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

Improve cooling of drivetrain

This site may earn commission on affiliate links.
Putting on some temp sensors on the radiators or coolant lines would be ideal. The question is really whether Tesla engineered "just enough" cooling for ordinary driving and temperatures or if we're up against a motor or inverter cooling limit. If additional cooling of the existing cooling system would help, it wouldn't be tough to add temperature controlled high velocity fans to the existing cooling radiators that would come on when the heat load is high. If that isn't enough, one could replace the radiators with something larger. There's a fair amount of extra room in the front of a Model S.
 
According to 'sources familiar with the situation', who I spoke to at a track event...

The algorithms which trigger power-limiting are not mechanical sensor based - they are software based. If more than X amount of energy is used in less than Y minutes, power limiting kicks in. It's a very simplistic formula, which doesn't look at any actual temp values.
If you search track threads, you'll see people confused why power-limiting kicks on track at the same time when it's 50 deg vs 90 deg.

Before people start modifying the cooling systems on the car, you'll need to get around the software nannies...

2013 TESLA Model S on the track at at Willow Springs International Raceway - YouTube
 
Last edited:
I'm guessing there's a good reason for this. There are windings on the inner rotor, and there are pretty hefty currents in those windings (induced by magnetic fields from the outside stator). There is no electrical connection of any kind to the outside of the motor; indeed the only mechanical connection is the shaft and bearings.

With this arrangement, there is no simple way to monitor the temperature of those windings. Nor is there any easy way to cool them.
 
I've read there is a way of overriding the traction control, for example, to do burnouts, by removing a fuse. I think it was Jay Leno who mentioned that.

Is there no way to similarly override the drive thermal control system? I suppose someone who is serious about track competition would feel free to burn up a motor or two to find out where the limit really is. Or to install their own sensors. Even if it isn't possible to actually sense the temperature of the rotor and windings, it should be possible to sense the heat being radiated by them to the surrounding case, eh? Not that I'm about to engage in that sort of activity, mind you!

According to 'sources familiar with the situation', who I spoke to at a track event...

The algorithms which trigger power-limiting are not mechanical sensor based - they are software based. If more than X amount of energy is used in less than Y minutes, power limiting kicks in. It's a very simplistic formula, which doesn't look at any actual temp values.
If you search track threads, you'll see people confused why power-limiting kicks on track at the same time when it's 50 deg vs 90 deg.

Before people start modifying the cooling systems on the car, you'll need to get around the software nannies...

2013 TESLA Model S on the track at at Willow Springs International Raceway - YouTube
 
The algorithms which trigger power-limiting are not mechanical sensor based - they are software based. If more than X amount of energy is used in less than Y minutes, power limiting kicks in. It's a very simplistic formula, which doesn't look at any actual temp values. If you search track threads, you'll see people confused why power-limiting kicks on track at the same time when it's 50 deg vs 90 deg.
I see this as another index that the problem is overheating of a component which temperature cannot be directly measured. I can't say tesla was trying to save a few temperature sensors so they went 'the blind way'. Rotor is such a component where measuring temperature while functioning is rather involved. Hence they once measured how much heat the indirect cooling is capable of taking away. When there is more heat generated than it can be removed, they know the rotor must start getting too hot and they limit the power input.

Hollow shaft (end) proves nothing.
 
I see this as another index that the problem is overheating of a component which temperature cannot be directly measured. I can't say tesla was trying to save a few temperature sensors so they went 'the blind way'. Rotor is such a component where measuring temperature while functioning is rather involved. Hence they once measured how much heat the indirect cooling is capable of taking away. When there is more heat generated than it can be removed, they know the rotor must start getting too hot and they limit the power input.

Hollow shaft (end) proves nothing.

Well, in evidence against my earlier thoughts that they might be, the post by doug on THIS PAGE at 10-26-2011, 03:15 PM states that a conversation he had with JB says the same: they aren't cooling the rotor despite the patent.
 
Under full throttle you are pushing 320kW through the motor and generating heat.
Under regen braking you are pulling 60kW through the motor and generating heat.
So was I. Reread the part you quoted. You think I'm doing > 1000 Wh/mi average over a 20 minute period on the street? You give me too much/little credit, sir.

- - - Updated - - -

You'd need to do some careful measuring on the racetrack to see if you get a slightly longer time on the track before your drivetrain became power limited due to heat. I don't think it would be perceptible without measuring down to the second.
What I said was:
It didn't seem to help much with acceleration limiting
I think we're agreeing. Elaborating...

The question as I understood it was whether setting regen to Low would improve the track experience w/r/t hitting the acceleration limiter to a meaningful degree. I'm saying that it does not. I think you're agreeing because you're saying the improvement would required fine measurement to identify. If it requires fine measurement to identify, then I don't think it's a meaningful degree of improvement.

Said another way: if the limit hits after 1 lap for driver X on track Y pretty regularly, meaningful would be at least 1.5 laps of fun mode -- not 1.0347 laps.
 
Thanks for even more input guys! Great resources on this forum.

As it's not my car, I'll have to wait for the next opportunity to test things and theories out. Hopefully the owner is buying some R-tires for the summer season now, to get some better grip for the trackdays.

As for the power pr. minute forumla; I think(sadly) that this may be right, although I hope not. I remember the car used longer time to "cool down" after a long heat of driving, rather than just two-three laps. So if this forumla exists, I thing limiting power to say i.e. 280kw probably would keep the car going without getting the limitier, much longer. But then again, how do you limit the power. Put it in range mode?
 
I did a 5 lap autocross enduro race and was just noticing a tiny power limit on my last lap (4:55 total). I was running w/ regen in standard just to try and keep my brakes cooler since 2 areas of the course are >65mph down to 20-25ish.
For autocross for me, the Model S accel limiter isn't an issue for me.

For higher speed activities (lapping), it's a kill joy and is holding me back from fully experiencing the tracks. This, frankly, is starting to make me consider buying an ICE for such activities. :( This makes me unhappy.
 
So was I. Reread the part you quoted. You think I'm doing > 1000 Wh/mi average over a 20 minute period on the street? You give me too much/little credit, sir.

Well, my track average over 30 miles was 1178 Wh/mi. This was with normal regen, hard friction braking, and pedal to the floor... all as required!

I was thermally limited after three laps or so. Next time I plan on setting regn to low and see if that gains another lap or so at full power.

(pic was posted before but here it is again):
IMG_2408.jpg
 
Well, my track average over 30 miles was 1178 Wh/mi. This was with normal regen, hard friction braking, and pedal to the floor... all as required!
Yah. I think the highest I've seen for a sustained (30+) mile period was around 1700 Wh/mi.

My point was that anything over 1000 Wh/mi for 20+ minutes will likely easily run into the limiter.


Sidenote: If you're willing to share, what location? Sounds like a fun place!