I had been wondering if PMAC was maybe better for a while based on the fact that other big companies like Nissan, BMW, and Hyundai were using them when they could have done full induction if they thought that was better. I recall at some point people had been critical saying it was because "motor control software for full induction is harder". But Tesla doesn't appear to need to switch for that reason - they have basically mastered motor control software for induction motors.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
[Spoiler Alert + Mild Speculation] Tesla has created a monster!
- Thread starter omgwtfbyobbq
- Start date
-
- Tags
- Model 3
Funny, the document starting all this frenzy popped up as the first response when I did this search:
I am trying to understand what "both" means... People assume it means dual motor, but maybe something else? Activated by both accelerator and brake pedal perhaps?
I am trying to understand what "both" means... People assume it means dual motor, but maybe something else? Activated by both accelerator and brake pedal perhaps?
Hmm, this CARB doc does suggest "both" means front and rear wheels:
https://www.arb.ca.gov/msprog/arbecert/files/database/ecert_ldv_data_requirements.pdf
https://www.arb.ca.gov/msprog/arbecert/files/database/ecert_ldv_data_requirements.pdf
3Victoria
Active Member
Who knows. My (unlikely) theory (guess): all 3s are AWD. The RWD ones use the second motor only for regen.
JRP3
Hyperactive Member
My (likely) theory is the lack of drive shafts and wiring to the front wheels means the document is not accurate.
Thomas Edison
Active Member
Thomas Edison
Active Member
Naw, I pointed out in another thread that there would be a driveshaft in this photo if there was a front motor.
It's only your theory if incorrect, which is very likely. Otherwise get in line.
Not only is the photo strong evidence against the car being built with a front motor, but Tesla would probably substantially decrease the sale of AWD if that feature was software upgradeable.
Some really cheap people might even wait until they were literally stuck before paying the upgrade cost. "Marge, were stuck. Get out the credit card".
Besides which, why in the world would they make people wait a year to pay them for something they are shipping now?
insaneoctane
Well-Known Member
I have to say there are many details about the Model 3 that haven't been fully revealed and this thread really provided the first evidence of that with the EPA certs. I'd like to provide another example, less paper-trail to follow, but uses fairly sound physics.
I opened a thread asking for folks to estimate 1/4 mile times of the LR because I wanted to see if the LR would be competitive with the BMW 340i for that measure (turns out, probably not). I didn't get a satisfactory answer, so I had to not be lazy. Using first order physics I figured if I could reasonably predict the Model S for both 0-60 and 1/4 mile, then I'd have a good enough model to predict the Model 3's 1/4 mile time and since I already know it's 0-60 time I would be able to tweak any differences between MS and M3....
Ok, if you don't want to know the details of the equations, skip down a few lines
So when I plugged in all the values necessary for the Model S (different tires, mass, frontal area, cd, motor curve, final drive, etc), I got what I would consider pretty damn close:
When I used all the M3 data (tire size, mass, etc.) it came up short from 0-60. It was predicting 0-60 in 5.84 seconds. We know the LR (which is what all the parameters are based off) should be at 5.1 seconds, so what are my levers? Even driving the transmission efficiency to 100% (impossible), only gets me to 5.5 seconds. Lower resistance tires can't get me there. I can't change mass, tire size, final drive ratio, frontal area, or Cd because we actually know these things. I'm open for suggestions of what to try, but my only option was to go to the torque/power curve. Again, the curve is completely flat for basically ALL of the 0-60, so you can't really say it's how I modeled it. I found that if I applied a 1.14x multiplier to the output curve I came in at 5.1 seconds. That would mean that instead of 192 kW motor, it's more like 219 kW (346 HP)! By the way with that 1.14x multiplier I now get:
I opened a thread asking for folks to estimate 1/4 mile times of the LR because I wanted to see if the LR would be competitive with the BMW 340i for that measure (turns out, probably not). I didn't get a satisfactory answer, so I had to not be lazy. Using first order physics I figured if I could reasonably predict the Model S for both 0-60 and 1/4 mile, then I'd have a good enough model to predict the Model 3's 1/4 mile time and since I already know it's 0-60 time I would be able to tweak any differences between MS and M3....
Ok, if you don't want to know the details of the equations, skip down a few lines
- Using Newton's law, F=ma ; a = F/m
- F used to accelerate is force of motor MINUS resistive forces
- Resistive forces, Fresis = Frr + Fdrag where:
- Frr = rolling resistance
- Frr = Crr x v
- Crr = Coeff of rolling resistance (often expressed as ~ 30 x Cdrag)
- v = velocity
- Fdrag = aero resistance
- 0.5 x Cd x A x p x v^2
- Cd = 0.23 (Coeff of drag)
- A = 2.23 m^2 (frontal area, pulled this from some hypothetical threads)
- p = 1.29 kg/m^3 (rho, air density)
- v = velocity
- Frr = rolling resistance
- Force of motor
- Lots of torque and power curves for Teslas floating around with numbered axes. I dropped a scale on one and created a table. For the 3 I scaled the values according to the information available (258 HP from EPA document = 192kW). I then created an M3 motor curve that looked just like the MS, but was 72% of the peak values. With such a flat curve, I think this stands to reason, but will likely be a point of debate
- Fmotor = torque * neff * rfinal / rwheel
- torque is dependent on RPM and in Nm
- neff = transmission efficiency. I used 0.95
- rfinal = final drive ratio (I used 9 from EPA report)
- rwheel = rear wheel (m) I used 235/40R19 to calculate radius for M3 (.3353 m)
- Lots of torque and power curves for Teslas floating around with numbered axes. I dropped a scale on one and created a table. For the 3 I scaled the values according to the information available (258 HP from EPA document = 192kW). I then created an M3 motor curve that looked just like the MS, but was 72% of the peak values. With such a flat curve, I think this stands to reason, but will likely be a point of debate
- Resistive forces, Fresis = Frr + Fdrag where:
- Now for a given change in velocity (ie at 10MPH vs 5 MPH), I can look up the motor torque and calculate both the acceleration and delta t (time in seconds). I can also calculate delta d (distance)
- So in 1 MPH increments, I'm able to calculate accelerations and distance traveled which gives me what I want
So when I plugged in all the values necessary for the Model S (different tires, mass, frontal area, cd, motor curve, final drive, etc), I got what I would consider pretty damn close:
- 0-60 = 4.45 sec
- 1/4 mile = 13.1 sec @ 107 MPH
When I used all the M3 data (tire size, mass, etc.) it came up short from 0-60. It was predicting 0-60 in 5.84 seconds. We know the LR (which is what all the parameters are based off) should be at 5.1 seconds, so what are my levers? Even driving the transmission efficiency to 100% (impossible), only gets me to 5.5 seconds. Lower resistance tires can't get me there. I can't change mass, tire size, final drive ratio, frontal area, or Cd because we actually know these things. I'm open for suggestions of what to try, but my only option was to go to the torque/power curve. Again, the curve is completely flat for basically ALL of the 0-60, so you can't really say it's how I modeled it. I found that if I applied a 1.14x multiplier to the output curve I came in at 5.1 seconds. That would mean that instead of 192 kW motor, it's more like 219 kW (346 HP)! By the way with that 1.14x multiplier I now get:
- 0-60 = 5.1 sec
- 1/4 mile = 13.9 sec @ 101 MPH
- I know my first order estimates do NOT consider traction, I'm ok with that for now.
- Something is off with the data, they still aren't coming clean with this car. Either more powerful motor than advertised, lower mass, or ???
- This message is to further the thread...."Tesla has created a monster!" and I don't think they are ready to fess up yet!
- If your interested in the calcs, here's the google sheets:
- Tesla quarter mile calcs.xlsx
SageBrush
REJECT Fascism
Why does the SR Model 3 have a slower 0-60 than the LR model ?
It would seem to imply that the SR battery cannot output 3.5C for 5 seconds. Or that the motor is software limited.
It would seem to imply that the SR battery cannot output 3.5C for 5 seconds. Or that the motor is software limited.
insaneoctane
Well-Known Member
voip-ninja
Give me some sugar baby
I have to say there are many details about the Model 3 that haven't been fully revealed and this thread really provided the first evidence of that with the EPA certs. I'd like to provide another example, less paper-trail to follow, but uses fairly sound physics.
I opened a thread asking for folks to estimate 1/4 mile times of the LR because I wanted to see if the LR would be competitive with the BMW 340i for that measure (turns out, probably not). I didn't get a satisfactory answer, so I had to not be lazy. Using first order physics I figured if I could reasonably predict the Model S for both 0-60 and 1/4 mile, then I'd have a good enough model to predict the Model 3's 1/4 mile time and since I already know it's 0-60 time I would be able to tweak any differences between MS and M3....
Ok, if you don't want to know the details of the equations, skip down a few lines
- Using Newton's law, F=ma ; a = F/m
- F used to accelerate is force of motor MINUS resistive forces
- Resistive forces, Fresis = Frr + Fdrag where:
- Frr = rolling resistance
- Frr = Crr x v
- Crr = Coeff of rolling resistance (often expressed as ~ 30 x Cdrag)
- v = velocity
- Fdrag = aero resistance
- 0.5 x Cd x A x p x v^2
- Cd = 0.23 (Coeff of drag)
- A = 2.23 m^2 (frontal area, pulled this from some hypothetical threads)
- p = 1.29 kg/m^3 (rho, air density)
- v = velocity
- Force of motor
- Lots of torque and power curves for Teslas floating around with numbered axes. I dropped a scale on one and created a table. For the 3 I scaled the values according to the information available (258 HP from EPA document = 192kW). I then created an M3 motor curve that looked just like the MS, but was 72% of the peak values. With such a flat curve, I think this stands to reason, but will likely be a point of debate
- Fmotor = torque * neff * rfinal / rwheel
- torque is dependent on RPM and in Nm
- neff = transmission efficiency. I used 0.95
- rfinal = final drive ratio (I used 9 from EPA report)
- rwheel = rear wheel (m) I used 235/40R19 to calculate radius for M3 (.3353 m)
- Now for a given change in velocity (ie at 10MPH vs 5 MPH), I can look up the motor torque and calculate both the acceleration and delta t (time in seconds). I can also calculate delta d (distance)
- So in 1 MPH increments, I'm able to calculate accelerations and distance traveled which gives me what I want
So when I plugged in all the values necessary for the Model S (different tires, mass, frontal area, cd, motor curve, final drive, etc), I got what I would consider pretty damn close:
Not shabby IMO. Great, so let's try the M3! This is where it got interesting......
- 0-60 = 4.45 sec
- 1/4 mile = 13.1 sec @ 107 MPH
When I used all the M3 data (tire size, mass, etc.) it came up short from 0-60. It was predicting 0-60 in 5.84 seconds. We know the LR (which is what all the parameters are based off) should be at 5.1 seconds, so what are my levers? Even driving the transmission efficiency to 100% (impossible), only gets me to 5.5 seconds. Lower resistance tires can't get me there. I can't change mass, tire size, final drive ratio, frontal area, or Cd because we actually know these things. I'm open for suggestions of what to try, but my only option was to go to the torque/power curve. Again, the curve is completely flat for basically ALL of the 0-60, so you can't really say it's how I modeled it. I found that if I applied a 1.14x multiplier to the output curve I came in at 5.1 seconds. That would mean that instead of 192 kW motor, it's more like 219 kW (346 HP)! By the way with that 1.14x multiplier I now get:
Summary:
- 0-60 = 5.1 sec
- 1/4 mile = 13.9 sec @ 101 MPH
- I know my first order estimates do NOT consider traction, I'm ok with that for now.
- Something is off with the data, they still aren't coming clean with this car. Either more powerful motor than advertised, lower mass, or ???
- This message is to further the thread...."Tesla has created a monster!" and I don't think they are ready to fess up yet!
- If your interested in the calcs, here's the google sheets:
- Tesla quarter mile calcs.xlsx
Interesting as I posited a few days ago that the horsepower to weight ratio was too low for a 5.1 second time and that it would have to be 300+ horsepower.
I was then roundly condemned as a voodoo witch doctor for my speculation.
SageBrush
REJECT Fascism
Does it ? I somehow got is stuck in my head that 96s was a standard battery config.
Perhaps the battery is amp limited ?
Probably, now that I think about it.
I would vote that there is more power available than 192kW/ / 258 HP
Could it be it default runs in an efficiency mode with that power (and doesn't do 0-60 in 5.1) unless you get it into some kind of Sport mode?
I don't think there is any switch we saw for sport/insane type alternate modes, but maybe there is some not yet published way to bump up the HP when you want quicker 0-60 ?
Has anyone heard anything about special launch modes?
Could it be it default runs in an efficiency mode with that power (and doesn't do 0-60 in 5.1) unless you get it into some kind of Sport mode?
I don't think there is any switch we saw for sport/insane type alternate modes, but maybe there is some not yet published way to bump up the HP when you want quicker 0-60 ?
Has anyone heard anything about special launch modes?
Maybe it gives 192kW when it knows it is being tested on the EPA cycle, but then 'uncorks' the rest when it sees it is being driven by an owner. Oh wait... VW already tried that trick... Nevermind.
Oh wait... VW already tried that trick... Nevermind.AC induction motor does not dissipate power while spinning when it is not energized, but PMAC motors does.
Spinning magnets magnetize iron and there is hysteresis loss.
So dual motor with PMAC is not as straightforward as with AC induction motor as MS has.
Spinning magnets magnetize iron and there is hysteresis loss.
So dual motor with PMAC is not as straightforward as with AC induction motor as MS has.
Similar threads
