Charge increases overnight, every night

[hugh_jassol]

I have a LEMR 3, which i charge to 80% every night. I also have my app set to notify me on charge completion.

I get home from work and plug in... when the charge is complete, it I get a notification on my phone that the car has completed charging and has 206 mi range - this is usually around 10pm. At 7am when I unplug and get in the car to go to work, it has 210 mi range displayed. This has been consistent for me for the past 2-3 months - before that it used to tell me the range was 210 at charge complete and it would still be that in the morning.

This is also reflected in my teslafi data. At end-of-charge, it reports 206 mi range, and at start of drive the next morning, 210 mi.

Any idea what's going on? Can't be temperature as my garage is actually cooler in the morning than when charging is complete...

 

[tccartier]

These acronyms are getting crazy what the hell is an Lemr 3 I get the model 3 okay ...

 

[BluestarE3]

These acronyms are getting crazy what the hell is an Lemr 3 I get the model 3 okay ...

Mid-Range.

 

[AlanSubie4Life]

Any idea what's going on? Can't be temperature as my garage is actually cooler in the morning than when charging is complete...

It could be temperature and the way the SoC estimation curves operate in your car. Open-circuit lithium-ion battery voltage looks like it goes DOWN with increasing temperature. In order not to misinterpret this, the BMS needs to know the temperature, and adjust its estimation of the battery capacity even as the voltage changes overnight. It's not always perfectly steady or accurate - and it's possible it is either just wrong, or the SoC estimation has actually decided MORE energy is available from your battery at a slightly lower temperature. Whether it is right or not? Hard to know... The summary is that it is "complicated." (I don't know why your battery is doing this.) There may also be some other reason other than this particular factor...battery voltage might "rebound" for some reason...

You could try charging much more slowly and see whether your result changes. Just as an experiment.

Lemr 3

Limited Edition Mid Range Model 3, is the interpretation I have heard.

 

[hugh_jassol]

You could try charging much more slowly and see whether your result changes. Just as an experiment.

Good idea - maybe I'll try that. The odd thing is that it worked normally for the first 6 or so months I've had the car. This is a recent (and extremely repeatable) phenomenon.

 

[AlanSubie4Life]

Good idea - maybe I'll try that.

Cool.

The odd thing is that it worked normally for the first 6 or so months I've had the car. This is a recent (and extremely repeatable) phenomenon.

Not all that odd, unless you have not updated the software at all. The weights and formulas and whatever else they do with the BMS is probably changing all the time based on data they gather from batteries in the field, as they age and they better understand the variations between packs. They have to introduce new twists in the SoC estimation algorithm to account for unanticipated effects and outliers. And they probably push these updates to people...and it changes the results you see. Again, totally guessing. All I know is it is a very non-straightforward problem to solve.

 

[justsomeguy]

This happens to me too sort of. I get insane phantom drain at home, but my charge level increases while i'm parked at work.

I think Tesla is doing it to me on purpose so I have no idea how much phantom drain my car actually has.

 

[darth_vad3r]

I've been using the API recently to watch my car finish charging and check on it after charging.

There's definitely some shifting or lagging voltages, or recalibration/re-estimations going on after the charge is complete.

While charging the range in miles (visible in the API with 2 decimal places) will typically increment by a set amount at a time, like 0.48 miles.

e.g. 190.00, 190.48, 190.96...

Now let's say it stopped charging at 190.96. A few minutes later it might say 191.44, then 190.98 (note this is 0.02 more than it ended at).
I think the additional 0.48 ticks up after charge complete are re-estimating after voltages have settled.
I think the smaller adjustments (0.02 difference vs 0.48 differences) are related to re-calibrating the total capacity.

Now that I've noticed this, I was planning on taking a closer look to see if it matters if it's plugged in or not after charge complete to see this micro-adjustment behaviour.

In terms of increase, I've seen as much as 1-2%, say 81% or 82% a couple hours after hitting an 80% charge limit (and being unplugged in this case).

 

[AlanSubie4Life]

I've been using the API recently to watch my car finish charging and check on it after charging.

There's definitely some shifting or lagging voltages, or recalibration/re-estimations going on after the charge is complete.

While charging the range in miles (visible in the API with 2 decimal places) will typically increment by a set amount at a time, like 0.48 miles.

e.g. 190.00, 190.48, 190.96...

Now let's say it stopped charging at 190.96. A few minutes later it might say 191.44, then 190.98 (note this is 0.02 more than it ended at).
I think the additional 0.48 ticks up after charge complete are re-estimating after voltages have settled.
I think the smaller adjustments (0.02 difference vs 0.48 differences) are related to re-calibrating the total capacity.

Now that I've noticed this, I was planning on taking a closer look to see if it matters if it's plugged in or not after charge complete to see this micro-adjustment behaviour.

In terms of increase, I've seen as much as 1-2%, say 81% or 82% a couple hours after hitting an 80% charge limit (and being unplugged in this case).

Interesting. Isn’t the battery voltage also available in the API? Might be interesting to see how it behaves and how it correlates with the other data. Though presumably the relationship will be pretty obfuscated.

 

[darth_vad3r]

Interesting. Isn’t the battery voltage also available in the API? Might be interesting to see how it behaves and how it correlates with the other data. Though presumably the relationship will be pretty obfuscated.

No, you need the CAN bus monitoring to see brick voltages ... I'm not at that level of monitoring (yet anyways ... maybe one day)

 

[AlanSubie4Life]

No, you need the CAN bus monitoring to see brick voltages ... I'm not at that level of monitoring (yet anyways ... maybe one day)

I see. I just thought I saw it reported in TeslaFi stuff I see posted here - but maybe that only gets reported when at a Supercharger (must be...no other explanation that I can think of).

 

[darth_vad3r]

I see. I just thought I saw it reported in TeslaFi stuff I see posted here - but maybe that only gets reported when at a Supercharger (must be...no other explanation that I can think of).

There is a voltage in the API, but it is the charging voltage. I see that during L2 charging. You can see power, current, voltage. No decimal places on any of these though.

 

[AlanSubie4Life]

There is a voltage in the API, but it is the charging voltage. I see that during L2 charging. You can see power, current, voltage. No decimal places on any of these though.

I assume you mean the AC power/current/voltage (from your efficiency analysis).

I’ve seen 400V in TeslaFi data screen captures. I guess it is provided during the Supercharging, similar to the AC data which you can see during L2.

 

[hugh_jassol]

In terms of increase, I've seen as much as 1-2%, say 81% or 82% a couple hours after hitting an 80% charge limit (and being unplugged in this case).

This is exactly the behavior I see as well. It's 80% at charge complete.. then 81% 8 hours later. I haven't checked in between.

 

[darth_vad3r]

I assume you mean the AC power/current/voltage (from your efficiency analysis).

I’ve seen 400V in TeslaFi data screen captures. I guess it is provided during the Supercharging, similar to the AC data which you can see during L2.

Ya it doesn't distinguish between AC or DC voltage in the api, it just calls it charger voltage, so it depends on what you are connected to. Could be 120, 240, or high 300s to 400.

This is what it looks like:

      "charge_current_request": 30,
      "charge_current_request_max": 30,
      "charger_actual_current": 30,
      "charger_phases": 1,
      "charger_pilot_current": 30,
      "charger_power": 6,
      "charger_voltage": 190,