"Better" does not mean less heat. Which is the thing you keep saying 800V does, and pointed me to a person at Tesla that supposedly says 800V is better, but all I can find is that it's tradeoffs.
Here's the problems with non-experts constantly talking about 800V, and using basic electronics to go "half the current, so less heat!"... That's only true if you keep the same size wires, and your motor doesn't have more inductance. But this is not what an engineer does in a production vehicle. They take advantage of the 800V and make the wires smaller.
And then you get the really silly ones that believe in an 800V architecture the current in a given battery cell is half. That's when you know they don't understand anything at all about how EV architectures or batteries work.
If you're going to keep acting like 800V is clearly always a thermal advantage when on a track, please point me to the efficency numbers of the drive unit for an 800V system at 100% rated power and a 400V 100% rated power. The only place heat comes from is this difference. Please link to actual data that shows a large enough gap in efficency that thermals are easy on 800V and impossible on 400V. I mean, we have ICE cars that are only 30% efficent and they don't overheat yet it seems like we're saying 98% is a disaster and 98.4% is great.
Or it could be that Tesla just doesn't do enough thermal managment at all, because all they care about is 1/4 mile or less, and going to 800V won't magically fix this philosophical difference.
Obviously my job is not designing motors nor inverters or batteries, nor is yours very likely.
However I would like to mention that I did doctorate level physics and mathematics before and after school (Arts et Metiers in France one of the best world wide engineering school, general engineering master diploma + 1 year in physics specialization after).
Was 22 years ago though, so I would not be honest saying that I did not loose a lot of things on the theorical side since then. Yeah I spent hours on Maxwell differential and integral equations, calculate flux, charges even calculated tensors, review electrical circuits and experimentation
Not the point here
Very familiar with motors with my job though. I spent days and night configuring, troubleshooting and datalogging drives, servodrives and servomotors, dealing with robots, replacing components on drives after troubleshooting with my meter....
So I have a pretty good understanding on how drives and electric motors work, how to wire them and configure them from scratch depending of the application.
800V is more expensive for some components and necessitate a redesign, that's why Tesla is not using it on its cars, and because for mainstream cars it does not bring a significant increase in efficiency. Because also Tesla is cheap overall, which is good and bad...
But 800V brings :
- Like you said smaller wires
- Better efficiency inverters and lighter : SiC mosfets or IGBT, wires, diodes, transformer/and or inductors should see less current for the same power. we need to take in consideration the fact that an inverter has a sweet spot for voltage/current and obviously manufacturer are optimizing for low loads. When your drive is fully loaded efficiency goes down, and lower voltage architecture will suffer more I^2R losses. You have several ways of driving an electric motor, most of the time it's frequency vs voltage. 800V architecture requires WBG semi conductors.
- Higher voltage motors : It is true that in most cases the efficiency gain for the higher voltage is not always significant, but overall the consensus in the industry is that higher voltage motors always have a slight efficiency advantage over lower voltages ones. However for the same power a higher voltage motor can be smaller, so there is a weight advantage. Less mass can be easier to cool, less thermal inertia ?
- Potentially less loss in the batteries during high loads. Note that Tesla uses 7104 cells in the Model S and 4416 cells in the Model 3, cylindrical design. Porsche, for the same battery size, has only 396 cells, VERY likely a lot less wiring and potential loss at high loads because of that. Maybe Tesla battery disposition is less efficient for cooling as well ? though intuitively I would have said it's easier to cool down smaller cells than big ones, but it depends on how the cooling is done and the passage way of the glycol through that.
It would be ideal that we could have studies on 400V vs 800V but there is nothing available online, I doubt we have the exact specs of drive units from Tesla or Porsche / KIA /Hyundai, they are probably secrets...
What we can see is that 800V cars outperform Tesla as far as thermal management. And aftermarket tuners cannot cool down Tesla as well. The Pikes Peak UP Plaid overheated after 3min30 during the last race. Several nurburgring onboard laps with Plaids show power reduced to 350KW after 3min30 as well.
Tesla Model 3 P overheats after around 5-6 min on the nurb... Sustainable power is 220KW on this one. With frunk delete and additional cooler we may reach 10 min of 3P ? The EV6 GT can do 2 laps yes like the Ioniq 5N, confirmed by several journalists while it was tested. I believe the 5N is even better.
BTW Formula E uses 900V architecture for their powertrain, Rallycross/Nitrocross Group E uses also 870V, another examples higher voltage architecture is used for high performance applications...
So to summarize : 800V seems to be better because of several objective points and outside observations based on the performance of competitors, by how much, I'm not sure, as the thermal management strategy/design is obviously different..
Also, as Nabush points out, the Plaid was thermally limited during its record lap. How much, we can't really tell for sure, but it was obvious in a number of places. Even without the thermal limitation, I don't think the Plaid would have been able to turn a 7:07, but it certainly would have been quite a few seconds faster.
