What is the effect of air-conditioning on battery mileage?

What is the effect of air-conditioning on battery mileage?

Does anyone know, or have an educated opinion, of the effect of air conditioning on the battery mileage of a Tesla M3; or alternatively, how many kilowatts the air-conditioning uses at hypothetically 95°?

Frank99 | 15 agosto 2017

There's no real difference in efficiency between having the A/C run off a belt in an ICE, and having the A/C driven by it's own electric motor - one might argue that the belt is more efficient since you don't have to deal with the 90% efficiency of the alternator, and the 90% efficiency of the A/C motor.

I do believe that efficiency is the main advantage you get from a motor-driven A/C, though. For a luxury ICE with a belt-drive A/C, the A/C is geared to provide effective cooling even when the engine is at idle at a stop light (my Civic, however, foregoes that and doesn't generate effective cooling until I'm moving at speed). That means, unfortunately, that the A/C is spinning much faster than necessary when driving down the road. Spinning faster than needed takes more power than necessary, reducing efficiency. Importantly, it also means that the A/C compressor has to be able to deal with engine RPMs ranging from 800 to 8000.
With a motor-driven A/C compressor, you can design a more efficient compressor because you can control the speed that it turns at. It might have variable speeds, but the variable speeds are controlled by the compressor design - rather than having to deal with a 10:1 speed range, perhaps a 2:1 is all that's necessary to provide comfort for the occupants.

The only reason that most ICE vehicles use a belt driven A/C is cost. It's cheaper to put a pulley on the front of the compressor and let the owner pay for the inefficiencies than it is to put a motor and electronics on the front of the compressor.

Rocky_H | 15 agosto 2017

Heh, I remember my dad's Mitsubishi Mirage (same as Dodge Colt). When you turned the air conditioner on, it felt like the car hit something.

That is why I was so pissed off at the methodology Mythbusters used to "test" the air conditioning versus windows down effect. They were using huge V8 SUVs. Those have such massive engine power, the air conditioning is a barely perceptible fluctuation in the noise and can't really be tested. In a small car with a small engine, the A/C compressor can be strongly felt, as it's a huge drag as a percentage of the tiny engine power.

msmith55 | 19 agosto 2017

The battery temperature must be 10 C minimum, and about 30 C optimum, but temperatures above 30 C degrade battery life , so there is a narrow range where battery is happy. Charging is not allowed below 10 C, so the battery heater, 6 kWh comes on and heats battery before charging can start! That means in a cold climate, you need a level 2 charger to keep the battery happy and charging.

Carl Thompson | 19 agosto 2017

To put it in real world terms you can expect to get close to the rated range in an EV with outside temperatures between 60 degrees Fahrenheit and 80 degrees Fahrenheit (along with careful driving). Optimum is right around 70 degrees or so. Less than 60 and the car will have to do more work to heat the battery and you are more likely to use the cabin heater to heat yourself which takes a big chunk out of your efficiency. Above 80 and the car will have to do more work to cool the battery and you are more likely to use the cabin air conditioner to cool yourself which also detracts from efficiency.

The efficiency of ICE vehicles is a lot less sensitive to outside temperature so this may come as a surprise to some new EV buyers.


Frank99 | 19 agosto 2017

Interestingly, Tesla's range calculator:
indcates that range increases all the way up to 110F as long as you don't use the A/C, so I presume that means that they aren't actively cooling the battery up to that temperature.

Carl Thompson | 19 agosto 2017


Good catch. I should have pointed out in my post that the efficiency drop at high temperatures is much smaller than the efficiency drop at low temperatures. In my EVs I haven't noticed any better range at 80 or 90 degrees than there is at 70 without AC but maybe I simply haven't driven enough at those temperatures without AC to notice.

I'm also curious why range would go up with temperature? Obviously extra energy is not being created from nowhere. Is it that you are simply able to extract more of the electricity from your battery at 100 degrees than at 70 degrees and thus when you charge your battery you'd need to put more energy in to fill it up? That would suggest that any apparent increase in efficiency at higher temperatures is illusory and not real (though the increase in range is real).


JeffreyR | 19 agosto 2017


JB's post (it has links to earlier work too)

Cold Weather Driving
TMC Discussion in 'Blog Archive' started by Doug_G, Jan 15, 2015.

Pro tip: pre-condition your car using the app before you unplug. Also, AC is more efficient than having windows down especially at highway speeds.

Bighorn | 19 agosto 2017

I've not noticed much difference with the windows vs AC. Yesterday, I drove over 100 miles around the Yellowstone and Gallatin area with the pano and two front windows all the way open, speeds 60 and up and averaged under 275 Wh/m

Bighorn | 19 agosto 2017

Full disclosure--about 1800 ft drop net, but the tunes were at 11:)

MarylandS85 | 19 agosto 2017

Lol @bighorn
I’m sure those speakers used almost a good kW of power. In other news, I know a good audiologist I can refer you to, if you’re interested.

Bighorn | 19 agosto 2017

Two things:
Didn't have any problem with the deer.
Insurance doesn't cover hearing aids, but it does cover cochlear implants, so I'm going all in!

DTsea | 19 agosto 2017

Hot day air density is enough lower than cold dahy that it makes a difference.

That is why more home runs are hit on hot days too.

Carl Thompson | 20 agosto 2017


Good point. Higher temperatures and higher humidity both decrease air density which would reduce drag on the vehicle. I wonder if that's enough to account for the full difference Telsa's calculator reports?


Haggy | 21 agosto 2017

" But it doesn't seem like there's anything inherent in an ICE design that says it must use an AC compressor driven by the engine and can't use a separate motor."

But if you do that, it means that the belt that would turn the pulley for the compressor would still turn the pulley for the alternator, which will have to generate the electricity. With a Tesla, it would be impossible to use the main motor to power the AC.

Carl Thompson | 21 agosto 2017

"With a Tesla, it would be impossible to use the main motor to power the AC."

Why would it be impossible? You would just add a pulley for the compressor. No reason it couldn't be done.


topher | 21 agosto 2017

"With a Tesla, it would be impossible to use the main motor to power the AC"

No, it wouldn't. Just stupid.

Thank you kindly

Frank99 | 21 agosto 2017

>> No reason it couldn't be done.
Nope, no reason at all.
But the user experience would be terrible - the A/C would completely stop at every stoplight, even worse than an ICE. You could get around that by having the car shift into neutral at each stop, run the motor to keep the A/C on, then shift into gear to go. Seems easier to just give the A/C it's own motor...

Haggy | 22 agosto 2017

"No, it wouldn't. Just stupid."

If you have to redesign the motor, making it in effect a different motor, and redesign the AC system, you haven't used the main motor to power the AC but come up with a different motor. There's no place to put a flywheel, and assuming the compressor is somewhere up front like on the Model S, and you don't want a belt dangling under the battery, or you don't want to raise the battery (and entire body) to get legal clearance, you'd still need a system of multiple pulleys and belts and there's probably no room for any of that. You'd really need to redesign the car, in which case it wouldn't be using the main motor of this car, but designing a different car that does that. Or you could work it off the front motor and not have AC on RWD cars. But that still leaves the issue of having the motor on when the car is idle. Do you think dual torque converters would work? Or should the motor wind down to zero momentarily, have dual clutches engage, and then get back up to speed when you are ready to drive again? When I say it would be impossible, I'm talking about the current main motor. If you want to redesign the car, you could do whatever you want, but if nobody will buy it it would be pointless. .