[updated with *] P85D 691HP should have an asterisk * next to it.. "Up to 691HP"

[stopcrazypp]

Perspective doesn't come into this.

By Tesla's own account, P85D owners who order Ludicrous aren't getting new drive units (though they are getting new PEM firmware.) They're getting new contactors and a smart fuse for their battery pack. No new motor windings. No change to pack voltage. In short, nothing you would expect to influence the power fade factor due to shaft speed.

Given that changing one variable - peak current - is bringing the P85D close to what would unambiguously be meeting the originally announced specification, what do you think that suggests?

I'm talking more about: x = 0-60, y= hp motor power, z = 1/4 mile time (you can add more promises like top speed etc).

A peak current increase directly implies a peak power increase.

Not necessarily. It depends on what rpm they apply that peak current and how much sag they are getting at that point too. For example, for the Roadster they applied the highest current at lower rpm to boost acceleration. We unfortunately don't have current and voltage curves mapped to rpm, so we don't actually know where Tesla is applying peak current.

So far, so obvious. We're not comparing it to any ICE car, or to a BEV with rapid power fade at speed. In any case, I expect the motors in P85D to exhibit power fade at high speed due to back-EMF, exactly like any other induction motors. None of this is relevant to the disputed figures.

Knowing simply only a peak power number does not tell you how to expect the car to perform in the high end. Getting back to the point: if they can extend the amount of time the car stays at peak power, that will help high end acceleration without changing the peak number. On the flip side if they only increase peak power for a short period near the lower rpms, that will not help acceleration enough on the top end. I bring up the ICE example because without fail, power continues to climb as rpms climb, so expecting a better 1/4 mile with a higher peak power tends to work out. However, BEVs have a different power curve (either plateaus very early because of battery limits, or it falls rapidly at the high end because of motor limits), so peak power doesn't tell you that.

Edit: more directly, at launch the 1/4 mile number was already known but no one was able to tell from that that the car didn't make_ 691 up shaft power. So it does not directly follow.

Meaning it is reasonable to expect that the car is fitted with motors that will, at some point in the performance envelope, actually output 691HP. If they don't, that's something that ought to be mentioned explicitly - especially given that all previous Tesla vehicles had done precisely this and all ICE vehicles are legally required to.

Please show me where there is a law that says that and how it applies to an EV. I have yet to seen this argument made. I know automakers are required by law (at least in the US) to show the fuel economy, foreign parts content information, safety ratings, options and equipment installed, but nowhere on the legally required new car sticker does it show horsepower.

Which is misleading, and a betrayal of the normal intent of engine power ratings.

Saying it is misleading and saying it is impossible is two different things. I hope we have agreement least that it is possible for Tesla to use such a measure. And getting back to the main point: isn't it more misleading/worse to knowingly advertise a shaft power number that is impossible to achieve? I don't believe Tesla would do that.

 

[JER]

I'm talking more about: x = 0-60, y= hp motor power, z = 1/4 mile time (you can add more promises like top speed etc).

So?

Not necessarily...

Given that peak current limit is the only variable being changed, yes necessarily.

Raising the current cap and adjusting the PEM control laws to exploit it means - pretty much by definition - that when run flat out the motor will draw more current whenever it is not limited by some other factor, and in doing so develop more power.

Those limiting factors are torque limiting at low speeds to prevent wheelspin, and back-EMF preventing full power at high shaft speeds. (There's also thermal throttling, but that's not relevant here.) That leaves a space in the middle where more current flows, more power is generated and more acceleration occurs.

Knowing simply only a peak power number does not tell you how to expect the car to perform in the high end.

When all other variables are the same - which they are - it tells you to expect higher performance at high speed ("high" being above whatever the transition speed from traction limited to power limited acceleration was.) Above thirty miles an hour in this case. It won't be higher everywhere in that part of the envelope, but it will be somewhere, and it won't be lower anywhere. Remember: All other variables the same.

As I said, if you won't take my word for this, take Elon's. If you're going to keep arguing the toss over this, argue with him.

We're not comparing wildly different cars here. Were comparing delivered P85Ds with a hypothetical, almost identical P85D with the battery current cap relaxed by 200A and PEMs reprogrammed accordingly. A car that will no longer be hypothetical when people start getting the upgrades. A car that, based on what we know about P90D, we can project will almost meet the specification announced last October.

Please show me where there is a law that says that and how it applies to an EV.

Products must be as described. With ICEs, there are established rules on how engine power can be legitimately measured, if a figure is given. The overriding principle of those regulations is that claims must be representative of what the engine actually does in the car.

The power figures quoted for Model S variants are all representative of what actually happens in the car, except for the PxxD.

EVs have not yet been around long enough for decent regulations to exist, but they will. This is exactly the kind of chicanery they'll have to address.

In the meantime it is possible for Tesla to do this with the PxxD. It is also possible for them to not do it, which is precisely what is going on with every other variant of Model S - including the dual motor ones - in the design studio today. Why do you suppose that is?

 

[cynix]

Did Tesla really list ECE R85 as the standard the used for the P85D or was it for another Model S variant? I never saw it stated. Was it only in Europe which would make sense since it's a European standard.

It's in the owner's manual:

 

[dsm363]

Products must be as described. With ICEs, there are established rules on how engine power can be legitimately measured, if a figure is given. The overriding principle of those regulations is that claims must be representative of what the engine actually does in the car.

And that's the problem. A lot of assumptions were made about what Tesla meant. This hp issue is obviously the most important metric in the world to some and yet there is no defined and independently verified standard way to test these electric motors apparently. If hp is the only thing that really mattered to me I would have wanted to confirm these specs prior to ordering.

 

[Krugerrand]

If hp is the only thing that really mattered to me I would have wanted to confirm these specs prior to ordering.

You would also have verified that ICE hp was the same as/equivalent to EV hp motor power - if hp was the only (or main) thing that really mattered to you.

 

[stopcrazypp]

In the meantime it is possible for Tesla to do this with the PxxD. It is also possible for them to not do it, which is precisely what is going on with every other variant of Model S - including the dual motor ones - in the design studio today. Why do you suppose that is?

Actually my point has been that it has been going on for EVERY VARIANT of the Model S. Remember, the first incident where "motor power" was discovered to not mean shaft power was not actually with the P85D, it was with the S60 vs the S85. Under a scheme where shaft power is the measure, it is impossible for both of them to be rated the same 380hp "motor power". Yet they are. David Noland immediately noticed this the day the numbers were change to "motor power" (in mid October 2014). People did not notice it for the P85D until considerably later (in March 2015).

The same deal goes with S60D vs S85D (both rated 376 hp motor power), S70D vs S85D (both rated 514 hp motor power), and currently S70 vs S85 (both rated 382hp motor power). Given the different net power numbers advertised for 60/70 vs 85 battery variants, it is impossible for them to share the same motor power numbers if motor power means shaft power.

I spent a considerable amount of time gathering all the combined motor power numbers Tesla has ever advertised (US site, UK site might have sliley different numbers), in case you want to see the data:
http://www.teslamotorsclub.com/show...91HP/page119?p=1115469&viewfull=1#post1115469

 

[sorka]

It's in the owner's manual:

View attachment 93543

Interesting. If you add up the kW of the large and small performance motor, you get 543 KW or 1696 amps at 320 volts which far exceeds even the new Ludicrous fuse limit let alone the Insane 1300 amp fuse limit. It's obvious that what they did is measure one motor on the battery and then the other motor on the battery rather than at the same time. Also, 320 volts is a 20% drop and the test only allows a 5% maximum drop in voltage before you hit your current limit.

- - - Updated - - -

Actually my point has been that it has been going on for EVERY VARIANT of the Model S. Remember, the first incident where "motor power" was discovered to not mean shaft power was not actually with the P85D, it was with the S60 vs the S85. Under a scheme where shaft power is the measure, it is impossible for both of them to be rated the same 380hp "motor power". Yet they are. David Noland immediately noticed this the day the numbers were change to "motor power" (in mid October 2014). People did not notice it for the P85D until considerably later (in March 2015).

The same deal goes with S60D vs S85D (both rated 376 hp motor power), S70D vs S85D (both rated 514 hp motor power), and currently S70 vs S85 (both rated 382hp motor power). Given the different net power numbers advertised for 60/70 vs 85 battery variants, it is impossible for them to share the same motor power numbers if motor power means shaft power.

I spent a considerable amount of time gathering all the combined motor power numbers Tesla has ever advertised (US site, UK site might have sliley different numbers), in case you want to see the data:
http://www.teslamotorsclub.com/show...91HP/page119?p=1115469&viewfull=1#post1115469

Not every variant. The S85 claimes 373 hp. It makes 375 at the wheels. It's more than Tesla claimed. The 85D claims 417 hp but makes 504 at the battery. Clearly far more than claimed. The P85 claimed 470 hp (at the motor shaft) and dynos 435 hp at the wheels which is clearly more than claimed unless the drivetrain loss is less than 7%.

The S60 on the other hand claimed 380 hp and based on reported REST numbers and it's 5.9 second 0-60 is not even close to that number. Nobody is screaming about the S60 because nobody it bought for performance and they're not going to care. That doesn't make it right, but it does mean it's not going get any attention.

 

[Laserbrain]

In the meantime it is possible for Tesla to do this with the PxxD. It is also possible for them to not do it, which is precisely what is going on with every other variant of Model S - including the dual motor ones - in the design studio today. Why do you suppose that is?

I think its a scam from Tesla's marketing departement. It is really rather obvious:

Plus using no rollout with the 70D and 85D but suddenly using rollout with the P85D... I mean... come on. Really?

Drag strip acceleration with rollout...which has nothing - NOTHING - to do with a real world physics based 0-60 measurement of speed and time. And you do not have to be Sheldon Cooper to see this.

 

[Torben_E]

Sorka...

You know the 400 V is at the DC level. That drops to 340-350 V when max AMP is pulled (20 A x 0,40 Ohm per cell x 74) as well as the battery disharges (SoC og say 25%).

I would like to suggest that the DC to 3 phase AC inverter always feed the motor with 320 VAC. So wheter the inverter is feed with 400 VDC or 375 VDC or 350 VDC it keep the 320 VAC on the output side. The current however it will then control together with the frequency to try to keep constant FLUX. The FLUX will not be able to be kept constant (as the rpm of the motors increases) and as a consequence the torque will eventually start to drop. Maybe at 9.000-11.000 rpm.


Is there anyone that can confirm this or has the datasheet of the inverter ?


Torben_E

 

[MuppetOwl]

The Tesla specs from the service caralog and manual shows the engine is 320V ?

 

[jerjozwik]

Plus using no rollout with the 70D and 85D but suddenly using rollout with the P85D... I mean... come on. Really?

Drag strip acceleration with rollout...which has nothing - NOTHING - to do with a real world physics based 0-60 measurement of speed and time. And you do not have to be Sheldon Cooper to see this.

tesla is marketing the 85D and P85D to different groups. 85D is just a good representation of 0-60. for the P85D they are going against serious performance cars and need to have accurate 0-60 so, being a automotive company in the united states they use the conventional 0-60 process for the united states... which is based on drag performance which means with rollout.

I care about 0-60, I knew about rollout long before thus thread, I will order a p85d regardless because its an impressive machine.

the wife, care little about 0-60 and wants an 85d for the range...

 

[Andyw2100]

tesla is marketing the 85D and P85D to different groups. 85D is just a good representation of 0-60. for the P85D they are going against serious performance cars and need to have accurate 0-60 so, being a automotive company in the united states they use the conventional 0-60 process for the united states... which is based on drag performance which means with rollout.

I care about 0-60, I knew about rollout long before thus thread, I will order a p85d regardless because its an impressive machine.

the wife, care little about 0-60 and wants an 85d for the range...

This would have been a fine approach for Tesla to have taken if they had also included little decoder rings, so anyone looking at the specifications could have figured out just when Tesla was doing what and for whom. Failing that, words would have done the trick.

Thankfully, as of less than a week ago, Tesla is now using words.

 

[jerjozwik]

they did not even hand out those decoder rings at the D event. oh well...

 

[stopcrazypp]

Interesting. If you add up the kW of the large and small performance motor, you get 543 KW or 1696 amps at 320 volts which far exceeds even the new Ludicrous fuse limit let alone the Insane 1300 amp fuse limit. It's obvious that what they did is measure one motor on the battery and then the other motor on the battery rather than at the same time. Also, 320 volts is a 20% drop and the test only allows a 5% maximum drop in voltage before you hit your current limit.

320V may be the AC voltage, not necessarily the DC test voltage. I remember for the Roadster, they used the PEM to boost output current higher than what was coming out of the battery (although obviously the output voltage would be lower).

The Roadster PEM can output 850A for 1.5 (1.0 version was only 640A).
http://www.teslamotors.com/blog/engineering-update-powertrain-15

The battery can output 215kW. At nominal 3.6V*99 = 356.4V that is 603A. At minimum 3.0V*99 = 297V that is 724A.
http://my.teslamotors.com/roadster/technology/battery
http://www.teslamotors.com/blog/bit-about-batteries

Also I would agree they likely measured the motors separately. However, the spec does not specify how to do it for a power train with two motors, so perhaps they thought they can just add it together.

I don't know if they necessarily used a battery either because nowhere in the spec does it require a battery to be used as the DC source, while it does require the factory equipment to be used for the connected accessories.

Not every variant. The S85 claimes 373 hp. It makes 375 at the wheels. It's more than Tesla claimed. The 85D claims 417 hp but makes 504 at the battery. Clearly far more than claimed. The P85 claimed 470 hp (at the motor shaft) and dynos 435 hp at the wheels which is clearly more than claimed unless the drivetrain loss is less than 7%.

The S60 on the other hand claimed 380 hp and based on reported REST numbers and it's 5.9 second 0-60 is not even close to that number. Nobody is screaming about the S60 because nobody it bought for performance and they're not going to care. That doesn't make it right, but it does mean it's not going get any attention.

What I mean is the same exact scheme was used for all the variants, not that there were no variants that can make the motor power. The reason all the S85, P85, 85D variants are able to make the motor power is because the motor/inverter is obviously the bottleneck in those cases (we know from P85D that the 85kWh battery pack can support ~550hp).

In any scheme that takes into account battery limitations while measuring shaft power, S60/S70 should have less "motor power" than S85 and S60D/S70D less than S85D. However, if you look at the numbers at every given point in time, 60/70 kWh variants are always the same "motor power" as the 85kWh equivalents. I see no other way to explain it other than the motor power is not a measure that takes into account battery limitations.

I realize the other point about how "motor power" numbers have been adjusted with time: 380hp to 382hp for S60/S70/S85, 376 hp to 514hp for S60D/S70D/S85D, 470hp F / 221hp R to 503hp F / 259hp R for P85D/P90D, but that seems orthogonal to the above phenomenon. There are lots of other possible explanations for those power tweaks (Tesla underestimating power at first test, new firmware that pushes more power from same motors, or changes in assumed test parameters).

 

[Andyw2100]

Also I would agree they likely measured the motors separately. However, the spec does not specify how to do it for a power train with two motors, so perhaps they thought they can just add it together.

I don't know if they necessarily used a battery either because nowhere in the spec does it require a battery to be used as the DC source, while it does require the factory equipment to be used for the connected accessories.

Don't you see how what you are suggesting may have happened makes Tesla look much, MUCH worse than what most of us "complaining" about the missing horsepower are suggesting probably happened?

We're suggesting that Tesla had the best of intentions, and then just couldn't meet the HP specifications without some new hardware (though it's yet to be seen whether even with the new hardware the HP specs will be met, but in any case, Tesla will come closer to meeting them.) On the other hand, you seem to be suggesting that Tesla was attempting to test the motors in any way possible to make the numbers come out as high as possible, while still meeting whatever rules may or may not exist with respect to testing parameters, EVEN THOUGH THEY KNEW THE MOTORS WOULD NEVER BE ABLE TO MAKE THAT AMOUNT OF POWER AT ONE TIME IN THE CAR, AS DELIVERED.

Can't you see how that makes Tesla look much worse?

In our scenario, Tesla just made a mistake in coming up short, and now is making additional mistakes with respect to how to deal with the initial mistake. But in your scenario Tesla had the intention to mislead people from the outset, with engine specs they knew the cars would not make.

As much as I've come to trust Tesla less and less over the last few months, over a variety of issues, I have not come to think so little of them that I could conceive of them attempting to purposely mislead people and trick them into buying their cars. But that is essentially what you're suggesting they did by suggesting they were trying to push the limits of the testing parameters, while knowing the parameters bore little relation to what the reality of the situation was.

Again, luckily no one from Tesla has attempted to make the argument that you are making. I sincerely hope they never do!

 

[lolachampcar]

OMG
This thread is still breathing. This horse has died and been reincarnated several times by now.

 

[Andyw2100]

OMG
This thread is still breathing. This horse has died and been reincarnated several times by now.

I think the only way activity on this thread is going to die down is if Tesla addresses the questions about the P85D's horsepower. They've had plenty of opportunity to do so, yet they have remained silent.

 

[DFiveK2]

OMG
This thread is still breathing. This horse has died and been reincarnated several times by now.

I can't figure out if people are arguing with each other or all agree with each other and I also can't figure out what people think a good solution would be to the perceived problem. Maybe everyone can gather their thoughts and start a concise new thread.

 

[Andyw2100]

I can't figure out if people are arguing with each other or all agree with each other and I also can't figure out what people think a good solution would be to the perceived problem. Maybe everyone can gather their thoughts and start a concise new thread.

--we're still arguing with each other.
--we don't know what a good solution would be. A good start would be Tesla acknowledging the isssue.
--if a new thread was started it would take about ten minutes before there were three pages of posts all saying, "Why do we need another new thread on this?"

 

[stopcrazypp]

Don't you see how what you are suggesting may have happened makes Tesla look much, MUCH worse than what most of us "complaining" about the missing horsepower are suggesting probably happened?

We're suggesting that Tesla had the best of intentions, and then just couldn't meet the HP specifications without some new hardware (though it's yet to be seen whether even with the new hardware the HP specs will be met, but in any case, Tesla will come closer to meeting them.) On the other hand, you seem to be suggesting that Tesla was attempting to test the motors in any way possible to make the numbers come out as high as possible, while still meeting whatever rules may or may not exist with respect to testing parameters, EVEN THOUGH THEY KNEW THE MOTORS WOULD NEVER BE ABLE TO MAKE THAT AMOUNT OF POWER AT ONE TIME IN THE CAR, AS DELIVERED.

Can't you see how that makes Tesla look much worse?

It doesn't to me if they intend that "motor power" number to show the upgrade options as I suggested. I've said multiple times that with hindsight they should have two separate numbers advertised at the start in October 2014, but I don't see the issue with advertising a "motor power" number (which they are continuing to do).

Also, gearing aside (which will changed the numbers slightly) adding the two numbers together is a good rough approximation of total power. For the motors with the same gearing, it would be the correct method. If they need to take gearing into account and test two motors at the same time, they would need a motor test bench that can synchronize the two and that is a lot more complicated than just testing the motors individually (esp. given unlike hybrids, the motors don't have a shared transmission shaft but rather is synchronized through the road, so even the final drive contribution via the wheels will change the actual number).

Keep in mind that this is a motor rating, not a battery rating (perhaps they should have another line for battery rating, like Ford does).

As for not using a battery as a DC source, I think the advantage of that is it allows consistent and repeatable testing (that's probably why the spec doesn't require using a battery either). Using a battery means they have to cart in a different battery for every variant (60kWh, 70kWh, 85kWh, 85kWh Ludicrous, 90kWh), charge and discharge it to maintain the necessary voltage levels, plus the variations among individual batteries (even of the same capacity/part number) will change test results. If they use a regulated DC power supply they can set the voltage and current parameters directly and not have to worry about any of that.

None of the above reasons are sinister or say that Tesla has a motive to inflate their numbers.