Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Will Elon ever give us a software unlock for the LR AWD?

  • Yes

    Votes: 35 32.7%
  • No

    Votes: 72 67.3%

  • Total voters
    107
This site may earn commission on affiliate links.

MountainPass

Active Member
Global Vendor
Mar 2, 2018
2,103
4,458
Toronto, Canada
Listings
SOC-Featured-Image.jpg


We finally got an AWD car on our dyno (thanks to our friend Shereef for lending us his P3D) and it was great to find that these cars will work with a Dynapack hub dyno without issue! After seeing the difference between the standard and long-range RWD Model 3, we were very interested to see how much power loss would occur relative to battery state of charge with the massive amount of power the P3D makes.

Unfortunately, we quickly found out that the car exceeds the limits of our dyno! The front pods of our dyno are able to absorb 2000Nm of torque, and this is just below what the front motor of the P3D puts out. After maxing out the dyno and getting the red screen of death a few times, I had to carefully ramp the power in at the start of the pull until the front motor went into field weakening, at which point it was possible to see the full power output. So when you’re looking at these graphs below, understand that the torque below field weakening (where the torque is flat) would have been slightly higher – around 5% more. The peak power would have also been slightly higher, close to 5%.

Any differences that show on the dyno graph in the “flat torque” region, specifically on the 45% SOC pull, are because I was slowly ramping my foot into the throttle, and not due to the battery state of charge. The torque would have been around 500lb-ft (4865lb-ft at the axles) on each run up until the field weakening point. If you’re not sure what field weakening is, read about it here!

As we suspected, as the battery SOC drops the power drops off significantly, far more than the RWD cars. This is because the voltage sag is significantly greater with the much larger amount of current being consumed by both motors.

Tesla-Model-3-P3D-SOC-Dyno-Test.png


Interestingly between 95% to 75% there isn’t a lot of power drop off – around 17 horsepower. We can assume the power at 100% would be very similar to 95%, perhaps a few horsepower more. Call it 20 horsepower for 25% SOC drop.

The drop from 75% to 45% – a 30% drop in SOC (only 5% more than the example above) – results in a massive 64 horsepower difference! So it’s pretty clear that if you’re trying to put in some good times on the track, don’t bother unless your SOC is above 75%.

That being said, similar to the RWD Model 3, if you’re just talking about power below 70km/h (45mph), this remains fairly constant regardless of the battery state of charge.

You can watch the video below to hear the dyno pods scream in agony and also get some information about the torque split front to rear (it’s around 66% rear and 33% front). Check it out here:

"
 
Given that Tesla rates the power of the rear motors the same for RWD and P models, is it possible that the regular AWD model is getting motors "binned" for not being as good? Front motors are rated the same. Or is it all just software?
 
Given that Tesla rates the power of the rear motors the same for RWD and P models, is it possible that the regular AWD model is getting motors "binned" for not being as good? Front motors are rated the same. Or is it all just software?

All software, at least up until earlier this year when most AWD cars began being delivered with a different (by PN) rear motor.

Even as recently as a month or two ago though we had reports of folks going to pick up a P3D- out of inventory and finding an AWD car (but with the original rear motor PN in it) waiting for them, that got software updated to P...sometimes right then sometimes a couple days later via OTA update.
 
@MountainPass

Couple questions: (people already asked about the firmware version ...was this 36.1?)

1) At 3:00 in the video - you say 95% is 465HP (and that is what is shown on the dyno screen marker). This does not align with your plots. (95% SoC in your plots above shows 485HP if I am not mistaken!) From street testing, we expect closer to 455-460 HP to the wheels at that state of charge - with version 36.1 - though there is reason to think that may have been a slightly low result due to battery condition/temperature, so wondering what the deal is with the right-hand y-axis. I kind of wonder whether something happened when moving all the curves to the same graph with same axes on both sides (there is no difference in the left and right y-axes as it turns out...)

2) Do you account for any losses when plotting the results, or is this all "at the wheel" measurements? (I suspect it is all at the wheel, notwithstanding the apparent scale/plotting issue from point 1 ) . Obviously the torque is converted - could you explain the scaling you used (9:04:1, etc.) to get it the torque translated back to your plotted results?

3) You should consider generating a plot of motor efficiency vs. load & speed! (You'd need to log data from the CAN concurrent with the dyno output, so maybe it would be tricky.) ;)
 
Last edited:
All software, at least up until earlier this year when most AWD cars began being delivered with a different (by PN) rear motor.

Even as recently as a month or two ago though we had reports of folks going to pick up a P3D- out of inventory and finding an AWD car (but with the original rear motor PN in it) waiting for them, that got software updated to P...sometimes right then sometimes a couple days later via OTA update.

And then there's even people who ordered the AWD, got delivered by Tesla what they thought was an AWD, but actually received a P3D-. This guy in this Tesla forum thread here took his AWD to the track and was baffled how he was getting 11.8's in the quarter mile, and even provided time slips verifying such.
 
  • Like
Reactions: Knightshade
I have heard claims that the power is pulled less at lower SoC when in Track Mode, so it would be an interesting thing to check, to see if that is true.

According to this (Skip to around 10:45 into the video):


The water pumps running at around 35% of max flow burn about 0.3kW along with the MCU and screen being on. So any kind of parasitic loss because of track mode would be insignificant.
 
According to this (Skip to around 10:45 into the video):


The water pumps running at around 35% of max flow burn about 0.3kW along with the MCU and screen being on. So any kind of parasitic loss because of track mode would be insignificant.

I was not suggesting that the Track Mode features would be parasitic in any significant way. I was saying that they may allow a higher discharge rate when in Track Mode, even at lower SoC. No idea whether it is true, or physically whether it is possible. With lower voltage at lower SOC you’d expect some power reduction would be inevitable, though to some extent you might be able to throw more current at the problem.

But mostly I want to get confirmation that the max power measured here was 465HP and with which software it was done.
 
I would imagine it is physically possible for the AWD variant to sustain constant output for a good portion of the packs range. But this is much less likely on the performance. The voltage sag under load isn’t that bad on the 3P though, I’d say most likely it could sustain the power requirements, presumably there’s a sensible reason why it does not.

I’d like to see how pack temperature impacts performance as well. I’ve been waking up to a car full of regen dots lately, and it’s slow for the first few punches. After a bit of hammering and regen it appears to be back to normal for the next several hours.
 
@MountainPass

Couple questions: (people already asked about the firmware version ...was this 36.1?)

1) At 3:00 in the video - you say 95% is 465HP (and that is what is shown on the dyno screen marker). This does not align with your plots. (95% SoC in your plots above shows 485HP if I am not mistaken!)

Pretty sure the part in his post where he said it would be about 5% more peak power than what the graph shows due to dyno issues should explain this.
 
Pretty sure the part in his post where he said it would be about 5% more peak power than what the graph shows due to dyno issues should explain this.

The peak power may well have been higher without those issues, but they did not fabricate data - they measured a maximum of 465HP on their dyno, which may have been a number with some field weakening. At least that is my understanding. Though the peak HP is achieved just beyond field weakening onset so maybe not much higher.

But in any case, if you watch the video, the line they use the cursor on at 3:00 looks like it matches the 95% curve above. Thus was 465HP peak. Maybe it does not match; maybe it is a different dyno pull or something. But mostly it just looks like the HP axis in the plot in the OP is not right.