about the HQ-status for the car instead (to avoid waking up the car). But I have had TeslaFi miss the start of a journey, so whatever it is doing isn't getting info from HQ when the car is actively on the move (or my TeslaFi settings need some tweaking). The particular one is "deep sleep" which I can set for e.g. night-time and then its almost certainly going to miss any journey made in that interval

 

[Chanser]

Zappi, M3 and Teslamate working fine. My M3 has been sleeping for 24+ hours.
I only set a schedule on my Zappi.

 

[init6]

So my opinion, it's our monitoring tools that are exacerbating the time the car stays awake. Neither tool will wake the car, but both have built in timers to suspend polling the vehicle details as this keeps an awake car awake.

If the OP is using the default Streaming API method on Teslamate then Teslamate doesn't talk to the car and doesn't poll the vehicle. It only reads the stream coming from Tesla's servers.

 

[GRiLLA]

If the OP is using the default Streaming API method on Teslamate then Teslamate doesn't talk to the car and doesn't poll the vehicle. It only reads the stream coming from Tesla's servers.

Sorry that's irrelevant. All apps only read data from the Tesla servers, calling some of those endpoints will still keep the car awake as the Tesla servers makes a call to the car in response to the request.