Model 3 Motors on the Tesla Parts Catalog

[AlanSubie4Life]

So it isn't like different units perform differently. Binned units could be more effecient and have longer longevity but the cars' 0-30 time is only really determined by torque setting and vehicle weight. For quarter mile the max power matters some.

For the most part I agree. And torque is determined by the current. I just have the datapoint where my car pulled power (though peak torque was still boosted like @SleeperService, just dropped at 45mph rapidly ) until I switched to track mode, even at 92% SoC. Possibly that was just battery system related. I duplicated the same 0-60 run time without track mode several times with steadily decreasing SoC...first run with track mode at closer to 90% SoC was faster between 45-60 mph, same elsewhere. I’ve posted the data elsewhere.
So not clear to me what exactly caused that (and I don’t think there is any way to know). It COULD be motor efficiency in conjunction with battery power limitations, or it could be just battery power draw limiting, unrelated to the motor. Haven’t tried again with subsequent updates which have tweaked high power run time; might make some difference. Not really inclined to try...

When I say “no way to know” - I mean from that particular instance. In general it would have been interesting to see what the car said was being drawn from the battery in each run. Obviously power draw was higher in Track Mode than normal mode for my car, though. You’d then have to compare to data from another P3D+ to see whether for the exact same conditions and run time time, whether the power draw was any lower (that might point toward a motor efficiency difference).

 

[Krash]

...And torque is determined by the current...

The torque (and acceleration) from 0-30 is limited by a software setting called max torque. So if you mean it draws less current when going slower that is correct. But the idea that battery or inverter differences make a difference in performance during that range is incorrect.

Above 60 mph where the max power limits are in effect and the max torque limit doesn't govern performance, more effeciency means more power to the wheels. I never considered it before but the raven S/X may have better performance simply because the car may realize that power to the front PMSR (raven) gets delivered to the wheels more efficiently when the car is power limited.

 

[peakshaving]

The torque (and acceleration) from 0-30 is limited by a software setting called max torque. So if you mean it draws less current when going slower that is correct. But the idea that battery or inverter differences make a difference in performance during that range is incorrect.

Above 60 mph where the max power limits are in effect and the max torque limit doesn't govern performance, more effeciency means more power to the wheels. I never considered it before but the raven S/X may have better performance simply because the car may realize that power to the front PMSR (raven) gets delivered to the wheels more efficiently when the car is power limited.

I read somewhere that that's exactly why the raven S/X have better range - the PMSR is more efficient at lower speeds/ power limited scenarios. What are the benefits of having an induction motor at this point? I expected Tesla to just put two PMSR motors in all their cars and just put different gearing ratios for different speeds.

 

[Saghost]

I read somewhere that that's exactly why the raven S/X have better range - the PMSR is more efficient at lower speeds/ power limited scenarios. What are the benefits of having an induction motor at this point? I expected Tesla to just put two PMSR motors in all their cars and just put different gearing ratios for different speeds.

PMSR motors don't have much in common with the permanent magnet motors used by much of the rest of the industry, but one thing they do share is that it'll drag and induce currents if it's spun with no power.

The induction motor you can just put to sleep - no current in the stator means no current in the rotor, and it just sits and spins on its bearings. The others, not so much. If you don't impose currents, the magnets will still generate their own.

So one PMSR and one induction motor is probably the most efficient possible configuration for an AWD road going car. You run the PMSR all the time, and the induction motor only when you need more power or more traction.

 

[kengchang]

I read somewhere that that's exactly why the raven S/X have better range - the PMSR is more efficient at lower speeds/ power limited scenarios. What are the benefits of having an induction motor at this point? I expected Tesla to just put two PMSR motors in all their cars and just put different gearing ratios for different speeds.

Can't sleep PMSR, not going to happen with 2 PMSR setup

 

[peakshaving]

PMSR motors don't have much in common with the permanent magnet motors used by much of the rest of the industry, but one thing they do share is that it'll drag and induce currents if it's spun with no power.

The induction motor you can just put to sleep - no current in the stator means no current in the rotor, and it just sits and spins on its bearings. The others, not so much. If you don't impose currents, the magnets will still generate their own.

So one PMSR and one induction motor is probably the most efficient possible configuration for an AWD road going car. You run the PMSR all the time, and the induction motor only when you need more power or more traction.

Oh I didn't know that. I thought that since the rotor of SR motors is just a solid piece of steel it wouldn't induce any current if you don't magnetize the windings but you're right! Since these permanent magnets there's always a field that'll cause current to be generated. Thanks for enlightening me.

I suppose the next generation Tesla and the industry would want to achieve would be pure SRM motors without any permanent magnets and just eliminating the torque ripple with really fancy motor controls. Saves the cost of the magnets and enables freewheeling.

Few questions for you knowledgeable engineers-

I've heard that these motors (SRM) lose torque at higher RPMs. Is that true? I can't see why it would since as long as the switching happens fast enough to keep up with the rotor I don't see why there would be any loss of torque at higher speeds.

Are the permanent magnets that Tesla uses on their PMSRM just there to smooth out the torque ripple at low rpm?

 

[Saghost]

Oh I didn't know that. I thought that since the rotor of SR motors is just a solid piece of steel it wouldn't induce any current if you don't magnetize the windings but you're right! Since these permanent magnets there's always a field that'll cause current to be generated. Thanks for enlightening me.

I suppose the next generation Tesla and the industry would want to achieve would be pure SRM motors without any permanent magnets and just eliminating the torque ripple with really fancy motor controls. Saves the cost of the magnets and enables freewheeling.

Few questions for you knowledgeable engineers-

I've heard that these motors (SRM) lose torque at higher RPMs. Is that true? I can't see why it would since as long as the switching happens fast enough to keep up with the rotor I don't see why there would be any loss of torque at higher speeds.

Are the permanent magnets that Tesla uses on their PMSRM just there to smooth out the torque ripple at low rpm?

I'm not anything resembling an expert in this field, but what I took away from the only technical articles on the subject I've seen is that the additional poles created by using the magnets (ferrite, not rare earth) added to the efficiency significantly in addition to helping the torque ripple.

Switched reluctance motorsan old design is suddenly relevant again | ElectricMotorcycle.com | The Electric Revolution Starts Here

Tesla Model 3 Motor — Everything I've Been Able To Learn About It (Welcome To The Machine) | CleanTechnica

 

[Krash]

...I've heard that these motors (SRM) lose torque at higher RPMs...

I don't know about pure switched reluctance motors, but the induction motors and likely the PMSR motors reach limits of power at higher RPMs. The power draw actually falls below the max power limits.

The leading theory on this is Back EMF. My (limited) understanding is that larger motors raise that curve before BackEMF kicks in but of course are heavier and more expensive.

 

[AlanSubie4Life]

For sure AWDs are software locked to limit the maximum current to the rear motor

Different models of cars have different max torque and max power limits

And torque is determined by the current.

The torque (and acceleration) from 0-30 is limited by a software setting called max torque.

I was just saying the software difference (especially from 0-45mph) between AWD and P3D limits the current that is delivered from the battery to the motors on the AWD version. (Especially the rear motor.)

HP = Torque * Speed

Power drawn from the battery is just I*V, and V is approximately fixed for a given SoC.
So if efficiency is η:

Torque*Speed = η * I * V

So, Torque = ( η *V / Speed) * I

So for any speed up until the HP limit, Torque is determined by the battery current. (Everything else is approximately a constant for a particular speed.) So the software setting of max torque is limiting the battery current, at least indirectly.

That's all I was saying, and I think we're in agreement for the most part. I think the key thing you're saying is that they have means of measuring and controlling torque, and the current limit just follows, up until the battery power limit.

So if you mean it draws less current when going slower that is correct.

Yes, you draw current proportional to speed, up until you hit the power limit, on the P3D. And the current is lower at every speed in the AWD (I think). I suppose the AWD keeps increasing its current for a bit longer since the eventual AWD/P3D HP difference isn't that much I think (???). I haven't ever paid much attention to the plots for the AWD and exactly how their power ramps up - pretty sure it's just a shallower slope. But I do know above 60mph there isn't that much difference between AWD and P (but at that time you're past the power peak I think - again, haven't really looked exactly at where the power peak is).

But the idea that battery or inverter differences make a difference in performance during that range is incorrect.

I wasn't saying that the battery mattered below the power limit.
On my car, I only saw a difference in track vs. non-track mode once I got to the HP limit (you can tell where that is because that is where acceleration starts to drop). Here's my comparison that I was alluding to:

P3D+ Acceleration

Above 60 mph where the max power limits are in effect and the max torque limit doesn't govern performance, more efficiency means more power to the wheels.

Yes, that's why I was wondering whether an efficiency difference was responsible for the difference in track mode vs. non-track mode for the results above.
I think you can see from the post above that for the P3D, the power limit is in effect above ~45mph.

 

[TimothyHW3]

I don't know about pure switched reluctance motors, but the induction motors and likely the PMSR motors reach limits of power at higher RPMs. The power draw actually falls below the max power limits.

The leading theory on this is Back EMF. My (limited) understanding is that larger motors raise that curve before BackEMF kicks in but of course are heavier and more expensive.

So could you please explain when the induction motor is sleeping? Maybe in city driving at constant accel. then it sleeps? Or if I am cruising at 80on the highway, then I use only the RWD, is that it? Or does it work to help with traction and maybe if there is water on the road? Would be nice to know, maybe it will help me drive the consumption down a bit.

 

[Saghost]

So could you please explain when the induction motor is sleeping? Maybe in city driving at constant accel. then it sleeps? Or if I am cruising at 80on the highway, then I use only the RWD, is that it? Or does it work to help with traction and maybe if there is water on the road? Would be nice to know, maybe it will help me drive the consumption down a bit.

On my Raven X, the induction motor sleeps most of the time.

Full regen below about 60 mph wakes it up slightly (at 50 mph full regen was about an 80% forward split. At 70 mph it was all the front switched reluctance motor.)

It wakes briefly on acceleration from a stop, but goes back to sleep quickly if the acceleration is gentle

Obviously, a hard accelerator stab will wake it at any speed. Mild and moderate acceleration generally doesn't, though my dataset is still fairly narrow.

 

[Wikzor]

Hi guys,
Noticed that you wanted pic as proof, here it is:

Btw regarding the LC-part, couldn't that be referring to LC filter?

Thanks! Do you have a picture?

So they have revved it to a -F on the motor...

I filled in some of the pattern below, it might need to be corrected.

At the moment the 990 motor sighting reported earlier is unverified as there was no picture.

Updating build dates provided by @EricUSC earlier in this thread:

Rear Motor
Build Date 5/18: 1120980-00-C
***Changed from C to D sometime between 5/18-8/18***
Build Date 8/18-12/18: 1120980-00-D (blank label 9/18-10/18 followed by font change)
Build Date As early as 2/19- As late as 4/19 : 1120980-00-E ( To Sweden - I saw one example, we may have other pictures...)
Build Date As early as 4/19-?: 1120980-00-F (need a picture)

Rear Inverter Plate
Build Date 5/18: P1085693-00-E
Build Date 5/18-9/18: P1085693-20-E
***Changed from 20-E to 30-E between 9/18-10/18***
Build Date 10/18-2/19: P1085693-30-E (end point indeterminate, Sweden delivery build date 2/19)
Build Date 4/19: P1085693-00-F (tentative pending a picture)

Thanks! Do you have a picture?

So they have revved it to a -F on the motor...

I filled in some of the pattern below, it might need to be corrected.

At the moment the 990 motor sighting reported earlier is unverified as there was no picture.

Updating build dates provided by @EricUSC earlier in this thread:

Rear Motor
Build Date 5/18: 1120980-00-C
***Changed from C to D sometime between 5/18-8/18***
Build Date 8/18-12/18: 1120980-00-D (blank label 9/18-10/18 followed by font change)
Build Date As early as 2/19- As late as 4/19 : 1120980-00-E ( To Sweden - I saw one example, we may have other pictures...)
Build Date As early as 4/19-?: 1120980-00-F (need a picture)

Rear Inverter Plate
Build Date 5/18: P1085693-00-E
Build Date 5/18-9/18: P1085693-20-E
***Changed from 20-E to 30-E between 9/18-10/18***
Build Date 10/18-2/19: P1085693-30-E (end point indeterminate, Sweden delivery build date 2/19)
Build Date 4/19: P1085693-00-F (tentative pending a picture)

 

[TimothyHW3]

What Model is that SR+, AWD or RWD?

 

[Wikzor]

It's an AWD

 

[AlanSubie4Life]

Thanks. Maybe eventually we’ll get more pictures of more recent builds and fill in which cars get 990 motors.

So far this is the only one we’ve seen.

I doubt the LC refers to LC filter, but really I have no idea.

 

[voltronhb]

Had a drive unit replaced for our Model 3 LR RWD last month. The DU part was ASY,3DUR,MOSFET-HC (1120980-00-F). Anecdotally, seems like I’m getting better efficiency. It could be since our prior DU was failing.

 

[p-f-g]

But what are the "P-TRAIN" versions in the catalog, is this new ?

Model 3 Rear Drive Unit part#

• p-f-g
• Jul 15, 2019

 

[scubastevo80]

But what are the "P-TRAIN" versions in the catalog, is this new ?

Model 3 Rear Drive Unit part#

• p-f-g
• Jul 15, 2019

Vishay Siliconix website seems to indicate that the primary difference (at least that I can tell) between the P-TRAIN MOSFET inverters are that the 840 has max drain source voltage of 500v, and the 630 has a max of 200v. I'm not sure how those stack up to the other 3 in terms of max voltage and efficiency - perhaps one of the elec engineers here can comment.

 

[mongo]

Vishay Siliconix website seems to indicate that the primary difference (at least that I can tell) between the P-TRAIN MOSFET inverters are that the 840 has max drain source voltage of 500v, and the 630 has a max of 200v. I'm not sure how those stack up to the other 3 in terms of max voltage and efficiency - perhaps one of the elec engineers here can comment.

Any switching device need to handle greater than pack voltage, so > 400V minimum + derating. The P-TRAIN may just be shorthand for power train, and RMN re-manufactured. The 840 vs 630 may be peak currents: 840A * 350V = ~400hp, 630A*350V = ~300hp.

 

[MrHopsing]

LR AWD
VIN 224###
Norway

Interesting thread, especially now with the talks of AWD > P3D- update.