SOC vs power and torque output

[insaneoctane]

Pixel peeking reveals:

HP vs SOC:

SOC    HP
90    325
75    319.5
60    302.7
45    294.1
30    281.4
15    269.5

MPH where TQ drops below 316 lb-ft:

SOC    MPH
90    44.7
75    43.7
60    41.6
45    40.5
30    38.8
15    36.7

 

[D3xDt3Reaction]

It's interesting that the power drops off so much at higher RPM at all SOCs. For example the power drops from a peak of 325hp to 250hp at 11,000rpm at 90% SOC. Presumably this is due to the inverter/motor, not the max current out fo the battery, or it would indicate a massive loss in efficiency (and would require huge waste heat dissipation capabilities!).
When are you guys getting a P3D?

Seems like the drop off is probably just due to the properties of electric motors: when the motor is spinning it acts like a generator and creates a back EMF. At some point the EMF becomes large enough that the current falls off as the battery can’t “keep up”: current = (battery voltage - back EMF) / winding resistance.

In the “flat” part of the curve it seems like the drive unit is in a current limiting mode based on motor capability and traction (on the road). So basically I think we’re transitioning from being motor limited at low speeds to being battery limited around 45mph.

Someone more knowledgeable can correct me if I got it wrong.

 

[Daniel in SD]

Seems like the drop off is probably just due to the properties of electric motors: when the motor is spinning it acts like a generator and creates a back EMF. At some point the EMF becomes large enough that the current falls off as the battery can’t “keep up”: current = (battery voltage - back EMF) / winding resistance.

In the “flat” part of the curve it seems like the drive unit is in a current limiting mode based on motor capability and traction (on the road). So basically I think we’re transitioning from being motor limited at low speeds to being battery limited around 45mph.

Someone more knowledgeable can correct me if I got it wrong.

From what I can tell from a cursory Google search is that the maximum power drops off but the efficiency does not drop off. So what I'm having trouble reconciling is why the power is dropping off even at low SOC. It seems like the power would be flat at low SOCs instead of dropping off if it were being limited by the battery.

 

[hoang51]

I wish we had 0-60 and 1/4 mile times at each SOC to turn this graph into something that more people understand quickly. Don't get me wrong, I love the graph....

I'm motivated to do this in the name of science. I have Dragy and a location nearby for the runs. I just need to negotiate with the wife and my 11 month old son for free time. And maybe pray to the Supercharger gods that they won't ban me from going back and forth powering up many times.

 

[insaneoctane]

I'm motivated to do this in the name of science. I have Dragy and a location nearby for the runs. I just need to negotiate with the wife and my 11 month old son for free time. And maybe pray to the Supercharger gods that they won't ban me from going back and forth powering up many times.

Great! Can't wait to see your results. Be sure that the battery and ambient temperatures are similar and recorded. Not sure where to see battery temperature at the moment I will look into that.

 

[D3xDt3Reaction]

From what I can tell from a cursory Google search is that the maximum power drops off but the efficiency does not drop off. So what I'm having trouble reconciling is why the power is dropping off even at low SOC. It seems like the power would be flat at low SOCs instead of dropping off if it were being limited by the battery.

Sorry if I wasn’t clear. My comment above was only about back EMF at increasing RPM and wasn’t related to SOC.

However, at a lower SOC, the battery voltage is lower so we should transition from the motor limited state to the battery limited state at a lower RPM. The battery can’t provide as much current since the max current is related to battery voltage and internal resistance.

 

[gavine]

Power (watts) = volts x amps. At lower states of charge, battery voltage is lower, thus less total power. As I understand it, Tesla increases the amps at lower states of charge in order to try and maintain a consistent power output (to a certain point of course).

 

[D3xDt3Reaction]

Power (watts) = volts x amps. At lower states of charge, battery voltage is lower, thus less total power. As I understand it, Tesla increases the amps at lower states of charge in order to try and maintain a consistent power output (to a certain point of course).

Yes, and that point should be where the current is constrained by (battery voltage divided by battery internal resistance) as opposed to max motor current.

 

[hoang51]

Great! Can't wait to see your results. Be sure that the battery and ambient temperatures are similar and recorded. Not sure where to see battery temperature at the moment I will look into that.

Lol, you'll be forced to wait not because of me but getting out of the house to do these runs take great timing skills plus I need to do it at night when my son is long asleep, among other things. Hard but not impossible. Just don't know when it can be done. First step for me is to have the battery SOC to maybe 30% and work my way up to 90% at 15% increments, making it 5 runs total.

 

[MountainPass]

Lol, you'll be forced to wait not because of me but getting out of the house to do these runs take great timing skills plus I need to do it at night when my son is long asleep, among other things. Hard but not impossible. Just don't know when it can be done. First step for me is to have the battery SOC to maybe 30% and work my way up to 90% at 15% increments, making it 5 runs total.

That's what we did, it would have been too involved to drain the battery between pulls, it was easier just to charge it up. Either way, it was a full day of work! Hopefully, your charging situation will be better than ours in the shop.

 

[hoang51]

Here are my scientific test results from my 2018 Tesla Model 3 Performance with Performance Upgrade (e.g., 20" rims, bigger brakes)...

First off, the settings: AC and fan off. All windows up. Fog & headlights on. Acceleration on sport. Steering on standard. Regen brake on standard. Traction control slip start off. Creep on. USB music on. Car curb weight is 4072 lbs. My weight is about 148.4 lbs. Child seat Maxi Cosi weight of 17.5 lbs in the back seat (absent of child!). About 4237.9 lbs total, give or take a few pounds. No spoiler and no badge (no extra 50 HP?). Car's firmware/software is 2018.32.2 3817fdd. I used Dragy 2.1.8 (app version) to record my timing runs, so no video recording bullshit of counting frames. I set Dragy to collect data @ 25, 25, 45, 55, and 65 MPH in addition to what were normally collected at 10, 20, 30, 40, 50, 60 MPH, with 1/8 and 1/4 mile.

Disclaimers: These runs were generally done near the same launch area for the stretch of the road. Not exactly the same as every time for quarter mile runs, because some results, Dragy "invalidated" them (valid threshold is less than -1.0% slope). Out of the 5 runs total, 3 were valid. The two invalid runs were due to slopes at -1.02% and -1.14%. Meh, I'll consider that as negligible for scientific results in this presentation. All valid runs were under -0.9% slope. I did not get a chance to retrieve battery temperature as requested by @insaneoctane , but at least I got the ambient temperature. I couldn't get 5 runs to have 5 results (i.e., 20%, 40% 60%, 80%, and 100% SOC) as originally planned due to running late and short on time (family time robbed it from me).

I did these ninja runs at night about more than a week ago. Please don't ask where and when specifically, it's going to remain a secret. Ambient temperature was in about the low 70s. Winds was listed calm, about 1 MPH.

First run: 71 miles, ~22.9% SOC, reaching 1/4 mile felt so slow and was incomplete, didn't even break triple digits because I saw a car on the road ahead of me, I had to slow down before reaching the car. 63 miles SOC when back at Supercharger with AC on at 74 deg. I decided to throw this result out in favor of a re-test on a different day.

Second run: 153 miles, ~49.4% SOC. 145 miles SOC when back to Supercharger with AC on at 74 deg. Did not feel the nerfed power like the first run. Used 5.8 mi, 2 kWh, 396 Wh/mi for the run loop.

Third run: 231 miles, ~74.5.% SOC. Braking after achieving quarter mile... Slowing down isn't going to happen quickly, which is pretty much expected for a heavy car in momentum. 222 miles when back to Supercharger with AC on. Used 5.8 mi, 3 kWh, 480 Wh/mi for the run loop. By this time, wife wants me home...

Fourth run: 272, ~87.7% SOC. Good thing that this last run is on the way home.

Re-run of first run on within last week in favorable weather terms (I almost ran it on a damp night). This time with 68 miles, ~21.9% SOC. Similar feeling as the first first run - felt slow compared to higher SOC, but still faster than my ICE cars. No music this time. Ambient temperature was in about the low 70s. Winds was listed calm, about 1 MPH. Results was a small improvement over the original first run, but not by much. Still as expected. This result superseded the original first run.

Notes: Each run depleted about 5-6 miles off of the reported battery range. So that's about 2% each 1/4 mile run. I still love the torque this car delivers every time I step on the acceleration pedal hard and fast. Sorry Honda VTEC, you got nothing on electric motors to snap people's neck back into the seat.

Observation: Concurs with others before me - in order to get the best times, SOC needs to be towards the full battery capacity. Otherwise, car will start to get slower times.

 

[insaneoctane]

Here are my scientific test results from my 2018 Tesla Model 3 Performance with Performance Upgrade (e.g., 20" rims, bigger brakes)...

First off, the settings: AC and fan off. All windows up. Fog & headlights on. Acceleration on sport. Steering on standard. Regen brake on standard. Traction control slip start off. Creep on. USB music on. Car curb weight is 4072 lbs. My weight is about 148.4 lbs. Child seat Maxi Cosi weight of 17.5 lbs in the back seat (absent of child!). About 4237.9 lbs total, give or take a few pounds. No spoiler and no badge (no extra 50 HP?). Car's firmware/software is 2018.32.2 3817fdd. I used Dragy 2.1.8 (app version) to record my timing runs, so no video recording bullshit of counting frames. I set Dragy to collect data @ 25, 25, 45, 55, and 65 MPH in addition to what were normally collected at 10, 20, 30, 40, 50, 60 MPH, with 1/8 and 1/4 mile.

Disclaimers: These runs were generally done near the same launch area for the stretch of the road. Not exactly the same as every time for quarter mile runs, because some results, Dragy "invalidated" them (valid threshold is less than -1.0% slope). Out of the 5 runs total, 3 were valid. The two invalid runs were due to slopes at -1.02% and -1.14%. Meh, I'll consider that as negligible for scientific results in this presentation. All valid runs were under -0.9% slope. I did not get a chance to retrieve battery temperature as requested by @insaneoctane , but at least I got the ambient temperature. I couldn't get 5 runs to have 5 results (i.e., 20%, 40% 60%, 80%, and 100% SOC) as originally planned due to running late and short on time (family time robbed it from me).

I did these ninja runs at night about more than a week ago. Please don't ask where and when specifically, it's going to remain a secret. Ambient temperature was in about the low 70s. Winds was listed calm, about 1 MPH.

First run: 71 miles, ~22.9% SOC, reaching 1/4 mile felt so slow and was incomplete, didn't even break triple digits because I saw a car on the road ahead of me, I had to slow down before reaching the car. 63 miles SOC when back at Supercharger with AC on at 74 deg. I decided to throw this result out in favor of a re-test on a different day.

Second run: 153 miles, ~49.4% SOC. 145 miles SOC when back to Supercharger with AC on at 74 deg. Did not feel the nerfed power like the first run. Used 5.8 mi, 2 kWh, 396 Wh/mi for the run loop.

Third run: 231 miles, ~74.5.% SOC. Braking after achieving quarter mile... Slowing down isn't going to happen quickly, which is pretty much expected for a heavy car in momentum. 222 miles when back to Supercharger with AC on. Used 5.8 mi, 3 kWh, 480 Wh/mi for the run loop. By this time, wife wants me home...

Fourth run: 272, ~87.7% SOC. Good thing that this last run is on the way home.

Re-run of first run on within last week in favorable weather terms (I almost ran it on a damp night). This time with 68 miles, ~21.9% SOC. Similar feeling as the first first run - felt slow compared to higher SOC, but still faster than my ICE cars. No music this time. Ambient temperature was in about the low 70s. Winds was listed calm, about 1 MPH. Results was a small improvement over the original first run, but not by much. Still as expected. This result superseded the original first run.

Notes: Each run depleted about 5-6 miles off of the reported battery range. So that's about 2% each 1/4 mile run. I still love the torque this car delivers every time I step on the acceleration pedal hard and fast. Sorry Honda VTEC, you got nothing on electric motors to snap people's neck back into the seat.

Observation: Concurs with others before me - in order to get the best times, SOC needs to be towards the full battery capacity. Otherwise, car will start to get slower times.View attachment 332863 View attachment 332864

Other than the first, low SOC, it looks really similar

 

[CrazyCoconut]

Here are my scientific test results from my 2018 Tesla Model 3 Performance with Performance Upgrade (e.g., 20" rims, bigger brakes)...

First off, the settings: AC and fan off. All windows up. Fog & headlights on. Acceleration on sport. Steering on standard. Regen brake on standard. Traction control slip start off. Creep on. USB music on. Car curb weight is 4072 lbs. My weight is about 148.4 lbs. Child seat Maxi Cosi weight of 17.5 lbs in the back seat (absent of child!). About 4237.9 lbs total, give or take a few pounds. No spoiler and no badge (no extra 50 HP?). Car's firmware/software is 2018.32.2 3817fdd. I used Dragy 2.1.8 (app version) to record my timing runs, so no video recording bullshit of counting frames. I set Dragy to collect data @ 25, 25, 45, 55, and 65 MPH in addition to what were normally collected at 10, 20, 30, 40, 50, 60 MPH, with 1/8 and 1/4 mile.

Disclaimers: These runs were generally done near the same launch area for the stretch of the road. Not exactly the same as every time for quarter mile runs, because some results, Dragy "invalidated" them (valid threshold is less than -1.0% slope). Out of the 5 runs total, 3 were valid. The two invalid runs were due to slopes at -1.02% and -1.14%. Meh, I'll consider that as negligible for scientific results in this presentation. All valid runs were under -0.9% slope. I did not get a chance to retrieve battery temperature as requested by @insaneoctane , but at least I got the ambient temperature. I couldn't get 5 runs to have 5 results (i.e., 20%, 40% 60%, 80%, and 100% SOC) as originally planned due to running late and short on time (family time robbed it from me).

I did these ninja runs at night about more than a week ago. Please don't ask where and when specifically, it's going to remain a secret. Ambient temperature was in about the low 70s. Winds was listed calm, about 1 MPH.

First run: 71 miles, ~22.9% SOC, reaching 1/4 mile felt so slow and was incomplete, didn't even break triple digits because I saw a car on the road ahead of me, I had to slow down before reaching the car. 63 miles SOC when back at Supercharger with AC on at 74 deg. I decided to throw this result out in favor of a re-test on a different day.

Second run: 153 miles, ~49.4% SOC. 145 miles SOC when back to Supercharger with AC on at 74 deg. Did not feel the nerfed power like the first run. Used 5.8 mi, 2 kWh, 396 Wh/mi for the run loop.

Third run: 231 miles, ~74.5.% SOC. Braking after achieving quarter mile... Slowing down isn't going to happen quickly, which is pretty much expected for a heavy car in momentum. 222 miles when back to Supercharger with AC on. Used 5.8 mi, 3 kWh, 480 Wh/mi for the run loop. By this time, wife wants me home...

Fourth run: 272, ~87.7% SOC. Good thing that this last run is on the way home.

Re-run of first run on within last week in favorable weather terms (I almost ran it on a damp night). This time with 68 miles, ~21.9% SOC. Similar feeling as the first first run - felt slow compared to higher SOC, but still faster than my ICE cars. No music this time. Ambient temperature was in about the low 70s. Winds was listed calm, about 1 MPH. Results was a small improvement over the original first run, but not by much. Still as expected. This result superseded the original first run.

Notes: Each run depleted about 5-6 miles off of the reported battery range. So that's about 2% each 1/4 mile run. I still love the torque this car delivers every time I step on the acceleration pedal hard and fast. Sorry Honda VTEC, you got nothing on electric motors to snap people's neck back into the seat.

Observation: Concurs with others before me - in order to get the best times, SOC needs to be towards the full battery capacity. Otherwise, car will start to get slower times.View attachment 332863 View attachment 332864

Thanks for the doing the tests! Great info!

 

[dfwatt]

Someone did dyno testing on the Model 3 both RWD LR and AWD P model, with some very similar looking results but I can't find it anymore? Does anybody know where on the forum it might be?

 

[FlyNavy01]

Someone did dyno testing on the Model 3 both RWD LR and AWD P model, with some very similar looking results but I can't find it anymore? Does anybody know where on the forum it might be?

It was AWD vs Performance. The video can be found here:

Looks like the P3D was making ~465 awhp at peak which is pretty damn good.

 

[dfwatt]

It was AWD vs Performance. The video can be found here:

Looks like the P3D was making 466 awhp at peak which is pretty damn good.

That's a great catch but I've actually never seen that before this was a printed graph may be done by him or by somebody else independently. Actually showed a peak horsepower of 475. The quoted 450 looks like it's an average across a bunch of RPMs that one might expect to see.

 

[FlyNavy01]

That's a great catch but I've actually never seen that before this was a printed graph may be done by him or by somebody else independently. Actually showed a peak horsepower of 475. The quoted 450 looks like it's an average across a bunch of RPMs that one might expect to see.

This is the only one I've ever seen. From the same source as above.

 

[dfwatt]

Thanks!

 

[dfwatt]

It was AWD vs Performance. The video can be found here:

Looks like the P3D was making ~465 awhp at peak which is pretty damn good.

What's interesting is that this is an HP rating at the wheels meaning that there is still some parasitic losses in the drivetrain, albeit probably far less than a comparable ICE car. Direct drive (lack of a true transmission) and particularly the lack of any awkward 90 degree driveshaft torque translation probably means that parasitic loss is only a few percent (guessing ~4-8% tops) vs. ICE RWD at typically 10-18% loss and ICE AWD a whopping 17-25% loss. This means that the Model 3 Performance version is generating the equivalent of 540 to 575 horsepower "at the flywheel" from a comparable ICE AWD. No wonder this thing romper-stomps ICE cars outside of exotics and hugely supercharged V8s! Its only relative weakness is when the speeds get above 60 mph and motor RPMs rise to the point where peak HP starts to decline.

 

[Frank89]

Here are my scientific test results from my 2018 Tesla Model 3 Performance with Performance Upgrade (e.g., 20" rims, bigger brakes)...

First off, the settings: AC and fan off. All windows up. Fog & headlights on. Acceleration on sport. Steering on standard. Regen brake on standard. Traction control slip start off. Creep on. USB music on. Car curb weight is 4072 lbs. My weight is about 148.4 lbs. Child seat Maxi Cosi weight of 17.5 lbs in the back seat (absent of child!). About 4237.9 lbs total, give or take a few pounds. No spoiler and no badge (no extra 50 HP?). Car's firmware/software is 2018.32.2 3817fdd. I used Dragy 2.1.8 (app version) to record my timing runs, so no video recording bullshit of counting frames. I set Dragy to collect data @ 25, 25, 45, 55, and 65 MPH in addition to what were normally collected at 10, 20, 30, 40, 50, 60 MPH, with 1/8 and 1/4 mile.

Disclaimers: These runs were generally done near the same launch area for the stretch of the road. Not exactly the same as every time for quarter mile runs, because some results, Dragy "invalidated" them (valid threshold is less than -1.0% slope). Out of the 5 runs total, 3 were valid. The two invalid runs were due to slopes at -1.02% and -1.14%. Meh, I'll consider that as negligible for scientific results in this presentation. All valid runs were under -0.9% slope. I did not get a chance to retrieve battery temperature as requested by @insaneoctane , but at least I got the ambient temperature. I couldn't get 5 runs to have 5 results (i.e., 20%, 40% 60%, 80%, and 100% SOC) as originally planned due to running late and short on time (family time robbed it from me).

I did these ninja runs at night about more than a week ago. Please don't ask where and when specifically, it's going to remain a secret. Ambient temperature was in about the low 70s. Winds was listed calm, about 1 MPH.

First run: 71 miles, ~22.9% SOC, reaching 1/4 mile felt so slow and was incomplete, didn't even break triple digits because I saw a car on the road ahead of me, I had to slow down before reaching the car. 63 miles SOC when back at Supercharger with AC on at 74 deg. I decided to throw this result out in favor of a re-test on a different day.

Second run: 153 miles, ~49.4% SOC. 145 miles SOC when back to Supercharger with AC on at 74 deg. Did not feel the nerfed power like the first run. Used 5.8 mi, 2 kWh, 396 Wh/mi for the run loop.

Third run: 231 miles, ~74.5.% SOC. Braking after achieving quarter mile... Slowing down isn't going to happen quickly, which is pretty much expected for a heavy car in momentum. 222 miles when back to Supercharger with AC on. Used 5.8 mi, 3 kWh, 480 Wh/mi for the run loop. By this time, wife wants me home...

Fourth run: 272, ~87.7% SOC. Good thing that this last run is on the way home.

Re-run of first run on within last week in favorable weather terms (I almost ran it on a damp night). This time with 68 miles, ~21.9% SOC. Similar feeling as the first first run - felt slow compared to higher SOC, but still faster than my ICE cars. No music this time. Ambient temperature was in about the low 70s. Winds was listed calm, about 1 MPH. Results was a small improvement over the original first run, but not by much. Still as expected. This result superseded the original first run.

Notes: Each run depleted about 5-6 miles off of the reported battery range. So that's about 2% each 1/4 mile run. I still love the torque this car delivers every time I step on the acceleration pedal hard and fast. Sorry Honda VTEC, you got nothing on electric motors to snap people's neck back into the seat.

Observation: Concurs with others before me - in order to get the best times, SOC needs to be towards the full battery capacity. Otherwise, car will start to get slower times.View attachment 332863 View attachment 332864

Excellent analysis! Are these numbers with 1 feet rollout or without? Have you seen any difference with newer firmware updates?

Also do you guys happen to know if a similar analysis has been done on AWD non-P and RWD?