Tesla P85D Insane Mode

[Ohbse]

It's more like 20A. You have to account for voltage sag. NCR18650BE would sag down to 3.2V at 20A.
And only for 5 seconds at a time. So they can deal with smaller conductors....

Great remark :wink:

Doc

Poking around... It looks like the P85D is rather over-rated. They probably rate for peak front + peak rear motor power. But the trick is not at the same time.

Proof:
1.) search for the dyno data on the web, its much lower then expected.

2.) a 2013 P85 (416hp rated) traps at 111.5mph (http://www.dragtimes.com/Tesla-Model-S-Timeslip-25833.html?r=1)
a p85D traps at 115mph. (see motortrend)

It's well known that trap speed is very indicative of power/weight and power. The numbers look more like 480-500hp, not 691, sadly.
That "only" comes out to ~14-16 amps per cell, which a NCR18650BE would handle just fine in 5-7 second bursts.
A BE would sag to 3.4-3.5 V at 15A in a quick burst at 90% SOC. (source review of probably a rewrapped black market chinese NCR18650BE: http://lygte-info.dk/review/batteries2012/Keeppower%20IMR18650%203200mAh%20(Black)%202014%20UK.html)
Do the math and that comes to 500 battery horsepower.

People (and Tesla marketing) appear to have arrived at the 691hp figure by adding together the ratings of the front and rear motors. This is obviously not very accurate. It might be representative of the power potential, but at this moment in time I don't think the battery is capable of delivering the full '691hp' under many circumstances. HOWEVER it really doesn't need to. The idea of validating horsepower numbers with trap speed doesn't really work with electric drive-train. The power curve of a well designed EV is wildly different from a high performance ICE engine. The power curve is more like a table with torque falling continuously as RPM climbs. This naturally doesn't result in amazing high speed performance and can skew the results of empirical testing. It can appear slow from the numbers (430 hp at the wheels in the case of the P85D) but the reality of driving it is completely different. What makes a car fast isn't peak horsepower, it's area under the curve. In the past when I was into ICE tuning I ran many 12 second 1/4 mile passes with a 1490cc 4 cylinder engine from a junkyard in a front wheel drive hatchback. This made a mere 110kw or so, however it made that power over a VERY broad RPM range. Geared appropriately it would embarrass many very expensive vehicles that made 3-6x the power (and cost 5-25x more money)

There's nothing not going to be much around that is going to out-accelerate the P85D on the street, certainly not with less than perfect surface quality. Yeah, you're probably going to lose if you're racing from a 50mph roll or higher but outside of street racing douchbags I don't think anybody will really care.

 

[flathill]

[/quote]
Poking around... It looks like the P85D is rather over-rated. They probably rate for peak front + peak rear motor power. But the trick is not at the same time.

Proof:
1.) search for the dyno data on the web, its much lower then expected.

2.) a 2013 P85 (416hp rated) traps at 111.5mph (http://www.dragtimes.com/Tesla-Model-S-Timeslip-25833.html?r=1)
a p85D traps at 115mph. (see motortrend)

It's well known that trap speed is very indicative of power/weight and power. The numbers look more like 480-500hp, not 691, sadly.
That "only" comes out to ~14-16 amps per cell, which a NCR18650BE would handle just fine in 5-7 second bursts.
A BE would sag to 3.4-3.5 V at 15A in a quick burst at 90% SOC. (source review of probably a rewrapped black market chinese NCR18650BE: http://lygte-info.dk/review/batteries2012/Keeppower%20IMR18650%203200mAh%20(Black)%202014%20UK.html)
Do the math and that comes to 500 battery horsepower.[/quote]

People (and Tesla marketing) appear to have arrived at the 691hp figure by adding together the ratings of the front and rear motors. This is obviously not very accurate. It might be representative of the power potential, but at this moment in time I don't think the battery is capable of delivering the full '691hp' under many circumstances. HOWEVER it really doesn't need to. The idea of validating horsepower numbers with trap speed doesn't really work with electric drive-train. The power curve of a well designed EV is wildly different from a high performance ICE engine. The power curve is more like a table with torque falling continuously as RPM climbs. This naturally doesn't result in amazing high speed performance and can skew the results of empirical testing. It can appear slow from the numbers (430 hp at the wheels in the case of the P85D) but the reality of driving it is completely different. What makes a car fast isn't peak horsepower, it's area under the curve. In the past when I was into ICE tuning I ran many 12 second 1/4 mile passes with a 1490cc 4 cylinder engine from a junkyard in a front wheel drive hatchback. This made a mere 110kw or so, however it made that power over a VERY broad RPM range. Geared appropriately it would embarrass many very expensive vehicles that made 3-6x the power (and cost 5-25x more money)

There's nothing not going to be much around that is going to out-accelerate the P85D on the street, certainly not with less than perfect surface quality. Yeah, you're probably going to lose if you're racing from a 50mph roll or higher but outside of street racing douchbags I don't think anybody will really care.[/quote]

--------------------------------END QUOTE----------

#1 Trap speed does not indicate peak HP.

#2 Tesla has always been honest spec wise (usually underrated). If they rated the car at 691 HP then the car can draw that much or more from the battery pack (depending on traction). In all the dyno runs I've seen of the P85D they are feathering the throttle. Did they disable traction control? What happens on dyno with traction control on? What happens on a dyno with traction control off? We also don't know if they set the atmospheric correction right on the dyno computer. The output of an electric motor is not dependent on altitude like a combustion engine. From the couple P85D dyno runs I've seen they are registering LESS peak power than the P85. This can't be right as the P85D is showing much HIGHER peak power draw on the dashboard than the even the very last P85's made with the new low impedance packs.

 

[Punx0r]

Being 4WD the P85D shouldn't be any harder to dyno than the P85, as the extra power is being applied to another axle, so traction shouldn't be any more of a problem.

Regarding atmospheric correction on a dyno, it's referenced to sea level. Most dynos are above sea level, so the correction increases indicated HP over what was actually measured. So if anything if an atmospheric correction factor has been wrongly applied to a Model S the dyno should over-read, if anything.

 

[flathill]

The dyno also corrects for temp and humidity but this would only change the results a few percent. You are right this cant account for the large discrepancy between what the dyno says and what Tesla says

 

[flathill]

Looks like I could be wrong trusting Tesla

"The company is already working on an update to the website to explain this distinction between net power and "motor power."

http://www.greencarreports.com/news/1095000_puzzling-new-power-numbers-for-tesla-model-s-whats-the-deal

Tesla need to clear this up ASAP

 

[okashira]

Looks like I could be wrong trusting Tesla

"The company is already working on an update to the website to explain this distinction between net power and "motor power."

http://www.greencarreports.com/news/1095000_puzzling-new-power-numbers-for-tesla-model-s-whats-the-deal

Tesla need to clear this up ASAP

No they don't really. manufacturers have been rating hybrids the same way for some time.
It's just a number. You don't hear anyone complaining about the car's performance.
The car is sold as an overall package.
We tried to figure out the power from the dash display. it's logarithmic and not very precise, but it seems to stop at about 480kw. That's probably battery V * I, but still, not too shabby.

Edit: sorry, it looks like ~432kw.
that would be 60.81 W per battery cell, I would assume the dash measures battery voltage x current.
a "grade a" ncr18650be would probably sag to ~3.25V at 20A or ~3.45V at 15A at ~90% charge (http://lygte-info.dk/review/batteries2012/Keeppower IMR18650 3200mAh (Black) 2014 UK.html So were looking at ~18.42A @ 3.3V.
18A for ~7 seconds is well within the surface charge regime of one of these cells

 

[Punx0r]

Yeah, that is pretty sleazy. I guess the marketing department has gotten its way on this one.

Claimed BHP for a car should be the same as for an ICE car: measured motor shaft power, including all ancillaries. Anything else is misleading and potentially false advertising. It would be like selling a 100HP ICE car as 200HP on the basis that if you replaced the fuel system with uprated parts and remapped the ECU it could theoretically produce that.

 

[liveforphysics]

If we know anything, it's that IF the early p85 cells were BE's (which I think you guys are mistaken on), then you KNOW the P85D's cells can not be BE's. Why do people continue to make this known incorrect conclusion...


You guys are smarter than this, why continue to act like fools.

 

[okashira]

If we know anything, it's that IF the early p85 cells were BE's (which I think you guys are mistaken on), then you KNOW the P85D's cells can not be BE's. Why do people continue to make this known incorrect conclusion...


You guys are smarter than this, why continue to act like fools.

The only thing foolish I see is those claiming unsubstantiated statements of opinion as fact.

 

[liveforphysics]

If we know anything, it's that IF the early p85 cells were BE's (which I think you guys are mistaken on), then you KNOW the P85D's cells can not be BE's. Why do people continue to make this known incorrect conclusion...


You guys are smarter than this, why continue to act like fools.

The only thing foolish I see is those claiming unsubstantiated statements of opinion as fact.

Did you miss that P85's got new cells mid 2014? It is the reason why the quarter mile trap speeds all jumped up and times went down after a certain production date in 2014.

That means, IF it were the BE cell before that date, it is no longer the BE cell used after that date, which would include all P85D's made.

I don't claim to know the name of the new cell, or the old cell, because my friend who works in Teslas battery engineering told me it's not a cell you can find a datasheet for or purchase aside from buying them already assembled into a Tesla pack.

 

[Punx0r]

I'd be neat if he could name the most similar cell that datasheets are available for and give an indication of how similar the Tesla cells are

 

[okashira]

If we know anything, it's that IF the early p85 cells were BE's (which I think you guys are mistaken on), then you KNOW the P85D's cells can not be BE's. Why do people continue to make this known incorrect conclusion...


You guys are smarter than this, why continue to act like fools.

The only thing foolish I see is those claiming unsubstantiated statements of opinion as fact.

Did you miss that P85's got new cells mid 2014? It is the reason why the quarter mile trap speeds all jumped up and times went down after a certain production date in 2014.

That means, IF it were the BE cell before that date, it is no longer the BE cell used after that date, which would include all P85D's made.

I don't claim to know the name of the new cell, or the old cell, because my friend who works in Teslas battery engineering told me it's not a cell you can find a datasheet for or purchase aside from buying them already assembled into a Tesla pack.

Please show some evidence on the trap speed jumps ? I haven't seen anything. The p85 I referenced was a 2013 model with 111.5 mph. And a 2015 model p85d with only 115mph. That's not much... sure the car is probably 150lbs heavier but it's already 4800lbs.

And your friend at Tesla is probably being over-confident.
Panasonic doesn't officially sell cells to ANYONE except in pre-built packs, except Tesla, so technically his statement is true.
That's why all the Panasonic cells are "grey market"
Tesla has no control over what Panasonic does with their batteries...
What would Panasonic do with all the Tesla rejects, of which there would be many thousands... I mean they have made 4 billion of the exact same cell.

I'll gladly eat my words if/when I pop open a pack and test a cell. Either way, I am sure they will make great cells for an e-bike.

 

[okashira]

After experiencing this first hand recently, the amazing part is how the acceleration hits immediately off the line. Traditional cars, even high performance ones, have output that ramps up. This is partly due to turbos and things spinning up and also just fundamental limitations to torque from a IC engine at low RPM.

The Tesla is just instantly 100% right off the line. It is basically like being dropped into free-fall at a carnival ride.

I wonder if the motor is sensored.

 

[riba2233]

It's an induction type motor (not BLDC), it probably has some sensors, but encoder type, not hall's.

@Luke: if they got better batteries, it can be just a higher grade of the same cell. BE is the only type that physically matches Tesla cells, and PD is the only type that matches cells in RAV4 pack. Also, they make the most sense, they have the high capacity, but also high current capability. They are only Panasonic cell that matches these factors... And also, if Panasonic had a much stronger cell with higher capacity, why would they only sell it to Tesla, they would lose a lot of potential money. Technology has it's limits, and companies like to brag about new accomplishment, especially in battery technology...

 

[Doctorbass]

I guess that TESLA also calculated a safety margin to keep battery life realistic so probably that the actual cells in the P85D are much better than what we think! estimating cell max rating based on few test bench and spec is far from the conditions the Ev real world use! I think they have to be very conservative!

Repetitive 65Watts draw on a single 18650 over the 8 years life warranty of the tesla is just incredible.. any actual 18650 cells would have significant life reduction in these conditions...

let say the P85D owner like to floor the pedal to every ride it make because he like the feeling... how could these cells endure that over 8 years old... lol

I know it's more about the average power the cells is used at but I guess that repetitive power burst also affect the cells life..

Let say they calculated that the cell will take a WOT during few sec 1% of time, if the power that this 1% represent 2000 time more cell deterioration than the rest of the 99% normal C rate use, this change a lot the cells life! even though t's just 1% of time..

Doc

 

[Hillhater]

...
let say the P85D owner like to floor the pedal to every ride it make because he like the feeling... how could these cells endure that over 8 years old...

Do you, or any one else you know,... Regularly "floor it". In your daily driver ?
I doubt few people do. 75%...85%, maybe sometimes, but full balls out max is asking for trouble from many sources !
The only guys that use their performance to the max , are track / strip racers, and even they know you don't hammer your equipment if it's not necessary.
So other than the initial " show off". sessions, and odd stop light blast, I doubt many Tesla owners hit their motors with max power very often !
.....they can see instantly what it does to their range ...

 

[liveforphysics]

It's an induction type motor (not BLDC), it probably has some sensors, but encoder type, not hall's.

@Luke: if they got better batteries, it can be just a higher grade of the same cell. BE is the only type that physically matches Tesla cells, and PD is the only type that matches cells in RAV4 pack. Also, they make the most sense, they have the high capacity, but also high current capability. They are only Panasonic cell that matches these factors... And also, if Panasonic had a much stronger cell with higher capacity, why would they only sell it to Tesla, they would lose a lot of potential money. Technology has it's limits, and companies like to brag about new accomplishment, especially in battery technology...

The leading edge of the battery industry is sadly entirely private, unless you happen to be an OEM of sufficient size to begin the conversation.

If you were sitting down right now with Panasonic's own sales team, and you had already signed the pages of NDA documents and other lawyer BS, you still wouldn't get awareness of the current flagship 18650 offerings unless you were dangling enough volume demand to tempt them. I can't think of the scale an ebike project would need to be to make that happen... lol

For a supplier relationship like Tesla's, where there own vehicle production numbers are restricted by the capacity of Panasonic's production lines, what benefit would they have in trying to increase demand?

 

[okashira]

Yep, the dash shows battery power:
https://www.youtube.com/watch?v=JEPBkjmS2uE
Check at 3:45 and he has the car parked, but it's drawing ~2-3kw to heat the battery. (he says 6kw, but he doesn't understand it's logarithmic.)

I did a little closer analysis of the power...
Based on two videos, one at 100% charge, shows ~390kw draw peak and 70% charge shows 370kw draw peak (assuming 40% past the half way point and 25% past the halfway point eyeballing the lines)

Only 55 W per cell when full. A fully charged ("grade A") NCR18650BE would only sag to ~3.4V-3.45V at that power for 5 seconds... for only 15.94 - 16.18A per cell.
At 70% SOC... My testing shows there is significantly less pulse DCIR then at near full charge (http://endless-sphere.com/forums/viewtopic.php?f=14&t=56549&start=200#p996072) for these type cells.
At 99% we had a 0.72V drop (4.14 [assumed] - 3.425) but the resting voltage at 70% is about ~3.87V and pulse DCIR is about 68% of that at full charge (quick measurements from the chart with interpolation, feel free to verify) so the drop should be .72*.68 = 0.4878V or 3.38V.

3.38V/3.425 * 390 = 384kw approx equal to 370kw, id say thats well within the error of all my estimates and measurments.

conclusion, 390kw peak battery power at full charge, ~365kw peak at 50% charge
assuming 94% efficiency, of which is probably a low estimate, tesla probably balances the load between the two inverter/motors so they are near peak - that's ~500 mechanical horsepower
And only 16A peak draw per cell.

And 16A draw from each battery. That's perfectly fine for a NCR18650BE for 5 seconds at a time (power tapers significantly by 100mph which is only 5 seconds from 45mph)

also add in that it's single speed transmission, no oil windage losses, no inertal losses from low gears, no flywheel/driveshaft, no time lost changing gears, and near peak power at 35mph, etc, so no 15% losses like a gasser... probably more like 3%.

There's your Tesla magic.

 

[Doctorbass]

Yep, the dash shows battery power:
https://www.youtube.com/watch?v=JEPBkjmS2uE
Check at 3:45 and he has the car parked, but it's drawing ~2-3kw to heat the battery. (he says 6kw, but he doesn't understand it's logarithmic.)

I did a little closer analysis of the power...
Based on two videos, one at 100% charge, shows ~390kw draw peak and 70% charge shows 370kw draw peak (assuming 40% past the half way point and 25% past the halfway point eyeballing the lines)

Only 55 W per cell when full. A fully charged ("grade A") NCR18650BE would only sag to ~3.4V-3.45V at that power for 5 seconds... for only 15.94 - 16.18A per cell.
At 70% SOC... My testing shows there is significantly less pulse DCIR then at near full charge (http://endless-sphere.com/forums/viewtopic.php?f=14&t=56549&start=200#p996072) for these type cells.
At 99% we had a 0.72V drop (4.14 [assumed] - 3.425) but the resting voltage at 70% is about ~3.87V and pulse DCIR is about 68% of that at full charge (quick measurements from the chart with interpolation, feel free to verify) so the drop should be .72*.68 = 0.4878V or 3.38V.

3.38V/3.425 * 390 = 384kw approx equal to 370kw, id say thats well within the error of all my estimates and measurments.

conclusion, 390kw peak battery power at full charge, ~365kw peak at 50% charge
assuming 94% efficiency, of which is probably a low estimate, tesla probably balances the load between the two inverter/motors so they are near peak - that's ~500 mechanical horsepower
And only 16A peak draw per cell.

And 16A draw from each battery. That's perfectly fine for a NCR18650BE for 5 seconds at a time (power tapers significantly by 100mph which is only 5 seconds from 45mph)

also add in that it's single speed transmission, no oil windage losses, no inertal losses from low gears, no flywheel/driveshaft, no time lost changing gears, and near peak power at 35mph, etc, so no 15% losses like a gasser... probably more like 3%.

There's your Tesla magic.

I like math ! but i'm too lazy to go too deep... Thanks for that exemple 8)

Doc

 

[liveforphysics]

The pack is ~96s74p, it has to be ~20A per cell due to sag, which is a steep slope near the cells saturation. This wouldn't make sense to do in a product with a warranty. The BE cell would barely run a p85 with longevity, let alone the p85d.

When you compare the trap speeds of the cars and see only a small difference, remember that one had substantially less time to be accelerating.

 

[gogo]

When you compare the trap speeds of the cars and see only a small difference, remember that one had substantially less time to be accelerating.

Yes. Trap speeds measure a small portion of an acceleration event. A previous poster drew errant conclusions from them.

This is what many thrill seekers fail to grasp about drag racing. Cars are expensive in the fun/dollar measure. The best bargain is two-wheeled with a long enough swingarm to keep from flipping over, or a wheelie bar. Speed is dangerous, so the acceleration thrills on a two-wheeler in the first 1/8 mile are the safest and best fun/dollar. Plus hanging onto the handlebars in the open air vs. being inside a cabin sitting sedately in a seat are just two different categories altogether.

 

[okashira]

When you compare the trap speeds of the cars and see only a small difference, remember that one had substantially less time to be accelerating.

Yes. Trap speeds measure a small portion of an acceleration event. A previous poster drew errant conclusions from them.

This is what many thrill seekers fail to grasp about drag racing. Cars are expensive in the fun/dollar measure. The best bargain is two-wheeled with a long enough swingarm to keep from flipping over, or a wheelie bar. Speed is dangerous, so the acceleration thrills on a two-wheeler in the first 1/8 mile are the safest and best fun/dollar. Plus hanging onto the handlebars in the open air vs. being inside a cabin sitting sedately in a seat are just two different categories altogether.

Indeed, I am glad I learned my lesson on cars early on with only $5000 spent (15,000 on my project, sold for $10,000!)
And you can build a bad-ass e-bike for $2000 total.

Anyway, I still stand by my claims - 500hp and 16A-17A per cell.
Here is what a car with a real 700hp does: https://www.youtube.com/watch?v=o_aB9JTiFVE
And that car has an automatic transmission, a heavy driveshaft to spin, oil windage losses in the engine, tranny and rear end to deal with, yet it pulls 127mph which is a huge jump over 114-115mph. The hellcat is quite a heavy car as well.

 

[zombiess]

Here is what a car with a real 700hp does: https://www.youtube.com/watch?v=o_aB9JTiFVE
And that car has an automatic transmission, a heavy driveshaft to spin, oil windage losses in the engine, tranny and rear end to deal with, yet it pulls 127mph which is a huge jump over 114-115mph. The hellcat is quite a heavy car as well.

The Hellcat is a fat bastard for no good reason, it's am embarrassment to Dodge for making such a heavy car.

Your statement about the automatic trans makes it sound like you think an auto trans is a negative, for drag racing a good auto trans is almost always faster than a manual trans. I've seen cars pick up 6-7 MPH just by switching. The Tesla is even better since it only has 1 gear and at least it's ridiculous truck like weight is because of the batteries.

 

[liveforphysics]

If you recall, Elon mentioned that he hadn't yet unleashed the full potential of the vehicle yet. Then later confirmed that a later update would improve the vehicles official 0-60 time to 3.1s. Keep in mind, he previously claimed it was 3.3, and when tested the vehicle achieved 3.1.

At least the way the vehicle was dyno'd here, it shows Tesla begins sharply reducing motor phase current as soon as 27mph with the current firmware. I think keeping the performance artificially restricted at this moment is to reduce the impact of the vehicle sales back-order log. If they catch up on sales of P85D, release the greatest performance increasing firmware update in the history of the automotive world to yet again go back to continuously back-ordered.

This about it, while you're production limited, sold out and backlogged for many months on orders, what do you gain by increasing demand? Makes sense to stage the consumer appeal when it's as effortless as an automatic OTA firmware push.

 

[Punx0r]

Bit of a bump because I read today (albeit from a couple of months ago) a comment by the CEO of Aston Martin that reminded me of this thread. The topic if the 800HP BEV Aston Martin Rapide supposedly due for release in two year's time. A zero-emission vehicle is necessary for AM to continue making ICE cars with large, powerful engines (which customers like) while meeting average fleet emissions requirements.

CEO Andy Palmer said of Tesla:

“We don’t do Ludicrous because Ludicrous speed is stupid,” Palmer said.

“I think that the fact that you could drive a few laps of a decent race course or race it around the Nordschleife [famed track in Germany] is much more interesting than doing 500 meters in Ludicrous mode.”

Article: http://www.autonews.com/article/20150818/OEM04/150819871/aston-aims-to-launch-800-hp-all-electric-rapide-in-2-years

In another article this is clarified as " the Aston's "insane mode comes as standard - no button required".

This implies peak power output that is sustainable for much longer periods than the Model S is capable of. It'd also be a win if the car is genuinely 800HP compared to the P85D's <500HP (despite being marketed as 800HP).

If nothing else it would be interesting to have a model of luxury sports car that's otherwise the same but available in either V12 ICE or BEV. How will they compare? Which will people pick after a test-drive?