Overall M3+ Efficiency w/ Detailed Data

[TomLee]

Hi All

First of all, I'm pretty new to the community as I only get my SR+ for about a month and half. After the delivery, I immediately noticed the overall efficiency was far below my expectation. I used to have a Honda CRV for the family. It was EPA rated for 352 miles or 560 km with a 15.3 gal or 58L tank. That roughly equals to 23MPG or 10L/100km. With most of the short city driving and some road trip, I achieved around 12.5L/100km which means I can drive 464 km with a full tank of gas. Let's say I got 464/560 = 83% of the average gas efficiency. I only needed to fill-up the tank once bi-weekly.

With 240 miles/386 km rated SR+, I was expecting to charge at most once a week giving 80% of the efficiency. However, I was disappointed as three charges were needed bi-weekly. I understood it's still far more cost efficient than the ICE car, but still, not up to my expectation. From the chart I have made for my SR+ below, my daily efficiency was anywhere from 38% to 72%, averaging 56%. I'm not sure if anybody get the same or achieve better efficiency.

Okay, talking about the data collection and other details:
1. Data collected before and on Jul 17th were from Teslafi. Since there were enormous discussion on how Teslafi contributes to the phantom drain, true or false, I turned off data polling from it on the 17th.
2. After 17th of July, I checked and only checked data from Tesla official App and Stats App twice a day, before and end of every day driving.
3. Outside temperature were from 22C/72F to 30C/86F during driving. Sometime parked outside, sometime indoor.
4. Max driving speed 110km/h or 70mph with around 30% of the time (highway driving) . But most of the time 55km/h or 35mph.
5. Wind was mild most of the time and it shouldn't affect the result.
6. Distance traveled up-hill roughly equals to the down-hill.

To the community, does these numbers seems normal to you? Is there any room for improvement and how?

 

[TomLee]

No comments?

 

[Diesel123]

What is your consumption on the actual computer in wh/mile? How far off of the “rated” are you when driving ?

Stock 18” aero wheels correct?

 

[TomLee]

What is your consumption on the actual computer in wh/mile? How far off of the “rated” are you when driving ?

Stock 18” aero wheels correct?

I remember the lifetime efficiency is around 150Wh/km or 240wh/mile on the screen. I don't see much difference between this and my recorded usage. I'm actually Okay with 80-90% of the driving efficiency. Most of the time I'm above that (faded line in my chart). What kills me was the sitting lost. Those drags the overall efficiency down to the toilet.

Yes, I'm using the stock 18" aero wheels with covers.

 

[AlanSubie4Life]

This is somewhat typical, but if you would like to improve it.

Try:

Turn off all cabin overheat protection.
Never use Sentry Mode
Unplug your TeslaCam drive
Disable all Apps (log out of Stats).
Gather all data by manually logging.
Do not access the car from the Tesla app unless you are about to begin a drive.

Then see how your results pan out. My last check of Stats unfortunately showed it did seem to increase vampire drain. (Though there may have been other factors at play - I was not able to eliminate all confounding variables.)


Not relevant, particularly, but:
To get the rated range, you need to see 215Wh/mi / 134Wh/km on the gauge in the car. You likely won’t be able to get this at 70mph of course.

 

[acarney]

I remember the lifetime efficiency is around 150Wh/km or 240wh/mile on the screen. I don't see much difference between this and my recorded usage. I'm actually Okay with 80-90% of the driving efficiency. Most of the time I'm above that (faded line in my chart). What kills me was the sitting lost. Those drags the overall efficiency down to the toilet.

Yes, I'm using the stock 18" aero wheels with covers.

That 240 wh/mile is about average with what ABetterRoutePlanner (it says 247 wh/mile @ 65mph, though it's slightly overestimating to make sure people aren't left stranded on their route) uses for the Model 3 SR+ data model for their range estimates, it also is a little bit better then some "real world" testing that some YouTube channels have done (and gathered crowed sourced data). On a full 50kWh battery that would mean roughly 208 miles of range instead of the rated 240 miles, or about 86% of what it's advertised as. That's not bad. This makes me think it's something else draining the battery instead of the actual driving efficiency

 

[darth_vad3r]

Not relevant, particularly, but:
To get the rated range, you need to see 215Wh/mi / 134Wh/km on the gauge in the car. You likely won’t be able to get this at 70mph of course.

215 is your Wh per displayed rated mile number for SR+? I did a longer trip recently and it was more like 125 +/-1 Wh/km (200-203 Wh/mi). Given the available info, my math seemed accurate, so, what’s with the big difference? BMS re-calibration mid trip? More errors from slow speed stop and go traffic for a portion? Or do you see something wrong with my method?

Numbers/math follow (do we have a thread for this going somewhere? ) ...

A) 120.05 - 121.65 dashboard km used.
B) 15.09 - 15.18 kWh energy used (in one leg that showed displayed range use of 121 +/-1 km, or 75 +/-1 mi).
C) Taking ranges A and B, calculate min/max consumption:

15.18 kWh / 120.05 km = 126.45 Wh/km = 203.5 Wh/mi max (actual rounds down to 203 with more precision)
15.09 kWh / 121.65 km = 124.04 Wh/km = 199.6 Wh/mi min (rounds to 200).
​

So my trip shows 200-203 Wh/mi for use of displayed miles (SR+).​

My Wh/mi of range added on the supercharger charge screen appears to be 219. For an LR I saw charging at a V3 on YouTube it appears to be 234.

Also 219 minus 7.76% is 202. 7.76% is a 4 kWh buffer if the pack is 51.5 kWh. That made me think my numbers were good.


Even more on methods...

A) I got the dashboard km used narrowed down to less than 1 mi of error (instead of 2 km or 2 mi) by comparing both different displayed units (km, mi) and taking mins of maxes, and maxes of mins, and converting the error ranges back to km.

B) Energy range (kWh) was calculated from a combination of the same trip displayed in both km and mi. I treat the km and mi ticks as accurate with a -0.05 adjustment, and use +/- 0.5 range on the consumption number, multiply all 4 variants, take the highest min, and the lowest max to get range of energy used. E.g. 130.2 km is treated as 130.15 (when the trip display ticks up and captures a snapshot). Since the trip in miles was over 0.1 km shorter, I added a conservative 15 W to the mile numbers to cover it being stale. My logic here is that the miles trip ticks up every 161m, so it’s going to be shorter than the trip in km more than 2/3 of the time. Best case is the mile trip is the more up to date one for 100m until the next km tick, worst case is they tick up simultaneously, and the average is that it’s winning for only 50m of the next 161m until it updates. In my case the km trip was 114m longer.​

 

[AlanSubie4Life]

215 is your Wh per displayed rated mile number for SR+? I did a longer trip recently and it was more like 125 +/-1 Wh/km (200-203 Wh/mi). Given the available info, my math seemed accurate, so, what’s with the big difference? BMS re-calibration mid trip? More errors from slow speed stop and go traffic for a portion? Or do you see something wrong with my method?

Numbers/math follow (do we have a thread for this going somewhere? ) ...

A) 120.05 - 121.65 dashboard km used.
B) 15.09 - 15.18 kWh energy used (in one leg that showed displayed range use of 121 +/-1 km, or 75 +/-1 mi).
C) Taking ranges A and B, calculate min/max consumption:

15.18 kWh / 120.05 km = 126.45 Wh/km = 203.5 Wh/mi max (actual rounds down to 203 with more precision)
15.09 kWh / 121.65 km = 124.04 Wh/km = 199.6 Wh/mi min (rounds to 200).
​

So my trip shows 200-203 Wh/mi for use of displayed miles (SR+).​

My Wh/mi of range added on the supercharger charge screen appears to be 219. For an LR I saw charging at a V3 on YouTube it appears to be 234.


Even more on methods...

A) I got the dashboard km used narrowed down to less than 1 mi of error (instead of 2 km or 2 mi) by comparing both different displayed units (km, mi) and taking mins of maxes, and maxes of mins, and converting the error ranges back to km.

B) Energy range (kWh) was calculated from a combination of the same trip displayed in both km and mi. I treat the km and mi ticks as accurate with a -0.05 adjustment, and use +/- 0.5 range on the consumption number, multiply all 4 variants, take the highest min, and the lowest max to get range of energy used. E.g. 130.2 km is treated as 130.15 (when the trip display ticks up and captures a snapshot). Since the trip in miles was over 0.1 km shorter, I added a conservative 15 W to the mile numbers to cover it being stale. My logic here is that the miles trip ticks up every 161m, so it’s going to be shorter than the trip in km more than 2/3 of the time. Best case is the mile trip is the more up to date one for 100m until the next km tick, worst case is they tick up simultaneously, and the average is that it’s winning for only 50m of the next 161m until it updates. In my case the km trip was 114m longer.​

I don't have an SR+, so I can only go based on video evidence that I've previously linked to here (I'm not going to dig it up now but it was a trip someone took from Colorado Springs to somewhere south of there on Youtube in an SR+). I can't explain why your number would be so far off (for all I know it could be correct, TBH). To me 200Wh/rmi seems far too low though. That would imply you'd only see 48kWh for a full discharge on the meter which seems too low with what we know about the 54kWh battery in the SR+ (see the EPA documents), even after accounting for meter errors and hidden reserves.

But, you're the one with the car. I would experiment again. Honestly I don't think you have to spend as much time as you do capturing as many significant figures as possible, but I don't really see anything wrong with it either, quickly reading through your methods.

I have heard that the meter does read low for very low consumption trips, which could be a partial explanation. I think @Zoomit referenced that post somewhere (though I don't think he was the one who said it). What was your displayed Wh/km (not Wh/rkm) on the trip meter for this test?

 

[jsmay311]

Questions:

1. How are you defining "Total Efficiency"?
2. What does the "AC Lost" part of the bar chart mean?
3. Do you have Cabin Overheat Protection or Sentry Mode on?
4. What type of charger (or outlet) are you using? (Volts? Amps?)

Looks like vampire drain decreased significantly after disabling TeslaFi on July 17. If that was indeed the root cause, you should look at the "Raw Data" page to try to figure out if your car was indeed failing to go to sleep, and if so, why. I've been able to use TeslaFi without increasing vampire drain much at all. Maybe having both TeslaFi AND Stats going at the same time causes problems(?).

 

[TomLee]

215 is your Wh per displayed rated mile number for SR+? I did a longer trip recently and it was more like 125 +/-1 Wh/km (200-203 Wh/mi). Given the available info, my math seemed accurate, so, what’s with the big difference? BMS re-calibration mid trip? More errors from slow speed stop and go traffic for a portion? Or do you see something wrong with my method?

Numbers/math follow (do we have a thread for this going somewhere? ) ...

A) 120.05 - 121.65 dashboard km used.
B) 15.09 - 15.18 kWh energy used (in one leg that showed displayed range use of 121 +/-1 km, or 75 +/-1 mi).
C) Taking ranges A and B, calculate min/max consumption:

15.18 kWh / 120.05 km = 126.45 Wh/km = 203.5 Wh/mi max (actual rounds down to 203 with more precision)
15.09 kWh / 121.65 km = 124.04 Wh/km = 199.6 Wh/mi min (rounds to 200).
​

So my trip shows 200-203 Wh/mi for use of displayed miles (SR+).​

My Wh/mi of range added on the supercharger charge screen appears to be 219. For an LR I saw charging at a V3 on YouTube it appears to be 234.

Also 219 minus 7.76% is 202. 7.76% is a 4 kWh buffer if the pack is 51.5 kWh. That made me think my numbers were good.


Even more on methods...

A) I got the dashboard km used narrowed down to less than 1 mi of error (instead of 2 km or 2 mi) by comparing both different displayed units (km, mi) and taking mins of maxes, and maxes of mins, and converting the error ranges back to km.

B) Energy range (kWh) was calculated from a combination of the same trip displayed in both km and mi. I treat the km and mi ticks as accurate with a -0.05 adjustment, and use +/- 0.5 range on the consumption number, multiply all 4 variants, take the highest min, and the lowest max to get range of energy used. E.g. 130.2 km is treated as 130.15 (when the trip display ticks up and captures a snapshot). Since the trip in miles was over 0.1 km shorter, I added a conservative 15 W to the mile numbers to cover it being stale. My logic here is that the miles trip ticks up every 161m, so it’s going to be shorter than the trip in km more than 2/3 of the time. Best case is the mile trip is the more up to date one for 100m until the next km tick, worst case is they tick up simultaneously, and the average is that it’s winning for only 50m of the next 161m until it updates. In my case the km trip was 114m longer.​

Wow, you really pay attention to the details! I have thought about the numbers behind the decimal but never mind doing anything. That 200Wh/mile is not somthing I can easily achieve as well.

I like details as well, but I'm more concerned about the overall efficiency: total kms traveled/total kms added. For the period of 19 days I recorded, I traveled 524km and added 930km worth of electricity in. That's about the same as what I calculated from each day: 55% average. I'm wondering if you have such data and what is your overall percentage?

 

[AlanSubie4Life]

Wow, you really pay attention to the details! I have thought about the numbers behind the decimal but never mind doing anything. That 200Wh/mile is not somthing I can easily achieve as well.

I like details as well, but I'm more concerned about the overall efficiency: total kms traveled/total kms added. For the period of 19 days I recorded, I traveled 524km and added 930km worth of electricity in. That's about the same as what I calculated from each day: 55% average. I'm wondering if you have such data and what is your overall percentage?

As mentioned above, for your issues, you need to minimize your losses when not driving. For that, try some of the suggestions above. I definitely think you should completely disable access by all apps. And you DEFINITELY can't use Cabin Overheat protection or Sentry Mode or TeslaCam (Teslacam seems to increase vampire drain in a couple of limited datapoints I have).

What @darth_vad3r is calculating is actually the Wh/rkm (rated km) and Wh/rmi (rated mile).

 

[TomLee]

Questions:

1. How are you defining "Total Efficiency"?
2. What does the "AC Lost" part of the bar chart mean?
3. Do you have Cabin Overheat Protection or Sentry Mode on?
4. What type of charger (or outlet) are you using? (Volts? Amps?)

Looks like vampire drain decreased significantly after disabling TeslaFi on July 17. If that was indeed the root cause, you should look at the "Raw Data" page to try to figure out if your car was indeed failing to go to sleep, and if so, why. I've been able to use TeslaFi without increasing vampire drain much at all. Maybe having both TeslaFi AND Stats going at the same time causes problems(?).

Thanks for asking the detail questions! I prepared, but no one asked
1. Total efficiency: total kms traveled/total kms of range being used. For me, it's around 55% so far.
2. AC Lost part was copied from Teslafi "conditioning section when available. Otherwise it was the difference between total range lost and the range lost from driving, parking and sleeping. Sleeping Lost was copied from Teslafi 'Sleep' section when available. Otherwise it was the one day's start range reading and previous' day end range reading. Parked range was copied from Teslafi 'Parked' section when available. Otherwise it was the readings from Phantom Drain Section of Stats App.
3. I don't have Cabin Overheat on, but Sentry was on. I can turn sentry off if on road trip, but for now, I want to know how much of impact it has.
4. I've been using 120V and 240V 12A~24A AC charger several times at home, office and mall. I charged at supercharge station some time especially on road trip as well.

Please enlight me where do you see the drain decreased from the data? if you are saying the parked range lost is significantly lower after 17th, I agree. But it could also due to the data collection method change. That number was not available in Stats app. The only thing comparable was the phantom drain section. So some numbers could shift to AC section which significantly increased. Otherwise, I see average daily lost of 13.1km before 17th and 15.3km after.

 

[TomLee]

As mentioned above, for your issues, you need to minimize your losses when not driving. For that, try some of the suggestions above. I definitely think you should completely disable access by all apps. And you DEFINITELY can't use Cabin Overheat protection or Sentry Mode or TeslaCam (Teslacam seems to increase vampire drain in a couple of limited datapoints I have).

What @darth_vad3r is calculating is actually the Wh/rkm (rated km) and Wh/rmi (rated mile).

Thanks for the suggetions. That will be my next step.

 

[AlanSubie4Life]

but for now, I want to know how much of impact it has.

It's huge. About 1 mile per hour.

I really think you need to completely eliminate apps. It's kind of like Heisenberg's Uncertainty Principle; if you watch the car too much, you affect the result. The easiest thing would be to log your rated miles before each charge, and after each charge (just keep a running total of how many you add). And then keep track of how many miles you travel.

Remember that your overall energy consumption (to the extent you care about it - does not appear to be your primary concern) will be about 10-15% higher than that result due to charging inefficiency & the error in the in-car meter.

That number was not available in Stats app. The only thing comparable was the phantom drain section.

I wouldn't trust any of these apps for their data collection. There are various issues with their methodology - and they can miss losses as well. Best thing is to just keep track of only the fundamental numbers yourself. It won't give you all the bells & whistles you want, but better than nothing - it will tell you your "total efficiency" as you define it.

 

[darth_vad3r]

I don't have an SR+, so I can only go based on video evidence that I've previously linked to here (I'm not going to dig it up now but it was a trip someone took from Colorado Springs to somewhere south of there on Youtube in an SR+).

Yes I remember your post on it, I had just forgotten what the number you calculated from it was, that's what I meant by "your number" (also didn't feel like diggint it up )

I can't explain why your number would be so far off (for all I know it could be correct, TBH). To me 200Wh/rmi seems far too low though. That would imply you'd only see 48kWh for a full discharge on the meter which seems too low with what we know about the 54kWh battery in the SR+ (see the EPA documents), even after accounting for meter errors and hidden reserves.

But, you're the one with the car. I would experiment again.

Oh I will I just don't always get the opportunity to do it on a longer trip

Honestly I don't think you have to spend as much time as you do capturing as many significant figures as possible, but I don't really see anything wrong with it either, quickly reading through your methods.

I definitely don't have to spend as much time as I do ... but I do

I have heard that the meter does read low for very low consumption trips, which could be a partial explanation. I think @Zoomit referenced that post somewhere (though I don't think he was the one who said it).

Yes, I remember this as well, I think he was quoting wk057 most likely. It was my recollection of this that made me surmise maybe my low number may be based on a bit of stop-and-go we hit on this trip.

What was your displayed Wh/km (not Wh/rkm) on the trip meter for this test?

My displayed trip consumption on the "since h:mm" trip for the one leg was 130 Wh/km (and 209 Wh/mi on the ever-so-slightly shorter version of the same trip displayed in miles instead).

So the km and miles both ticked down on the display MORE than my actual distance driven because I wasn't low enough at 209 Wh/mi based on my test (required 202 ~Wh/mi).

Numbers of note for my car:

1. 139.x Wh/km -- The rated line in the energy chart is at 139-140 Wh/km (both numbers can show solid line overlap or dotted line below or above).
2. 136.1 Wh/km - The rate at which km get added on the supercharger screen based on mapping kW power to km/hr charge speed
3. 125 +/- 1 Wh/km - The rate at which displayed dashboard km ticked down for my 1 trip experiment

 

[AlanSubie4Life]

Yes I remember your post on it, I had just forgotten what the number you calculated from it was, that's what I meant by "your number" (also didn't feel like diggint it up )

Obviously there are ways in which information could have been lost which would make the calculation from this video inaccurate. But here is the video I used for my calculations.

Tesla Model 3 Standard Range Plus Real-World Highway Range Test

 

[darth_vad3r]

Obviously there are ways in which information could have been lost which would make the calculation from this video inaccurate. But here is the video I used for my calculations.

Tesla Model 3 Standard Range Plus Real-World Highway Range Test

Ah, ya that one ... if only he'd changed the display to miles instead of %. Ya, I'm not sure how back-calculating the miles from the percent loss can impact things, especially given calibration issues and how much things can fluctuate. It went 28%, 27%, 28% while he was wrapping up his video. He started at 95%, it might have ticked up to 96% as he pulled away.

How do your numbers for your car work out? What's your rated line at in the energy screen, what's your charge Wh/mi rate, and what's your Wh/dashboard miles?

 

[AlanSubie4Life]

How do your numbers for your car work out? What's your rated line at in the energy screen, what's your charge Wh/mi rate, and what's your Wh/dashboard mile

1) Never pay attention to the energy screen line. Looks near 250Wh/mi when I last checked 6 months ago.

2) Never supercharge.

3) Wh/rmi is 230Wh/mi for P3D and AWD every single time I check it.

 

[bwilson4web]

With 240 miles/386 km rated SR+, I was expecting to charge at most once a week giving 80% of the efficiency. However, I was disappointed as three charges were needed bi-weekly.

Just trying to understand:

• "giving 80% of the efficiency" - is this the charge limit set on the battery? Normally I use efficiency to be "kWh_out / kWh_in". For a battery, efficiency would be the kWh consumption over the charging kWh.
  • If there is a battery charge limit set, what is the value? The reason is charging over 85-90% reduces the 'head room' needed to save regeneration. It leads to using the friction brakes to waste energy instead of maximizing the stopping energy put in the battery.
• "three charges were needed bi-weekly" - Is this 3 charges every 2 weeks or 3 charges, 2 times each week? The term "bi-weekly" can be ambiguous. In the first version, it averages 1.5 charges each week and the second case, 6 times each week. There is a difference.

Regardless, here is a graph showing my SR+M3 vehicle energy consumption as a function of speed and the expected range using 90% and 100% battery charge:

Knowing the speed is critical to making claims about kWh/mi or the inverse, mi/kWh. Use 1.6 * mi = km.

Bob Wilson

 

[jsmay311]

1. Total efficiency: total kms traveled/total kms of range being used. For me, it's around 55% so far.
2. AC Lost part was copied from Teslafi "conditioning section when available. Otherwise it was the difference between total range lost and the range lost from driving, parking and sleeping. Sleeping Lost was copied from Teslafi 'Sleep' section when available. Otherwise it was the one day's start range reading and previous' day end range reading. Parked range was copied from Teslafi 'Parked' section when available. Otherwise it was the readings from Phantom Drain Section of Stats App.

Oh. I thought by "AC" you were talking about A/C electric power, and "AC lost" referred to charging losses.

So are you not accounting for charging losses at all in your "Total efficiency" number? Yikes. I think that means you're underestimating (possibly by quite a bit) how bad your overall efficiency is.

3. I don't have Cabin Overheat on, but Sentry was on.

As mentioned above by AlanSubie4Life, this will have a big drain on your overall efficiency. Stop it.

4. I've been using 120V and 240V 12A~24A AC charger several times at home, office and mall. I charged at supercharge station some time especially on road trip as well.

OK. I asked because 240V charging is more efficient than 120V charging.

Please enlight me where do you see the drain decreased from the data? if you are saying the parked range lost is significantly lower after 17th, I agree.

Yes, that's what I was saying.