How many Kwh usable for driving are you getting on an 85 or 90 car? (I only get 75Kwh on my 90D)

[marcmerlin]

I know you don't get the full battery capacity for driving, some is kept for emergency reserve and some just not to run the batteries flat, which would damage them.
I also know that Tesla apparent does not want to disclose what the usable Kwh for driving is, for our cars (if I'm wrong, please correct me).

So on a new car, I used the battery percent meter, did a long trip, caculated percent loss and how many Kwh were used for that trip in the trip meter.
You then extrapolate to 100% use, and get an approximate Kwh usable for driving if things work well.
But in my car I'm only getting 75-76Kwh, which is way short of 90Kwh and unofficial numbers of 82Kwh usable for driving, that I heard.

I did my tests without stopping the car during the drive and with the battery charging and warm just before I left.
I did the test twice 2 days apart, uphill from Manteca supercharger to Kirkwood, CA, and downhill from Kirkwood back to cupertino, CA
Temperatures were not warm, but above freezing.


First leg: Manteca supercharger to kirkwood was 85% to 15%, i.e. 70%
used: 53Kwh on trip meter.
If 70% is 53Kwh, 100% is 75.7Kwh

Second leg: Kirkwood to home was 80% to 20%, 60% use showed 44.4Kwh on trip meter
That means 100% is 74Kwh
187 miles, 3h26, very good consumption of 238Wh/mi due to the big downhill.

Any idea why I'm getting so much less than the rated battery capacity?
Did others do similar tests and get unofficial numbers of Kwh usable for driving on their car?

Thanks.

 

[hockeythug]

Here we go again

 

[mgboyes]

Here we go again

Has this issue (the fact that 90kWh packs appear to have no more usable capacity then the 85s) been covered elsewhere? Would love a pointer to the thread - and sorry if my searching has been inept.

To be totally clear I am not concerned about

• battery degradation over time
• the inaccuracies of capacity measurement
• the fact that different people with different driving styles get different mileage ranges from their cars
• the fact that some of the capacity is reserved for anti-bricking
• the fact that the 90kWh pack quite probably does not have a 90kWh total capacity (just like the 85 is not actually an 85).

My specific issue (and that of a growing number of UK owners) is that the 90 is supposed to have 6% more usable capacity (i.e. 6% more range) than an 85 but nobody that I can find has ever actually experienced that. The usable capacity of a new 85kWh pack is well known to be about 77kWh but those with 90 packs in the UK are all finding that they can get no more than about 75kWh usable from theirs before they hit empty.

 

[marcmerlin]

Has this issue (the fact that 90kWh packs appear to have no more usable capacity then the 85s) been covered elsewhere? Would love a pointer to the thread - and sorry if my searching has been inept.

To be totally clear I am not concerned about

• battery degradation over time
• the inaccuracies of capacity measurement
• the fact that different people with different driving styles get different mileage ranges from their cars
• the fact that some of the capacity is reserved for anti-bricking
• the fact that the 90kWh pack quite probably does not have a 90kWh total capacity (just like the 85 is not actually an 85).

My specific issue (and that of a growing number of UK owners) is that the 90 is supposed to have 6% more usable capacity (i.e. 6% more range) than an 85 but nobody that I can find has ever actually experienced that. The usable capacity of a new 85kWh pack is well known to be about 77kWh but those with 90 packs in the UK are all finding that they can get no more than about 75kWh usable from theirs before they hit empty.

Thanks for your data point, at least it seems to help confirm that maybe I'm not crazy

Hockeythug: range vs battery vs usable battery has been discussed a lot, but I didn't find a thread accessing my exact concerns. Since you seemed to have rolled your eyes over my message, please provide the threads you think I should have found and read

Thanks,
Marc

 

[marcmerlin]

Actually if I am to believe the numbers from this thread Tesla's 85 kWh rating needs an asterisk (up to 81 kWh, with up to ~77 kWh usable) my 90D actually is getting slightly less usable Kwh than an 85D :-/

 

[jeffro01]

The problem is your math is completely unusable for what you're trying to use it for. You have no input regarding other sources of potential power draw. Your drawing conclusions based on incomplete data. I certainly concede that it appears, just from Jason's work, that 90 does not equal 90 and 85 does not equal 85. However, the discrepancies simply aren't that substantial.

Jeff

 

[whitex]

Unfortunately as jeffro01 said, the method described here has a lot of holes. I know it's tempting to do so and I tried something similar a while back - I even filed a bug with Tesla showing used and remaining capacity I captured with a gopro every 10 seconds, then graphing those numbers (my complaint was that I appear to have a ~1KW continuous draw that was not accounted for). Here are some of things I learned that you may find useful:
1. Not all power drain seems included in the reported consumption numbers
2. The % is scaled such that 0% will show up at the 0 remaining miles, which is far from actual 0% of the battery, and above the non-brick buffer.
3. The % is not always linear, 1KW on the top end will end up a different percentage than on the bottom end of SoC.

Checkout wk057's thread linked here earlier - it includes a lot more detailed approach to figuring this out.

 

[marcmerlin]

Hi Jeff and whitex, thanks for your answers.
I would actually like to know ways my math is wrong because if not, things are looking very bad.

1) other sources of power draw. If I'm short 4 to 6Kwh on a 3h drive, , that's up to 2000W unaccounted for. That's a *lot* of energy to run the headlights, car computers, radio, and whatever else might not be shown on the trip meter. I'm assuming that Teslas do not use 1500-2000W when driving, of power that isn't going to the drivetrain. Would you agree?

2) I know that 0% is not empty battery. I said that If I end up with a 76Kwh battery per my calculations, you add 4Kwh for anti bricking (as per Tesla's 85 kWh rating needs an asterisk (up to 81 kWh, with up to ~77 kWh usable) ) and that's still 80Kwh of battery out of 90Kwh.
Sure, I'm probably missing a few things, but that's 10Kwh missing, that's a *lot* of power, I could run my house for half a day on that (and I have a lot of crap in my house)

3) % not being linear. That one is a fair point, I was wondering about that one too.
But I measured 85% to 15% and 80% to 20%
That's a pretty wide range of the battery percentage graph. I agree I'm missing 30 to 40%, but do you think that would account for 8 to 10Kwh? That sounds like a big difference.

4) More importantly, why are people with 85D cars doing the same math than me, ending up with 77 or 78Kwh usable for driving (i.e. more than me)?

 

[apacheguy]

Need a CAN logger, stop extrapolating. This is not reliable. My 85 reports 75.4 kWh for a full charge.

 

[marcmerlin]

Need a CAN logger, stop extrapolating. This is not reliable. My 85 reports 75.4 kWh for a full charge.

I understand what a CAN logger is, but I'm not sure how it stops you from extrapolating.
If you're not driving the car from 100% to 1% or somesuch, you're going to be extrapolating, right?

 

[stopcrazypp]

I understand what a CAN logger is, but I'm not sure how it stops you from extrapolating.
If you're not driving the car from 100% to 1% or somesuch, you're going to be extrapolating, right?

The CAN logger reports what the car thinks it has in kWh. No need to extrapolate. However, this may still be subject to battery balancing leading to a lower reported kWh.

Even if you don't extrapolate, the trip meter does not give an accurate accounting of all energy consumed. All extrapolating does is adding additional error to that.

 

[marcmerlin]

So the CAN logger will give you Kwh left as opposed to percent left?
If it were that easy, people would know that their 85D really only has (apparently) 80Kwh of capacity.
Can you describe in more details exactly what you get from the CAN logger and how it is not extrapolating itself? as well as whether it gives you your real battery capacity when the car is 100% charged?

As for trip meter not giving you all energy consumed, again, I agree, but one more time, in a 3H drive, I'm off by 6 to 10Kwh, that's a *lot* of energy not going towards driving, more than is believable.
But let's turn that around: for normal driving with let's say radio, headlights, and other stufff you can't turn off, how many Watts go to those systems? From there, it's easy to compute how many Kwh go towards things other than drivetrain and compute an estimate of what it would be after 3H.

 

[apacheguy]

It reads the values directly off the BMS. There is no extrapolation on the user end.

 

[marcmerlin]

It reads the values directly off the BMS. There is no extrapolation on the user end.

Ok, then can Tesla read what charge the BMS thinks I have (in Kwh) if I charge the car to 100%?
That would lay the issue to rest, but my service manager said it wasn't possible when I asked him (hence my confusion)

 

[amallah]

I am also interested in some way to monitor my battery capacity and show that the increased price of my 90 is "worth it". I've had my 90D for a short amount of time, but I am not sure that the extrapolation method is really useful. If you did the same method on an ICE car, you'd be saying measure the mileage when the needle is at 3/4 tank, then at 1/4 tank and use that information to determine the size of the tank - I don't think that makes sense.

Hi Jeff and whitex, thanks for your answers.
I would actually like to know ways my math is wrong because if not, things are looking very bad.

1) other sources of power draw. If I'm short 4 to 6Kwh on a 3h drive, , that's up to 2000W unaccounted for. That's a *lot* of energy to run the headlights, car computers, radio, and whatever else might not be shown on the trip meter. I'm assuming that Teslas do not use 1500-2000W when driving, of power that isn't going to the drivetrain. Would you agree?

Other sources of power draw not accounted for in distance measurements - tailwinds/headwinds, elevation changes, exterior temperature vs climate control settings, tire pressure, speed travelled and driving style (acceleration/deceleration events). I think differences in any of those between two Tesla drivers could make up for 2kWh easily. If you go to evtripplanner.com and check their calculations, a 5mph headwind can reduce the range by over 15%.

I think it's probably more accurate to take the % at empty and if you can somehow get the amount of energy used to charge it back up to 100%. At least there are less confounds in that math.

 

[marcmerlin]

I am also interested in some way to monitor my battery capacity and show that the increased price of my 90 is "worth it". I've had my 90D for a short amount of time, but I am not sure that the extrapolation method is really useful. If you did the same method on an ICE car, you'd be saying measure the mileage when the needle is at 3/4 tank, then at 1/4 tank and use that information to determine the size of the tank - I don't think that makes sense.


Other sources of power draw not accounted for in distance measurements - tailwinds/headwinds, elevation changes, exterior temperature vs climate control settings, tire pressure, speed travelled and driving style (acceleration/deceleration events). I think differences in any of those between two Tesla drivers could make up for 2kWh easily. If you go to evtripplanner.com and check their calculations, a 5mph headwind can reduce the range by over 15%.

I think it's probably more accurate to take the % at empty and if you can somehow get the amount of energy used to charge it back up to 100%. At least there are less confounds in that math.

Sorry, most of your arguments do not apply. You are talking about range in miles, I'm talking about Kwh use.
Most of what you mentioned impacts range, but does not impact the Kwh used number and percentage of battery left, being farther away from reality

 

[stopcrazypp]

So the CAN logger will give you Kwh left as opposed to percent left?
If it were that easy, people would know that their 85D really only has (apparently) 80Kwh of capacity.
Can you describe in more details exactly what you get from the CAN logger and how it is not extrapolating itself? as well as whether it gives you your real battery capacity when the car is 100% charged?

The CAN logger tells you how much expected usable capacity there is. It doesn't count the brick part however, but people assume it is 4kWh for 85kWh pack (this 4kWh was determined from some other part) and add it to it to get the full capacity (note: the 4kWh may not necessarily apply to 90kWh pack). You can find more details in wk057's thread (although it is quite long).

Yes, the BMS is also extrapolating itself, but it accounts for all energy demand (unlike the trip meter) and it doesn't have to rely on a linear assumption of SOC (it can view voltage directly and has historic battery data to go with). However, even the BMS is not that accurate and can drift (I remember seen an example of 1.5% variance in estimated kWh in a span of a week).

The only true way to measure battery capacity would be to charge fully (and balance), then fully discharge (going past the brick protection part) while measuring the amount of energy discharged. Obviously, this is not practical for a typical owner to do, so the CAN logger is the closest thing.

 

[amallah]

Sorry, most of your arguments do not apply. You are talking about range in miles, I'm talking about Kwh use.
Most of what you mentioned impacts range, but does not impact the Kwh used number and percentage of battery left, being farther away from reality

You are right, I am talking about range in miles, but so are you if you are using the trip meter. If you're truly just trying to measure energy usage, then you should also stick to pure energy measurements - either energy you had to "refill with" or apparently the CAN logger has this information.

Ref: https://www.evtripplanner.com/InterpretingTeslaModelSEnergyReadings.pdf

 

[mgboyes]

Yes, the BMS is also extrapolating itself, but it accounts for all energy demand (unlike the trip meter)

A lot of people seem to believe that the trip meter does not account for all energy demand. They're wrong. The trip meter measures all energy used while the "ignition" is on. It won't factor in overnight losses, or preheating, but once you're sitting in the car and travelling, it's counting everything.

Equally I agree that a CAN logger is the best way to measure the total energy used.

 

[stopcrazypp]

A lot of people seem to believe that the trip meter does not account for all energy demand. They're wrong. The trip meter measures all energy used while the "ignition" is on. It won't factor in overnight losses, or preheating, but once you're sitting in the car and travelling, it's counting everything.

Equally I agree that a CAN logger is the best way to measure the total energy used.

I haven't been following this closely, but this is what I frequently hear. whitex up thread says there is an extra draw that is not accounted for by the trip meter:
How many Kwh usable for driving are you getting on an 85 or 90 car? (I only get 75Kwh on my 90D)

There's been discussion about what is missing (vampire draw even when on, AC/heater usage when stationary but on, etc).

Of course, this should be fairly easy to test for someone with a CAN logger. If it accounts for everything, then the trip meter numbers should match exactly with difference in capacity found by the CAN logger.