9kW "stator heater" ON when supercharging. Hypocritical waste?

[ucmndd]

We're talking about a dedicated 50 kW CHAdMO station. That's not applicable.

I for one don’t find it particularly surprising that Tesla isn’t spending a ton of developer time optimizing their DCFC code for CHAdeMO.

 

[Rocky_H]

I for one don’t find it particularly surprising that Tesla isn’t spending a ton of developer time optimizing their DCFC code for CHAdeMO.

Sigh. Let's go for the other format then--CCS. Harley Davidson dealerships have those 20 kW DC fast charging stations. Once Tesla starts selling the CCS1 adapter in North America, then we can point out how incredibly dumb this is for Tesla to boil the battery for a lowly 20 kW charging power. Tesla's Gen2 wall connectors supplied 20 kW and didn't need this insane level of heating.

The point is that this so-called "logic" is really awful and isn't specifically about the format of CHAdeMO or CCS or Supercharging. They need to do better.

 

[n2mb_racing]

Sigh. Let's go for the other format then--CCS. Harley Davidson dealerships have those 20 kW DC fast charging stations. Once Tesla starts selling the CCS1 adapter in North America, then we can point out how incredibly dumb this is for Tesla to boil the battery for a lowly 20 kW charging power. Tesla's Gen2 wall connectors supplied 20 kW and didn't need this insane level of heating.

The point is that this so-called "logic" is really awful and isn't specifically about the format of CHAdeMO or CCS or Supercharging. They need to do better.

Seems like a good idea, but they've got a lot of bugs to deal with before they probably get there. Software isn't exactly their best talent. Nor manufacturing. Hardware engineering, maybe?

 

[FalconFour]

Indeed, and the point here is that Tesla is the best at battery engineering, for sure, but nothing is perfect, especially in a system with so many moving parts.

Today, the logic seems to take the SOC% and applies it to the "ActiveHeat" target - that's all. Dumb linear mapping logic, regardless of charging rate. If DC charging, apply linear SOC%->ActiveHeat calculation. As SOC% increases, target increases. The result for slower charging stations is "it heats all the time"; the result for Supercharging is it blows right past that target and hits ActiveCool targets instead.

Really, I just think the logic should apply a distance-to-max-power logic to the code. Look at the difference between the station's max rate and the BMS max rate (which is part of the numbers it's always thinking about). If the station amps are far less than the BMS max, don't heat - it'll do no good and cause a slower/more wasteful charging experience. If the station max approaches or exceeds the BMS max, heat away!

Strangely, though, V3 Supercharging seems to just turn a blind eye to the BMS advertised max, and just blasts it with ludicrous amps exceeding its advertised max... not sure what the logic is there. It also doesn't seem to matter - heating only takes place for a very short period of time before the pack heats itself up and exceeds that target anyway.

Just a simple calculation of delta between EVSE max amps (which is, of course, advertised on the line) and BMS max amps, to determine heating, seems like a relatively quick fix on par with the USB media album art fix we recently got just prior to the holiday update (yay). (which also implies the fix in legacy code was simple enough to deploy for only 1 version between 2021.36.x and 2021.40.x).

Or just a switch, as simple as moving that existing control out of the service menu and into somewhere user-accessible It already has the "may impact charging performance" caveat right on it!

 

[Richbot]

So, I'll pour some fuel on this fire, because the "wtf are you doing, car?" isn't limited to 3/Y at superchargers and they may have some sort of oops going on more broadly with V3 logic

I can't recall when, but at some point mid-summer it seems software nerfed V3 charging on my Long Range Plus battery. It'll pop up to 203 for a few minutes at low SoC with preconditioning on the way to the charger, and yeet 10-12% into the battery, but then it'll throttle back quickly (down to 120ish by about 30% starting around 10-15) then down to 90-95kw for a good 15 minutes before creeping back up above 100 just in time for the SoC to be too high for it to accept any more.

I wonder, with 200+kw already generating a ton of heat, plus battery heating, if it's hitting some sort of weird "I should be cooling right now but I can't because I am told I must heat" limit or lockout. I haven't had a chance to do enough experimentation in lots of different conditions, but I never hear the AC running during V3 supercharging lately either, until the very, very end of the session if at all (and normally not going past about 60% anyway at an SC with the big battery)

I'm asking the service center to look at this which will probably yield some form of "could not reproduce, system is operating as designed" non-response. But it's not lining up with the charge curves I've seen, and it used to, and it seems 100% software. On a V2, it's a totally different story, and the curve seems to make sense peaking at 150, holding it for a good while, and tapering gradually and cooling kicking on mid-session

anyway I think something's f**ky beyond just the Model 3's.

 

[Gauss Guzzler]

...but you're still stubbornly (not even curious or trying to have a real conversation, but already-made-your-mind-up based on "Tesla is smart" and nothing more)

If heating were really "necessary" for charging a battery (despite all the overwhelming evidence to the contrary over years of past battery experience...

I didn't say "Tesla is smart".
I believe my exact words were: "You're not a battery designer"

When you objected, I asked you to show some of your data, even the simplest out-of-context plot like this would really help to back your assertion (this plot obviously contradicts your claim, doesn't factor in variable charge rates and isn't even for Tesla cells but sorry, it's all I could find).

Obviously you have more relevant data or you wouldn't be claiming to know more about battery chemistry than all of the engineers in the largest and most crucial department of the most valuable company in the world, right?

I mean, I could have asked you some really big questions, as you know, about how heating can cause or cure a variety of conditions. But I didn't. I just asked for a rough estimate of the preferred temperature for the weird 50kW charging case you cited. You clearly prefer a different temperature than Tesla, right? Well, what is it?

OK, fine, maybe I was right - you're not a battery designer and you don't actually know anything about batteries. That's cool. I don't either. So let's play blind-leading-blind and assume you are right about everything:
Of all the Wh that have ever been consumed by all of the Teslas in the world combined, what percentage of that power came from slow fast chargers? Is it significantly more than 0?

 

[FalconFour]

Well, the heating/cooling is controlled by 3 numbers the car is always thinking about: ActiveHeat, Passive, and ActiveCool. The 3 numbers never overlap (that'd be bad), e.g. ActiveHeat=130f, ActiveCool=120f = probably some kind of fault condition. The cooling/heating logic is derived from those - it takes the hottest or coolest temperature and reacts to it. I do wonder, though, what happens in that grey area, when it looks at the coolest (when heating) vs. the hottest (when cooling) numbers.

Given your situation, I'd actually think the charge port might be getting too hot (another thing it thinks about) - but I haven't actually taken any CAN measurements of that (worth a look). Those v3 stations pump an utterly batsh*t-crazy level of amperage through pins that (relatively) small; I always have been surprised what they get away with. I'd imagine that with age, the resistance increases and the temperature can play a larger part - especially since the equipment in the car isn't liquid-cooled like the Supercharger station's cables are (wonder what that whirring in the pedestal is about? ).

Given the absolute 40kW Supercharge-nerfing hell that 2016-era Model S's have experienced, I'm still happy to be getting 250kW at all on my 2018 LR 3, when I do (and I do). I just think that's a whoooole different can'o'worms than the heating bit.

 

[Richbot]

great now I have to bring a bucket of ice water and an arc flash suit on my next road trip

 

[FalconFour]

great now I have to bring a bucket of ice water and an arc flash suit on my next road trip

Or an ice pack and no crazy suits

no but really, act on measurements, not guesses. If you really want to dig like I have, get set-up with Scan My Tesla or tes-LAX, where you can get the measurements out of the car. That'll tell you if bringing an ice pack will do any good. Like I said, it's just my guess, and without data, a guess is as good as a wish... it can be an educated one, and it can lead to what data to measure, but you never know for sure without evidence/data.

Just like we've done with the data and measurements of wasteful heating

 

[Richbot]

indeed. If the tesla center doesn't find anything whcih I'm confident it won't, I suppose it's time for me to wonk out. Sad my wifi obd adapter software only works on my old cars

 

[rrolsbe]

I for one commend the author of this thread, Falconfour and others with constructive comments! No one in this discussion could possibly be 100% correct nor are the Tesla engineers always doing things the best possible way. I am the only one that is 100% correct all the time! Being an engineer (NOT a battery engineer) it sure appears that in certain DCFC scenarios a fair amount of an owners money (due to excessive heating) and time are being squandered. Even if charging at very elevated temps, above what improves charging speeds, helps with traction battery longevity is it enough to justify the added owner cost and energy/time wasted?? I for one monitor the Cell temps via SMT and try to arrive at a Tesla SC at around 100+F. With this method, I have not experienced any reduced charging speeds (but I could be wrong). One plus having a rear drive M3 the afterburner can only use 4KW of energy for battery heating during a SC session. I have not monitored battery heating parameters prior to and during V3 SC charging sessions but it appears it could be potentially wonky (like that word used in this thread) or possibly detrimental to the cells in the traction battery?

 

[rrolsbe]

Because charging begins instantly, at max speed of the station, and heating takes tens of minutes to achieve. Thus, for more than half the present charging session, there's little to no effect given to the battery by the heating... and yet the car imposes no charging rate limits. If heating were really "necessary" for charging a battery (despite all the overwhelming evidence to the contrary over years of past battery experience and "not wanting them to be hot"), the car would limit charging rate until the heating actually takes effect.

It does not. Charging takes place at the same rate whether it's hot or cold - except for near- or below-freezing cases - even when Supercharging, as the rate of charge causes the battery to heat up far faster than the battery heater has any effect on (and it quickly switches to cooling).

In all but "urban Supercharger" cases, the heating is unnecessary and works counter to the goal of charging, as it consumes unnecessary energy.

The fact that we've gone this far - where most of this information can be determined with data from the car combined with common sense - but you're still stubbornly (not even curious or trying to have a real conversation, but already-made-your-mind-up based on "Tesla is smart" and nothing more) sticking to this "you're wrong, I'm right" angle of argument... this'll be the last post of yours I reply to.

In all but "urban Supercharger" cases, the heating is unnecessary and works counter to the goal of charging, as it consumes unnecessary energy.

Can you please elaborate on your above sentence as it relates to an Urban SC charging experience I described in another thread (see below).

Thanks Very Much! Regards....

I believe the battery heating algorithm only takes into account that it is DC charging. Even if it is a 72KW urban charger makes no difference. I was sitting in a traffic jam in LA last Monday with an estimated arrival time to an Urban charger 90 minutes later when SMT showed battery temp of 127F and stator temp of 145F. When I set the navigation destination to a business across the street from that Urban charger, the stator temp started dropping. When I got within a mile of the charger, I switched the navigation back to the charger, charged at max of 72KW just fine.

 

[FalconFour]

In all but "urban Supercharger" cases, the heating is unnecessary and works counter to the goal of charging, as it consumes unnecessary energy.

Can you please elaborate on your above sentence as it relates to an Urban SC charging experience I described in another thread (see below).

Thanks Very Much! Regards....

I believe the battery heating algorithm only takes into account that it is DC charging. Even if it is a 72KW urban charger makes no difference. I was sitting in a traffic jam in LA last Monday with an estimated arrival time to an Urban charger 90 minutes later when SMT showed battery temp of 127F and stator temp of 145F. When I set the navigation destination to a business across the street from that Urban charger, the stator temp started dropping. When I got within a mile of the charger, I switched the navigation back to the charger, charged at max of 72KW just fine.

Hot battery charges faster at high SOC%. How on earth did you get to 127f? What SOC% did you start preconditioning at? Preconditioning also just hypnotizes it into "I must be DC charging " mode and sets DC charging targets while driving. But that 127f is really, really hot! Sounds like you started-off at 60% SOC or so already.

If it were not heated, and at 60f or below, it'd probably start tapering-off/slowing down (slowly, long ramp) at 80% SOC. With heating, it'll probably get to 95% before slowing down (fairly quickly, fast ramp). (guessing here, based on experience). If you only charged in this session from 60% to 80%, you'd never even notice hot vs. cold difference. (and isn't that the rub? wouldn't it be nice if it knew how long you wanted to charge to, to decide if heating is necessary? Oh wait, that IS a setting we give it )

 

[Gauss Guzzler]

Preconditioning doesn't necessarily "waste" power by heating the battery. It redirects any available heat from the motors, air conditioning, infotainment into the battery instead of the radiator. So if you're driving for a long time it'll take full advantage of this available heat to get the pack as close to optimal as possible. That's why it starts preconditioning hours early.

The battery isn't warmed to make fast charging possible, it's warmed to reduce the degradation that occurs during charging. This indirectly makes fast charging an acceptable lifespan compromise.

So in the case of a 50kW charger jumping right to full speed on a cold battery, it might just be that Tesla deliberately allowed a little extra cell damage to occur in the interest of saving time in these rare cases of slow fast charging.

 

[Rocky_H]

The battery isn't warmed to make fast charging possible, it's warmed to reduce the degradation that occurs during charging. This indirectly makes fast charging an acceptable lifespan compromise.

So in the case of a 50kW charger jumping right to full speed on a cold battery, it might just be that Tesla deliberately allowed a little extra cell damage to occur in the interest of saving time in these rare cases of slow fast charging.

Your insistence that this is deliberate is amazing. Are you still going to rationalize this with 20 kW charging too? It doesn't do this excessive heating with 20 kW from an AC wall connector. So if the 20 kW is converted to DC outside the car versus inside the car, why do you think that the battery needs to be excessively heated in one usage of 20 kW but not in the other usage of 20 kW? So why would you think it would decide that this extra degradation is OK in one 20 kW case but not the other? Sometimes if something makes no sense, it's because it's a mistake.

 

[MP3Mike]

It doesn't do this excessive heating with 20 kW from an AC wall connector.

Of course it doesn't. It can't as there is no 20kW AC charging on a Model 3..

 

[Rocky_H]

Of course it doesn't. It can't as there is no 20kW AC charging on a Model 3..

I know that. I thought it still does this messed up logic on all of the cars though, including the old Model S cars that can take 20 kW AC charging. That's what I was referring to in my example, but forgot to specify that.

 

[Gauss Guzzler]

I didn't "insist" that it was deliberate. I believe my exact wording was "might just be".

I also implied earlier that perhaps Tesla simply hasn't bothered to optimize the perfect balance of cell degradation vs. charge speed vs. charge efficiency for these rare slow-fast chargers as they don't see/expect a significant percentage of charging to occur this way so they erred on the side of battery health.

 

[googlepeakoil]

Neat, then heat it then. It's nowhere near the max battery power, and at Supercharging speeds, the battery heats itself up to ActiveCool temperature faster than anything else (then it's struggling to cool down).

So we have these scenarios:
- <=50kW CHAdeMO station: 4-8kW wasted (10-20% ish) on heating when it was never and will never be needed and it actively hinders the charging experience
- 72kW urban SC: 4-8kW (5-10%ish) on heating to retain speed through higher SOC%
- 120kW v2 Supercharging: 4-8kW (3-6%ish) on heating but often reaches its target by heat produced by battery charging; no active heating was ever needed
- 250kW v3 Supercharging: 4-8kW wasted (2-4%ish) on heating for the first minute or two (not even long enough for the coolant to heat up 1 or 2 degrees and get to the battery), but heat produced by charging is overwhelming - it then shuts off heating, and a few minutes later it's actively cooling to remove excess heat produced by charging

Because of the very lobotomized "if DC charging, then heat" logic, it meets the needs of only the middle use cases, but falls apart at both ends of other cases.

All we're asking is for a smarter system, or at least this neat little switch from Service Mode to be present in normal options:
View attachment 751555

Fact is, the cell voltage is the limit of charging rate, and Tesla's smart enough to know what the rate limit is. If you're not even close to hitting the rate limit, there's no need to be heating it to increase that ceiling.

Wow - is there really such a button? You say it's only in service mode?
I will intentionally "surprise" the car by navigating to supercharger in the last 5 minutes. This way you won't be heating the battery to 55C as it won't have time to get so hot! And I might not mind sometimes getting 70kW instead of 100kW if I'm taking a break to get a snack / stretch legs.
A few times while sitting on my drive looking around the superchargers on the map I've clicked on one to see cost / stalls / speed! Suddenly the car whirrs up it's fans and says "pre-conditioning car for supercharging". I'm not even in Drive mode!!! It's a bit OTT!