New Blog: Driving Range for Model S Family....

[gpetti]

Interesting to note that the EPA numbers for the 85D and the P85D have gone up. The 85D is now showing at 270 and the P85D at 253 (both on 19" wheels). A respective improvement of 5 and 3 miles.

There's still some lack of clarity here. Were the previous EPA numbers (85D=265, P85D+19"=250, P85D+21"=242) done with the "torque sleep" software or not? My guess would be yes and that the latest numbers are an improved version of the torque sleep software...

my reading of this note in the blog is that the EPA numbers, though not finally confirme, are definitely using the new software.

1 Values for 85D and P85D are pending final confirmation from the EPA and use new dual motor torque sleep control software available by the end of January 2015.

- - - Updated - - -

If you order now you won't get delivery until March.

Actually I think it's the end of February for P85D at present.

 

[breser]

my reading of this note in the blog is that the EPA numbers, though not finally confirme, are definitely using the new software.

1 Values for 85D and P85D are pending final confirmation from the EPA and use new dual motor torque sleep control software available by the end of January 2015.

Yes that's a given for the new numbers. I was pondering if the old EPA numbers (85D/P85D+19/P85D+21 265/250/242 respectively) were also with the torque sleep software (just an older less efficient version) or if only the new numbers are with the torque sleep software.

 

[Cottonwood]

I have not seen anyone post a 100% charge yet, but with ...80 software, I did a 100% charge today, and got 249 rated miles in my P85D.

 

[NOLA_Mike]

I have not seen anyone post a 100% charge yet, but with ...80 software, I did a 100% charge today, and got 249 rated miles in my P85D.

.81 software here and 249 is my full range charge in P85D.

 

[gpetti]

Yes that's a given for the new numbers. I was pondering if the old EPA numbers (85D/P85D+19/P85D+21 265/250/242 respectively) were also with the torque sleep software (just an older less efficient version) or if only the new numbers are with the torque sleep software.

I was assuming that this included the existing numbers but that they had not received final confirmation yet. I guess I'm reading too much into this;however, given no one is coming close to this efficiency using the current software it seems extremely likely that these numbers were using the new software.

 

[Zythryn]

I think this as an "engineering number".. as a best case brand new battery, range charged, perfect 70 degree weather, no traffic, flat highway, no wind and running the battery down to below "0".

I've done a lot of driving in California on busy freeways nearish 65 and have realistically never hit anywhere near 285. I would say a realistic any condition, non-range charge, don't worry take a side jaunt range of 200.

Or am I missing something?

In my S during 70 degree days I often get close to 300 on mostly highway trips.
It makes a big difference if your "nearish" 65 is 68 or 62.

This is with 19", primacy tires on cyclone wheels, no HVAC, dry roads.

 

[skilly]

Regarding torque sleep:
The AC induction motors in Teslas have no permanent magnets. In order to create torque, current is run through a large coil of wires in the stator (stator is the "body" part of the motor that does not rotate). This current passing through the stator wires creates a magnetic field, as opposed to the field being created by magnetic materials in permanent magnet motors.

The rotor (rotating central part of the motor) also has a magnetic field created in a similar way, but done so such that the magnetic force of the rotor opposes that of the stator. This causes the rotor to begin to rotate.

As the rotor rotates, the field must be continually changed with precise timing to allow the fields to properly oppose each other and generate torque.

Point is, if you remove the current, the rotor simply freewheels--rotates freely--on its bearings. Literally all you have to do is cut the power to the coils of wire. This is equivalent to shifting the car to neutral. Since there is no clutch, the rotor always rotates depending on axle speed.

Now, since running current (in the case of a Tesla a potentially large amount of current) generates heat and therefore uses energy, switching off the power to the motor saves that energy.

One more wrinkle: as a motor spins more quickly and gets to high RPMs, it generates something called "back EMF"...basically, the faster it turns, the more opposing force the motor places upon itself. (This is why the torque band of an electric motor starts to fall off at higher RPMs). Therefore, it is more efficient to run a motor at a lower RPM (like 3,000 RPMs instead of 17,000 RPMs), not only electrically (because of back EMF) but also mechanically (less friction in the bearings). At cruising speeds, the motors in the single motor Teslas are turning at high RPMs, but little torque is needed. In the dual motor Teslas, the minimal torque needed to maintain crusing speed can be achieved by shifting the power requirement to the motor that is geared to rotate more slowly. This can literally be done in an instant. Likewise, if you ask for power by pressing the pedal, the car (on the order of a few milliseconds) can deliver power to both motors simply by turning current back on.

Also, the less powerful front motor has less wire in the coils, therefore wastes less energy when current is run through it to generate torque.

this is really helpful - thank you!! This kind of information is why I weed through all the other "stuff"!! :wink:

 

[stopcrazypp]

The cars in the hands of customers not meeting the efficiency expected with the EPA numbers is a matter of missing the torque sleep.

Seems like it's my #3 option that is the correct one. The EPA sticker currently on the customer cars are not final and the discrepancies between the MPGe and range are likely from there.

3) The MPGe numbers on the sticker do NOT correspond to the range number and either one will have to be updated in the future.
What we do know is that there is a pending change that will improve efficiency, and that "normal" mode will return (probably referring to the same change), which might be related to this.

 

[gpetti]

I'm not sure what your point is. Sure, it's all going to work out, but why in the world did Tesla not just put this information out before they started delivering cars? I posted the following on the blog entry, immediately after reading it:

--
This is excellent information, and very, VERY good news for current and future P85D owners. It is presented quite clearly, and is certainly easy to understand. Thank you.
I would like to add, though, that Tesla could have prevented a lot of worry and concern being felt by new P85D owners over range numbers and energy usage figures that we were seeing that were not in line with our expectations if Tesla had simply made us aware that this January software update was in the works, and that our vehicles were not yet as efficient as they would be. Instead, we were given our key fobs, and our fantastic 0-60, 3.2 second Insane Mode masterpieces of technology, with not so much as a word of warning as to the fact that we should not yet expect to see the range or energy usage that was on the window stickers that came with our cars, or that had been on the Tesla Motors site when we ordered our cars. Frankly, this caused a great deal of anxiety and worry. At a time we all should have been exuberantly enjoying our brand new wonder-cars, many of us were wringing our hands, wondering if they were ever going to achieve the range we expected them to. This just wasn't necessary. I thank you, sincerely, for providing this information now, but can't say strongly enough how much better it would have been if the information had been offered a month ago.
Thanks again for building great cars!
--

Tesla told us, at the D announcement, that we would not have most of the autopilot features at the time our cars were delivered. And I don't think you saw anyone here complaining about a lack of autopilot features in their cars. (If you did, there certainly were not very many people complaining about that.) The issue here wasn't so much the lack of range, but rather the lack of communication. My guess is that Tesla didn't communicate this issue at the D announcement the way they did the Autopilot features because they hoped to have the software in the car at the time the cars were delivered. Great. That's admirable. But at some point, someone realized that wasn't going to happen. My concern is for whatever decision process kicked in at that point and resulted in the decision not to immediately put out the information they put out today. Clearly they had it available. This is the issue.

The lesson that should be learned is that the Tesla early adopters are a very understanding group, and will put up with a great deal, but we also want to be leveled with. Provide us with honest information, and we'll be Tesla's best ambassadors. Why wouldn't Tesla want to do that?

+ 1

 

[miimura]

Reversing the loading on the gears happens every time you go into regen anyway - I don't see why it would be any more of an issue here.

The only issue I can see with just dropping power to the inverter entirely is when you go to turn the motor back on. If there isn't an independent motor shaft position sensor, the phase response from the motor is the only way to know the phase angles, so if you turn the fields off you'd be unable to know exactly where to start them up again. If the car had to guess and guessed wrong, it would result in a sudden surge of acceleration or deceleration - I think.
Walter

There is no guessing when you are commutating a motor. There are Hall Effect sensors that sense the magnetic phase of the motor at each instant. Based on that information, the inverter can synchronize the phase and velocity of the already spinning motor.

 

[EarlyAdopter]

There is no guessing when you are commutating a motor. There are Hall Effect sensors that sense the magnetic phase of the motor at each instant. Based on that information, the inverter can synchronize the phase and velocity of the already spinning motor.

But how can Hall Effect sensors detect the magnetic phase when there is no current in either set of coils during "torque sleep" and hence no magnetic field to detect?

Perhaps some of the engineering challenge at hand is in leaving just enough current in the coils to sustain a detectable field to keep the phase aligned but not so much that you induce torque.

 

[Saghost]

There is no guessing when you are commutating a motor. There are Hall Effect sensors that sense the magnetic phase of the motor at each instant. Based on that information, the inverter can synchronize the phase and velocity of the already spinning motor.

Tesla uses induction motors. There are no permanent magnets, there are no brushes or commutators. As far as I know, there's no external phase angle sensor, hall effect or otherwise - and a hall effect sensor wouldn't read without a field generated by the inverter on an induction motor.
Walter

 

[Dan_]

Something I dont understand about the range of P85D.
Elon said that dual motor can help to improve of efficiency of the MS, thats why S85D has a better range than S85. (270mi vs 265mi)
But for P85D, its range is actually lower than P85 / S85 / P85+

Does anyone know what causes the decrease in efficiency (or range) of P85D?

Variant	EPA 5-cycle Range [miles]
85D	270
P85+, P85, 85	265
P85D	253
60	208

 

[Raffy.Roma]

@Dan

It's not a decrease of efficiency or range of the P85D. If you go and see in the previous blogs how the EPA 5-cycle range works you will see that there is a test involving acceleration. Now since the P85D has a tremendous acceleration (much more that the 85D) of course the range get decreased with respect to the 85D.

 

[breser]

It's not a decrease of efficiency or range of the P85D. If you go and see in the previous blogs how the EPA 5-cycle range works you will see that there is a test involving acceleration. Now since the P85D has a tremendous acceleration (much more that the 85D) of course the range get decreased with respect to the 85D.

The EPA test cycles specify the acceleration rates. The P85D's ability to accelerate at a high rate is not tested in the EPA tests. I'd go so far as to say that even an S60's full acceleration is not tested in these test cycles. Remember these are laboratory tests designed to be repeatable on all cars. Using high end capabilities like the P85D has would defeat that purpose.

This is not to say that accelerating quickly will not diminish the range of the P85D, it's just that this isn't reflected in the tests. If you do pedal to the floor accelerations you will certainly not get the EPA rated range. But that's true of all cars.

 

[Dan_]

@Dan

It's not a decrease of efficiency or range of the P85D. If you go and see in the previous blogs how the EPA 5-cycle range works you will see that there is a test involving acceleration. Now since the P85D has a tremendous acceleration (much more that the 85D) of course the range get decreased with respect to the 85D.

The EPA test cycles specify the acceleration rates. The P85D's ability to accelerate at a high rate is not tested in the EPA tests. I'd go so far as to say that even an S60's full acceleration is not tested in these test cycles. Remember these are laboratory tests designed to be repeatable on all cars. Using high end capabilities like the P85D has would defeat that purpose.

This is not to say that accelerating quickly will not diminish the range of the P85D, it's just that this isn't reflected in the tests. If you do pedal to the floor accelerations you will certainly not get the EPA rated range. But that's true of all cars.

I understand that the vehicle have different acceleration. But putting the acceleration factor aside, by looking at the table of the cursing range below, there are still no improvement over the rated range of P85D over P85/P85+.

From what I understand, by having a second motor at the front, it is supposed to improve the grip of the vehicle and therefore a better range. Any thoughts?

Variant	65 mph Range [miles]	75 mph Range [miles]
85D	295	249
P85+, P85, 85	285	242
P85D	285	240

 

[jpet]

Something I dont understand about the range of P85D.
Elon said that dual motor can help to improve of efficiency of the MS, thats why S85D has a better range than S85. (270mi vs 265mi)
But for P85D, its range is actually lower than P85 / S85 / P85+

Does anyone know what causes the decrease in efficiency (or range) of P85D?

If I understand it well, the S85D uses 2 new gen motors that are smaller and more efficient than the larger single motor in the S85.
The P85D only has one new gen motor and hence does not have the same efficiency benefit. That's a price to pay for the performance Model S.
You can't have it all but torque sleep will allow running the P85D on the more efficient front motor only, providing you can keep your foot under control.

I'm wondering if torque sleep will be available when not using cruise control...

 

[ScepticMatt]

I'm wondering if torque sleep will be available when not using cruise control...

It should be available as long as you don't use more power than what the front motor can provide.

 

[Todd Burch]

If I understand it well, the S85D uses 2 new gen motors that are smaller and more efficient than the larger single motor in the S85.
The P85D only has one new gen motor and hence does not have the same efficiency benefit. That's a price to pay for the performance Model S.
You can't have it all but torque sleep will allow running the P85D on the more efficient front motor only, providing you can keep your foot under control.

I'm wondering if torque sleep will be available when not using cruise control...

This sounds correct. As jpet said, the S85D uses the two smaller motors, which are more efficient than the one larger motor.

- - - Updated - - -

It should be available as long as you don't use more power than what the front motor can provide.

I agree...the car should be able to use torque sleep at any time, unless you demand more than 200ish-whatever horsepower than the front motor can provide. As far as I can tell, putting the rear motor to sleep should involve not much more than cutting power to the motor (which we do every time our pedal reaches the equilibrium point between regen and power generation). This can be done with no mechanical wear and in a matter of a few milliseconds.

 

[ScepticMatt]

I agree...the car should be able to use torque sleep at any time, unless you demand more than 200ish-whatever horsepower than the front motor can provide.

Well not quite "any" time. You still want awd in high traction situations (i.e. lowish speeds, cornering, jolts)
Otherwise the car would drive like a fwd car at low power levels.

As far as I can tell, putting the rear motor to sleep should involve not much more than cutting power to the motor. This can be done with no mechanical wear and in a matter of a few milliseconds.

While the end result is correct, you can only cut power once the motors is not under load.
So it's more like, detect cruising->shift power to front->cut power from back.