Tesla's 85 kWh rating needs an asterisk (up to 81 kWh, with up to ~77 kWh usable)

[DavidB]

Documents filed by Tesla containing "official data sheets" are, in my humble opinion, worthless. I want real data collected by (disinterested) 3rd parties. I'm not interested in reading mumbo jumbo that has no actual data that backs it up.

wk057 has, multiple times, called you out, saying, "put up or shut up", and you still post the same false data.

 

[vgrinshpun]

Documents filed by Tesla containing "official data sheets" are, in my humble opinion, worthless. I want real data collected by (disinterested) 3rd parties. I'm not interested in reading mumbo jumbo that has no actual data that backs it up.

wk057 has, multiple times, called you out, saying, "put up or shut up", and you still post the same false data.

This IS actual testing data. Duplicated/approved by TWO third parties: CARB and EPA.

What is your basis for calling it false data? It is most certainly not.

EDIT: I did not post this particular data before.

 

[DavidB]

This IS actual testing data. Duplicated/approved by TWO third parties: CARB and EPA.

You think Tesla always posts actual testing data? That's so cute. Naive, but cute. You think CARB or EPA actually duplicates the testing? Again, how cute.

 

[vgrinshpun]

You think Tesla always posts actual testing data? That's so cute. Naive, but cute. You think CARB or EPA actually duplicates the testing? Again, how cute.

As I've shown the data submitted by Tesla was duplicated by EPA in the Volkswagen test lab. What is cute is to refuse to acknowledge reality. But I forget, you actually do not read any REAL data.

And by the way, if you are so averse to anything coming from Tesla, did you mark "disagree" on @wk057 post attaching outdated Tesla descriptive materials, in the post I was replying to?

 

[shred86]

Documents filed by Tesla containing "official data sheets" are, in my humble opinion, worthless. I want real data collected by (disinterested) 3rd parties. I'm not interested in reading mumbo jumbo that has no actual data that backs it up.

wk057 has, multiple times, called you out, saying, "put up or shut up", and you still post the same false data.

To be fair, vgrinshpun asked why test results suggest a higher usable battery capacity than what wk057 shared. The reasoning was that "EPA numbers are BS" and the follow up post suggested it's because the data vgrinshpun was derived from power going into the charger, to which he showed it includes both (into the charger and actually depleted). Then it was suggested all of the values are calculated, not measured, which is where there seems to be the biggest disagreement and no one actually has factual data that shows if those test numbers are in fact calculated or measured*. When asked to show a Model 3 that has gotten anywhere close to 78 kWh, there is a video of one going to 75 kWh but that test was apparently flawed, which sadly I can't figure out why, lol. I've always thought the trip meter showed total energy used while in motion, so if anything it would always show less than actual kWh usage. Anyways, just trying my best to follow along but can some with a brain please explain to me how Ben Sullin's test was flawed.

* Edit: See post #1167. Seems to be pretty good evidence to suggest the tests were run on a four wheel dyno and not calculated.

 

[vgrinshpun]

To be fair, vgrinshpun asked why test results suggest a higher usable battery capacity than what wk057 shared. The reasoning was that "EPA numbers are BS" and the follow up post suggested it's because the data vgrinshpun was derived from power going into the charger, to which he showed it includes both (into the charger and actually depleted). Then it was suggested all of the values are calculated, not measured, which is where there seems to be the biggest disagreement and no one actually has factual data that shows if those test numbers are in fact calculated or measured. When asked to show a Model 3 that has gotten anywhere close to 78 kWh, there is a video of one going to 75 kWh but that test was apparently flawed, which sadly I can't figure out why, lol. I've always thought the trip meter showed total energy used while in motion, so if anything it would always show less than actual kWh usage. Anyways, just trying my best to follow along but can some with a brain please explain to me how Ben Sullin's test was flawed.

That is very objective description of the discussion we had.

Please note that in the previous posting i did not include pages that explicitly state what is actual test data. Take a look at my latest post, it includes data sheets which specifically label test results as such: "Range Test Results". These absolutely ARE the test results. Here is the M3 SR+ re-posted for convenience (link for the whole document is in my post above):

 

[shred86]

That is very objective description of the discussion we had.

Please note that in the previous posting i did not include pages that explicitly state what is actual test data. Take a look at my latest post, it includes data sheets which specifically label test results as such: "Range Test Results". These absolutely ARE the test results. Here is the M3 SR+ re-posted for convenience (link for the whole document is in my post above):

View attachment 399710

I saw those but initially I thought the argument was regardless of the title, "Range Test Results", they're still calculated values (so misleading label?). But then I saw the link to the full document which clearly states on page 4:

3.04 TEST PROCEDURES

SAE J1634 (as revised 2012‐10) was followed for all Range testing and SAE J2263 (as issued 1996‐10) was followed for Road load measurement.

SPECIAL TEST INSTRUCTIONS

This vehicle shall only be tested on a 4 wheel dyno.

Vehicle Settings

o Battery stabilized on the vehicle for 1000 miles.o Vehicle charged to 100% SOC
o Dyno Mode ON – This setting prevents TC faults.o Regen – Standard

o Climate OFF
o Headlights OFF
o Screen brightness set to default (50%)o Radio OFF

Instrumentation

o Battery voltage and Current measurement – Hioki 3390‐10o AC recharge – Hioki 3390‐10

...which also has pictures of the Model 3 on a four wheel dyno with the Hioki connections to the appropriate places, so I'm inclined to believe they actually tested this on a four wheel dyno and didn't calculate the values.

 

[David99]

The EPA does only 15% of the tests themselves. 85% is submitted by the car manufacturer. Most likely a simulation.

 

[DavidB]

The EPA does only 15% of the tests themselves. 85% is submitted by the car manufacturer. Most likely a simulation.

Here's a clue: If you see the exact same numbers, EPA did not do anything other than parrot what was given to them by the manufacturer.

 

[vgrinshpun]

I saw those but initially I thought the argument was regardless of the title, "Range Test Results", they're still calculated values (so misleading label?). But then I saw the link to the full document which clearly states on page 4:



...which also has pictures of the Model 3 on a four wheel dyno with the Hioki connections to the appropriate places, so I'm inclined to believe they actually tested this on a four wheel dyno and didn't calculate the values.

They absolutely did and do. I maintain a library of many of these applications from Tesla, which typically followed by confirmatory testing done by EPA - as I shown earlier in the case of Model S P85D, at the VW lab.

The reason I always post links to all documents I snap shot is because the original filing contains detailed description of data and how they were obtained. In a nut shell EPA testing includes running car on a dyno according to various EPA protocols (city, highway, etc.) while measuring energy out of the battery, (vehicle DC Wh/mile), depleting range (miles) and "event recharge energy" - energy required to recharge the battery.

 

[shred86]

Here's a clue: If you see the exact same numbers, EPA did not do anything other than parrot what was given to them by the manufacturer.

I'm probably just missing something here but the first 19 pages of this document, which is the "Request for issuance of a new certificate of Conformity to include a running change - Addition of a new variant to the RWD Model 3 Platform", is something submitted by Tesla to the EPA. This document specifies the conditions that were used to test the Model 3 under "Special Test Instructions" that says "This vehicle shall only be tested on a 4 wheel dyno." which includes the rest of the conditions, to include photos of a Model 3 on a four wheel dyno. The results of this test states the Tesla Model 3 LR battery capacity is ~78 kWh. On the attached pages, it shows the EPA results which I would agree, appear to just be calculated or simulated numbers. There's even a note that says "Fill in yellow shaded areas to compute range and AC wh/mi results". But that's besides the point... because it clearly appears Tesla actually tested their Model 3 on a dyno, unless they're flat out lying about the test instructions and the pictures are just fake, which I highly doubt.

 

[vgrinshpun]

The EPA does only 15% of the tests themselves. 85% is submitted by the car manufacturer. Most likely a simulation.

The data that I linked show that testing for Model S P85D was duplicated by EPA in the Volkswagen lab independently at the date other than the original testing by Tesla. The key is that coefficients that are entered into the dyno and its calibration need to representative of the actual load, and should be consistent between the tests, so variability is only due to the behaviour of the battery and drive unit of the car, not the dyno set.

I can always search EPA data base for the confirmatory independent testing data by EPA. The overall key is that manufacturer does not know in advance whether their data will be approved, or randomly verified through an independent test.

The largest difference between the manufacturer test and confirmation test by EPA I've seen so far was for Jaguar i-pace. EPA testing had much less charitable results. So they really enforce the verification of data by the manufacturers.

 

[vgrinshpun]

I'm probably just missing something here but the first 19 pages of this document, which is the "Request for issuance of a new certificate of Conformity to include a running change - Addition of a new variant to the RWD Model 3 Platform", is something submitted by Tesla to the EPA. This document specifies the conditions that were used to test the Model 3 under "Special Test Instructions" that says "This vehicle shall only be tested on a 4 wheel dyno." which includes the rest of the conditions, to include photos of a Model 3 on a four wheel dyno. The results of this test states the Tesla Model 3 LR battery capacity is ~78 kWh. On the attached pages, it shows the EPA results which I would agree, appear to just be calculated or simulated numbers. There's even a note that says "Fill in yellow shaded areas to compute range and AC wh/mi results". But that's besides the point... because it clearly appears Tesla actually tested their Model 3 on a dyno, unless they're flat out lying about the test instructions and the pictures are just fake, which I highly doubt.

I've addressed this before: the yellow shadowed area - Energy (Wh) and Distance (mi) as well as Recharge AC Watts are actual measured test data from the dynoed car. Other parameters on that sheet are calculated based on this test data.

 

[ucmndd]

Here is the summary, pertinent pages and link to actual test data of a physical battery pack. Pertinent test data are highlighted for clarity.

View attachment 399695

If I’m correctly understanding your interpretation of the numbers listed on these sheets as being actual physical measured capacity, you seem to be implying that the LR RWD and LR AWD actually have different sized batteries.

Do you actually believe that?

 

[vgrinshpun]

If I’m correctly understanding your interpretation of the numbers listed on these sheets as being actual physical measured capacity, you seem to be implying that the LR RWD and LR AWD actually have different sized batteries.

Do you actually believe that?

From what I see, including with my car, the battery capacity is not a rigidly fixed number, there is a variation due to the fact that testing involved two different cars with two different packs. I believe that nominally the battery capacity is approximately the same, but these two cars manufactured at different point in time, likely running different versions of BMS (note dates in the header row of the table) had actual measured battery capacity differ by roughly 1.28%.

 

[MarcG]

From today’s investor presentation on autonomy, Elon himself said “Model 3 long range has a pack of about 75kwh and short range about 50kWh”

 

[FlatSix911]

From today’s investor presentation on autonomy, Elon himself said “Model 3 long range has a pack of about 75kwh and short range about 50kWh”

The truth will set us free...

 

[Newscutter]

Ah... I see... we're arguing about minor variations in data and nothing actually useful to the average owner.

For those of us with 85 packs -- who aren't electrical engineers -- can @wk057 give us the layman version of the data?

Specifically, I've been concerned about a few stories of pre-AP 85s displaying miles above zero, some as high as 20ish, and dropping dead with no warning. I've adjusted my trip computing to increase my buffer accordingly, but obviously that affects range, charging time, etc. How is that explained with regards to the "kludge" buffer, etc.?

After nearly 5 years, I've noticed that the last few times I've attempted to full charge (or did successfully) the car would initially estimate something like "5 minutes remaining" then revert to a simple "charging" status FOR HOURS afterwards, while continuing to draw current.

My assumption has been that years of daily drives between 60-80% just threw off the algorithm and therefore my car's rated range would be reported inaccurately. This gives me some peace of mind re: degradation at the top (oh, it's reading lower than new because it's just off) but also less peace of mind regarding degradation at the bottom (oh, it says 40 miles remaining but it might go dark at 20 and surprise me).

If it's rebalancing/calibrating constantly.... then how trustworthy is the reported state of the battery at the top/bottom?

Don't need to get buried in the electrical details, just need an user's understanding of what my car is telling me (or not). Thanks!

I'd posted some screenshots of the charging behavior here: Tesla Pittsburgh on Twitter
And here: Tesla Pittsburgh on Twitter

Ah... I see... we're still just arguing about minor variations in data and nothing actually useful to the average owner, like the questions I asked above. So... Imma gonna go out here and measure my panel gaps to the micron and then see if that can be argued against the published specs from the manufacturer. Perhaps something of relevance will appear in the meantime.

I'd also like to state for the record that not ALL residents of PA are so stridently disagreeable as my colleague with the walls of text.

 

[vgrinshpun]

From today’s investor presentation on autonomy, Elon himself said “Model 3 long range has a pack of about 75kwh and short range about 50kWh”

I see, the “about” part, in your opinion, is useful in evaluating accuracy of actual test data..

 

[MikeBur]

I had to show ignored posts to understand the context of this resurrected thread, though I guess I have to say thank you to @vgrinshpun at this moment for their lack of ability to receive other data as they caused you to post this additional info in response. Wish people would read more carefully, eg bit on 3 data on CAN bus now padding real telemetry, plugged in power for testing, etc.
Not sure whether to repeat guidance on how to ignore noisy few now: makes for calmer reading, though we wouldn’t have gotten this gem if so.

Thanks for posting again Jason.

Asked the friend at Tesla who leaked a bunch of stuff to me about this, and he said "Just post it"...

So, here's a screenshot of an internal document on the Model 3 battery.

Now seriously. Just stop trying to argue this.