Calculate usable battery capacity based on rated miles values

[ljwobker]

Super informative thread. Thank you wk057 and others for careful analysis and testing - it's appreciated. I was curious to see if I could replicate some of the data so I ran the following (somewhat natural) experiment. Full range charge on a April 2014 S85, the battery was "full" for at least 3-4 hours before I started the trip. Total trip duration was about 28 hours, ambient temps ranged between 65-85F. No use of mobile app or anything else that I'm aware of during the duration, and no charging between when I left and returned home.

Trip meter showed zero miles remaining when I got home - it clicked over to zero about 500 yards from my house so it's not "deep" under zero by any meaningful amount. What I can't quite figure out is whether the kWh consumed since the last charge is wrong, or if there are things I'm not counting on here. If the full battery USABLE capacity is roughly 77.5kwh, did I really suffer almost 9kWh of various vampire drain in a little more than 27 hours? 9kwh in 27 hours would be a **constant** vampire drain of ~330watts, which seems awfully high. Anyone have data on this? Any info I'm missing here?

 

[David99]

If the full battery USABLE capacity is roughly 77.5kwh, did I really suffer almost 9kWh of various vampire drain in a little more than 27 hours? 9kwh in 27 hours would be a **constant** vampire drain of ~330watts, which seems awfully high. Anyone have data on this? Any info I'm missing here?

The trip meter in the car only runs and counts when the car is in motion. Any time you are stopped (even when the car is on) or parked the counter does not run even though energy is used. 300 Watt continuous use seems kind of high. I think mine uses about 180 Watt on average just parked. But then again, if during your trip you have some time sitting in the car running the AC, that might explain the higher energy usage not showing up.

 

[ljwobker]

180W parked??? -- even that seems really high. In my case the car wasn't touched other than when I was driving... are you saying that if I'm stopped for ANY reason -- even just at a traffic light -- the meter stops running? That seems really odd but I guess I can believe it... but in this particular case that whole trip is almost entirely devoid of stoplights/stop signs.. I would have been in the car moving for just over 3 hours, and there's no way there is more than 15 minutes of total zero-motion time in that trip. So I'm still missing a LOT of power from somewhere....

 

[AWDtsla]

180W parked??? -- even that seems really high.

Easily exceeded. Of course the car does usually sleep, which reduces the average draw, but there will be long segments with higher draw.

In my case the car wasn't touched other than when I was driving... are you saying that if I'm stopped for ANY reason -- even just at a traffic light -- the meter stops running? That seems really odd but I guess I can believe it... but in this particular case that whole trip is almost entirely devoid of stoplights/stop signs.. I would have been in the car moving for just over 3 hours, and there's no way there is more than 15 minutes of total zero-motion time in that trip. So I'm still missing a LOT of power from somewhere....

I believe it's still metering as long as you stay in drive. However I agree you're missing a lot of power. Just like every Tesla owner.

 

[ran349]

Super informative thread. Thank you wk057 and others for careful analysis and testing - it's appreciated. I was curious to see if I could replicate some of the data so I ran the following (somewhat natural) experiment. Full range charge on a April 2014 S85, the battery was "full" for at least 3-4 hours before I started the trip. Total trip duration was about 28 hours, ambient temps ranged between 65-85F. No use of mobile app or anything else that I'm aware of during the duration, and no charging between when I left and returned home.

Trip meter showed zero miles remaining when I got home - it clicked over to zero about 500 yards from my house so it's not "deep" under zero by any meaningful amount. What I can't quite figure out is whether the kWh consumed since the last charge is wrong, or if there are things I'm not counting on here. If the full battery USABLE capacity is roughly 77.5kwh, did I really suffer almost 9kWh of various vampire drain in a little more than 27 hours? 9kwh in 27 hours would be a **constant** vampire drain of ~330watts, which seems awfully high. Anyone have data on this? Any info I'm missing here?


View attachment 245715

Thanks for that data. How many rated miles did you have right after you charged to 100%, and then 3-4 hours later when you started the trip? Just so I understand, for your trip your actual driving time was roughly 4 hours, and the rest of the time the car was parked, so roughly 24 hours of parked time?

 

[ljwobker]

I saw 254 and 255 miles when the battery was "full". I don't have exact timestamps on those numbers but the "charging" light on my home charger had been off for at least a couple of hours before I departed. For the trip the actual driving time would have been a little less than 4 hours, and the car would have been parked (unoccupied, no app usage....) for about 24.

 

[ljwobker]

argh. 99% sure it was the cabin overheat protection kicking in. The car spent Sunday parked in the sun and would certainly have gotten hotter than 105F at some point. I can easily convince myself that the AC ran enough to gobble up that much energy...

 

[EV-Tech Exp]

argh. 99% sure it was the cabin overheat protection kicking in. The car spent Sunday parked in the sun and would certainly have gotten hotter than 105F at some point. I can easily convince myself that the AC ran enough to gobble up that much energy...

That does seem the most likely way in which that huge amount of energy will have been consumed.

 

[Texas]

I wonder if wk079 or one of the other technically accomplished members could advise me whether I should consider my 100 percent Rated Range to be 252 miles or 231 miles, given the following facts:

The car is a May 2015 P85DL that now has 15,600 miles on the odometer. When new, the 100 percent range indicated by the car was 253 miles. Currently it reads 252 miles. I tested it recently on a trip from Houston to Galveston, a distance of 153 miles with side trips. The computer recorded 308 Wh/mile for the round trip, not far from the EPA testing value of 310 Wh/mile.

At the end of the 153 miles, the battery percent remaining was 31 percent, which would be 78 miles, for a total 100 percent range of 231 miles, not 252 miles. I would have expected 39 percent remaining, or 98 miles, because my distance of 153 miles was 61 percent of the 252 miles 100 percent range.

On returning to Houston I recharged the car back to 100 percent, which consumed 47.1 kWh. The 100 percent reading was again 252 miles.

For whatever relevance it may have, I did not use Range Mode at any time in the trip, nor did I do any power launches. There were two people and a dog in the car and about 100 lbs. of miscellaneous gear. The car air conditioning was on for the whole trip, and outside temperatures were in the 88 to 90 degree range. We also had the radio system and daylight running lights on. The tires are the OEM 19” Michelins on Cyclone wheels inflated to 45 lbs. Wheels are in good alignment.

The car was parked, not plugged in, for most of two days in Galveston. Energy Saving is “On” but “always connected,” so this probably accounts for an unrecorded loss of 4 or more miles.

The only explanations I can come up for such a large discrepancy are (1) that the air conditioning is not included in the accounting of Rated Range, or (2) that my 100 percent range reading is very inaccurate. Neither seems plausible to me.

I would tend to rule out battery degradation as the cause. When the car was new, with 3500 miles on the odometer, I recorded a round trip in which I achieved only 220 miles with an average electric usage of 295 Wh/mile. Converting this to 310 Wh/mile and extrapolating another 33 miles to reach 253 miles of Rated Range yields the exact same 231 mile 100 percent Rated Range.

Does anyone have a good suggestion as to why my actual 100 percent range readings are so low as compared to the 253 and 252 mile 100 percent range reported by the car’s computer?

 

[jerry33]

Does anyone have a good suggestion as to why my actual 100 percent range readings are so low as compared to the 253 and 252 mile 100 percent range reported by the car’s computer?

Rated range is basically EPA miles. Drive like the EPA test and you will get rated miles. No A/C, heating, or weather is taken into consideration. Battery use when parked is not considered either. I use Ideal miles to get the battery SOC because they have changed the Rated range algorithm several time, but have not changed Ideal miles at all (as far as I can tell) so it's been more consistent over the years (In town, Ideal miles is very close to actual).

 

[AWDtsla]

I wonder if wk079 or one of the other technically accomplished members could advise me whether I should consider my 100 percent Rated Range to be 252 miles or 231 miles, given the following facts:

The car is a May 2015 P85DL that now has 15,600 miles on the odometer. When new, the 100 percent range indicated by the car was 253 miles. Currently it reads 252 miles. I tested it recently on a trip from Houston to Galveston, a distance of 153 miles with side trips. The computer recorded 308 Wh/mile for the round trip, not far from the EPA testing value of 310 Wh/mile.

At the end of the 153 miles, the battery percent remaining was 31 percent, which would be 78 miles, for a total 100 percent range of 231 miles, not 252 miles. I would have expected 39 percent remaining, or 98 miles, because my distance of 153 miles was 61 percent of the 252 miles 100 percent range.

On returning to Houston I recharged the car back to 100 percent, which consumed 47.1 kWh. The 100 percent reading was again 252 miles.

For whatever relevance it may have, I did not use Range Mode at any time in the trip, nor did I do any power launches. There were two people and a dog in the car and about 100 lbs. of miscellaneous gear. The car air conditioning was on for the whole trip, and outside temperatures were in the 88 to 90 degree range. We also had the radio system and daylight running lights on. The tires are the OEM 19” Michelins on Cyclone wheels inflated to 45 lbs. Wheels are in good alignment.

The car was parked, not plugged in, for most of two days in Galveston. Energy Saving is “On” but “always connected,” so this probably accounts for an unrecorded loss of 4 or more miles.

The only explanations I can come up for such a large discrepancy are (1) that the air conditioning is not included in the accounting of Rated Range, or (2) that my 100 percent range reading is very inaccurate. Neither seems plausible to me.

I would tend to rule out battery degradation as the cause. When the car was new, with 3500 miles on the odometer, I recorded a round trip in which I achieved only 220 miles with an average electric usage of 295 Wh/mile. Converting this to 310 Wh/mile and extrapolating another 33 miles to reach 253 miles of Rated Range yields the exact same 231 mile 100 percent Rated Range.

Does anyone have a good suggestion as to why my actual 100 percent range readings are so low as compared to the 253 and 252 mile 100 percent range reported by the car’s computer?

From experience with my car, I assume your real consumption was 352Wh/mi while displaying 308Wh/mi. But there could be more going on here.

On returning to Houston I recharged the car back to 100 percent, which consumed 47.1 kWh. The 100 percent reading was again 252 miles.

Measured at the wall? If so then it's more like the car is now considering 47.1*.9/173 = 240 Wh/mi to be a rated mile, which seems ridiculous. Do you have logs of what the car reported as added energy?

 

[Texas]

Jerry33, the EPA changed their rules. They now include heating and air conditioning elements in their testing requirements, https://www.fueleconomy.gov/feg/pdfs/EPA test procedure for EVs-PHEVs-11-14-2017.pdf .I haven't tried to interpret the requirements.

AWDtsla, you raise an interesting measurement point. The 47.1 kWh is what was recorded by the car and reflected on the left hand side of the instrument panel at the end of the trip. This entry was under the second line, "since last charge." It was also reported by my RemoteS app. I don't have any metering capability at the wall, just the HPWC. The wall charger runs at 80 amps through dual chargers in the car. I am told there is about a 10 percent inefficiency in this process, which is not reflected in the kWh hours I reported.

Mentioning power at the wall rather than in the car is of interest because the EPA measures electricity at the wall, not in the car, something I didn't know. The EPA says that "the battery is recharged from a normal AC source and the energy consumption of the vehicle is determined (in kW-hr/mile or kW-hr/100 miles) by dividing the kilowatt-hours of energy to recharge the battery by the miles traveled by the vehicle. The recharge energy includes any losses due to inefficiencies of the manufacturer’s charger" (emphasis added). Until reading your comment I didn't consider the possibility that the EPA range numbers were based on electric usage inclusive of charging losses. I'll have to go back and do some figuring to see how to factor this in to my numbers.

Also, please set forth the steps you would follow to arrive at 352 Wh/mile instead of the 308 Wh/mile the car's instruments recorded.

 

[AWDtsla]

Also, please set forth the steps you would follow to arrive at 352 Wh/mile instead of the 308 Wh/mile the car's instruments recorded.

Your car should be using 310Wh per rated mile. So (252-78)=173 rated miles consumed would be 173*.310=53.63kW and 53630/153=350Wh/mi. Now of course, you also charged 173 miles with only 47.1kW which is where things get weird. If you have logs of the charge session, I'd check if there was a jump somewhere in rated miles, indicated the BMS recalibrating. You should normally see the 310Wh per rate mile ratio reflected while charging, if it's not recalibrating. Just divide energy added by miles added.

I wonder however if maybe this is how Tesla is dealing with battery degradation. You can be assured that your degradation wasn't 1 mile in 3 years and 15k miles. Given the degradation isn't reflected in 100% charge, the miles magically disappear while discharging, and reappear again after charged?

 

[hiroshiy]

I wonder if wk079 or one of the other technically accomplished members could advise me whether I should consider my 100 percent Rated Range to be 252 miles or 231 miles, given the following facts:

The car is a May 2015 P85DL that now has 15,600 miles on the odometer. When new, the 100 percent range indicated by the car was 253 miles. Currently it reads 252 miles. I tested it recently on a trip from Houston to Galveston, a distance of 153 miles with side trips. The computer recorded 308 Wh/mile for the round trip, not far from the EPA testing value of 310 Wh/mile.

At the end of the 153 miles, the battery percent remaining was 31 percent, which would be 78 miles, for a total 100 percent range of 231 miles, not 252 miles. I would have expected 39 percent remaining, or 98 miles, because my distance of 153 miles was 61 percent of the 252 miles 100 percent range.

On returning to Houston I recharged the car back to 100 percent, which consumed 47.1 kWh. The 100 percent reading was again 252 miles.

For whatever relevance it may have, I did not use Range Mode at any time in the trip, nor did I do any power launches. There were two people and a dog in the car and about 100 lbs. of miscellaneous gear. The car air conditioning was on for the whole trip, and outside temperatures were in the 88 to 90 degree range. We also had the radio system and daylight running lights on. The tires are the OEM 19” Michelins on Cyclone wheels inflated to 45 lbs. Wheels are in good alignment.

The car was parked, not plugged in, for most of two days in Galveston. Energy Saving is “On” but “always connected,” so this probably accounts for an unrecorded loss of 4 or more miles.

The only explanations I can come up for such a large discrepancy are (1) that the air conditioning is not included in the accounting of Rated Range, or (2) that my 100 percent range reading is very inaccurate. Neither seems plausible to me.

I would tend to rule out battery degradation as the cause. When the car was new, with 3500 miles on the odometer, I recorded a round trip in which I achieved only 220 miles with an average electric usage of 295 Wh/mile. Converting this to 310 Wh/mile and extrapolating another 33 miles to reach 253 miles of Rated Range yields the exact same 231 mile 100 percent Rated Range.

Does anyone have a good suggestion as to why my actual 100 percent range readings are so low as compared to the 253 and 252 mile 100 percent range reported by the car’s computer?

Just to confirm, 153 miles is your whole round trip distance and
252 - (153 x (310 / 308)) = 98km of rated range you expected to have at the end of the round trip
Actually there's only 78 miles of rated range left

So if above assumption is correct, there was a discrepancy of about 20 rated miles, or 310 x 20 / 1000 = 6.2kWh of energy missing. AC usage while stopped (even at traffic signals) is not calculated in the dash display so 6.2kWh contains those and any standby electricity usage while parked. It is safe to say Tesla with power save and always connected consume 1kWh per day.

So 6.2 - 2(days) - 2kWh for AC stopped (because you did some side trips) is 2.2kWh.

You might have smart air conditioning on or cabin temp protection on. In such hot climate it would easily consume 2.2kWh. Also did you keep your fob close so that the car was unlocked, or use your Tesla app frequently (like a few times a day)? If so that also might consume 2.2kWh for two days.

Anyway I think your car's battery calculation seems not be far off.

 

[AWDtsla]

So 6.2 - 2(days) - 2kWh for AC stopped (because you did some side trips) is 2.2kWh.

You might have smart air conditioning on or cabin temp protection on. In such hot climate it would easily consume 2.2kWh. Also did you keep your fob close so that the car was unlocked, or use your Tesla app frequently (like a few times a day)? If so that also might consume 2.2kWh for two days.

Anyway I think your car's battery calculation seems not be far off.

Attributing the discrepancy to uncounted usage only makes sense if you disregard charge back to 100% with only 47.1kW. I'm sure the former happened, which makes the latter even more strange.

 

[Texas]

AWDtsla, you deserve a gold star. I went back and more carefully reviewed the notes I kept on my iPhone. I found that the 47.1 kWh was not the amount recorded when charging. It was the amount shown on the instrument panel, as were other important numbers.

So I have to correct my numbers to the values at the time of charging. The most significant change is the energy used: 54.3 kWh, not 47.1 kWh. The miles remaining also fell from 78 to 76, raising the mileage from 153 to 155. This changed due to losses accumulated while the car was parked in Houston waiting for electric rates to fall at night. This additional kilowatt loss probably resulted in part from being parked in the sun long enough for the Cabin Overheat Protection to kick in. The time between charging, I also realize, was three days, not two. Other unrecorded losses, in addition to normal traffic light stops, probably can be attributed to 30-40 minutes of stop and go freeway traffic on TACC, a small part of which was stopped. 54.3 kWh/155 = 350.3 Wh/mile. This is close to your predicted 352 Wh/mile.

At 350.3 Wh/mile, 155 miles expends 54.3 kWh, leaving (77.5-54.3) = 23.2 kWh of battery capacity remaining., or 75 miles at 310 Wh/mile, if you assume the battery is good for 77.5 kWh. This is very close to the metered 76 remaining miles recorded as noted above. The total range resulting from all of this accounting--an adjusted rated range--is 231 miles (155+76=231 miles).

Unless someone can punch a hole in this analysis, I believe this is the correct approach, although the accounting for unrecorded kilowatts (which I suggest should be called Phantom Kilowatts) in this instance is approximate. We have identified 7.2 Phantom Kilowatts (53.4 kWh -47.1 kWh) which would translate into 23 miles of range at 310 Wh/mile. These Phantom Kilowatts probably can be attributed to (1) three days of being parked, (2) one afternoon in which Cabin Overheat Protection was operating, (3) somewhat higher than average time stopped due to freeway traffic. If these 7.2 Phantom Kilowatts were used to propel the car rather than being expended in standby functions, it would have added 23 miles to the 231 miles.

Converting Phantom Kilowatts into mileage would result in a Rated Range of 254 miles, roughly the same value as the expected 252 mile Rated Range.

Given the difficulty of this bit of accounting, I would say that anyone who wants to accurately measure the Rated Range of his Tesla needs to take as much care as possible to avoid downtime in which Phantom Kilowatts can accumulate.

For those not familiar with the relevant principle, the Tesla system apparently accounts for energy used that reflects the testing done by the EPA. The EPA testing does not include energy used by the car while parked or stopped, which can be especially significant when the car is stopped with the air conditioning or resistance heating operating. These exclusions are what I refer to as “Phantom Kilowatts”. In the case above, the car recorded only 47.1 kWh used, but when I re-charged the battery, it took 54.3 kWh, not 47.1 kWh; the 7.2 kWh difference represents Phantom Kilowatts—i.e., power that did not play a part in propelling the car.

I should also acknowledge that Hiroshiy's comments were also useful in pointing me in the right direction.

 

[Texas]

Not sure what I can add but my experience is the discrepancy between the 'last charge' measurements and the rated miles as per the % and miles left (my chart line is 305Wh/mi an S75 in the thread should be 285Wh/mi)

Having run from genuine 100% via the last charge maths I used 53.2kW for 192 miles but when I re-charged back to 100% I had put back 62kW, mileage left was 28 based on EPA. Trip was 273kW/hr.

So a kW difference of 8.8kW from the two measurements?

I had a similar question when I compared two 100 percent readings in a P85D with a 252 mile Rated Range and a Rating of 310 Wh/mile. My difference was 7.2 kWh.

I was able to resolve this problem as shown in the above post dated May 19. The key analytical step is to identify energy used by the car for purposes not counted when the EPA rates the car. such as daily "vampire losses" and other uses such as air conditioning operation when the car is not in motion. Identify these "Phantom Kilowatts" and you should be able to resolve some otherwise inexplicable data readings.

 

[Rocky_H]

Jerry33, the EPA changed their rules. They now include heating and air conditioning elements in their testing requirements, https://www.fueleconomy.gov/feg/pdfs/EPA test procedure for EVs-PHEVs-11-14-2017.pdf .I haven't tried to interpret the requirements.

Can you provide real documentation to this? That link you gave is a two page document of their testing procedures, which I read thoroughly, and it makes no mention at all of running any kind of heating or cooling, so I remain skeptical.

 

[Texas]

You’re right, I provided the wrong internet citation. The correct announcement by the EPA adding air conditioning and heating to vehicle testing is found at: https://www.gpo.gov/fdsys/pkg/FR-2006-12-27/pdf/06-9749.pdf, at page 77872. Specifically, the agency announced that:

“The new test methods take into account several important factors that affect fuel economy in the real world, but are missing from the existing fuel economy tests. Key among these factors are high speeds, aggressive accelerations and decelerations, the use of air conditioning, and operation in cold temperatures.”

The new methods became applicable in 2007 for 2008 models. There is much more to the rule, but I have not attempted to analyze it in detail.

 

[appleguru]

Apologies for digging up this old thread, but this seems to have the most relevant info in it... Did we ever resolve the Usable state of energy vs state of charge question?

I checked my S75D today and noticed that when I was showing less than 1% battery left (3mi of range on the dash), that I still had an SoC of over 8%.

Is that all anti-bricking buffer? Energy that was “missing” due to the pack not getting this low in a while? Energy not availible due to temperature or other factors? (It was hot today, around 40C ambient). Other? Seems like a lot of buffer!