Battery charging and discharging losses

Battery charging and discharging losses

I’ve been very enthusiastic about simplicity and efficiency of electric motor in cars vs heavy and inefficient ICE engine. However one aspect I didn’t see much publicized is the inefficiency of the charging and discharging from Li-Ion battery pack. I wonder what information other forum members have to educate all of us.

To be specific here is an example. Assume my Model S batter pack is completely empty and has a capacity of 42kWh. I plug it in my 220V 50amps outlet in the garage and assume I want to charge it 100%. What will be my meter reading at the end of the charge? Can anyone approximate? I know it will be more than 42kWh, but how much more? The electricity will have to be used for cooling of the battery and for losses during electricity transfer to the Li-Ion cells.

Here is another question. Does anyone know if I choose to charge at 110V 15Amps standard outlet vs 220V 50 or 70amps will one charging method yield smaller electricity losses during charging?

Overall I think we need to account for charging/discharging inefficiencies when we talk about electric cars. ICE engine may be only 30% efficient, but we don’t lose a portion of the gasoline when filling up the tank.

I am a reservation holder of Model S, so please don't attack me for asking this question :)

David70 | July 21, 2011

I don't know specifically what the efficiencies are in either case, but one would expect charging at lower current and voltage would have smaller losses. I know that the internal resistance of Li-ion is lower than most other batteries, but don't know what the actual values are. If heating is typical Joule heating, then cutting voltage and current each in half should reduce losses to one-fourth as much.

In any case I seriously doubt that even high current and voltage charging would make the efficiency less than that of a gasoline or even a diesel engine.

Maybe someone else here knows the actual internal resistance of Lithium-ion cells. This is a question I posted long ago, but haven't yet had any response.

Brian H | July 22, 2011

Well, the general guideline is that the ~56kwh Roadster battery takes ~70kwh to charge. (approx) So figure 20% of input is lost to heat. How much is the charger and cabling, and how much the batteries, I have no idea.

Timo | July 22, 2011

Wasn't that 60kWh, not 70kWh? 10% is more like it, 20% sounds too high to me.

mnx | July 22, 2011
David70 | July 22, 2011

A follow up on nlukin's question.

The other numbers are for convenience in calculation.

The following description is for Panasonic 3.7 V 18650 batteries.

Internal impedance is less than 180 milliohms (0.18 ohms)

If charging is at a rate of 70kW and there are 7000 cells (it's very roughly that), then each cell is charged at a rate of 10W.
If the charging voltage higher than the average voltage of 3.7 V (e.g., 4V) the charging current would 2.5A and Joule losses would be ~ (2.5A)^2*0.18ohm = 1.125 W. That's about 11%. If you went to 140kW (which can easily be done by the Roadster for very short times),the losses would be about 45%.
Losses at 120V/15A (1.8 kW)would be under 3%.

Just FYI, since the Tesla doesn't (I believe) use the 3.7 V cells.

David70 | July 22, 2011

P.S. This is only battery losses, not losses in wiring, motor/generator, etc.

DC | July 22, 2011

I know you were just asking about the battery, but this goes over everything, and shows you how much more efficient an EV is over an ICE.

Ramon123 | July 28, 2011

Losses due to battery storage are typically around 25%.
The folks at GM actually provided data on how many kilowatthours from the wall were required to get so-and-so number of kilowatthours available to the motor. Pay no attention to motor
or drivetrain "inefficiencies" - those are included in the miles per kilowatthour. A book I have on solar panel systems with lead acid batteries states that these batteries have a 25% loss,
same as the li ion batteries used in the Volt.

nlukin | July 30, 2011

Thanks guys. Good info.

To sum it up 20-30% of electricity is lost during charge time. So, to charge 42kWh battery should be about 53kWh of electricity from the wall.
In CT, i am getting charged about 19 cents for 1kWa (combined delivery and service gen fee). So, 160 miles will cost me about $10 give or take. It is still much better than paying $60 for barely 300 miles on my Acura.

Robert.Boston | August 1, 2011

Make sure that you check your utility's rate structures. Many utilities offer residential customers a "time of use" rate (or something similar), under which you pay a lower price for energy during off-peak periods. Shifting to such a rate will reduce the cost of operating your EV, provided that you are careful to charge your car overnight.

BTW, you should check on these rate plans soon and request the change in rates sometime well before you take delivery of your EV, because most utilities need a few months to install the special meter required to put you on a time-of-use rate.

Brian H | August 3, 2011

Here (BC, Can.) it's <7¢/kwh, so your 160 mi. cost would only be <$4. Or <$7 per 300 mi. Just over 2¢/mi.

David70 | August 3, 2011

About the same here in Washington State Brian. I guess it's because a large enough fraction of our electricity comes from hydro and now wind. Of course, those are both actually from solar.

Brian H | August 4, 2011

You'd better hope as little as possible comes from wind. It's a kluge, very expensive if all actual (unsubsidized, fully-allocated) costs are included. And it plays hob with the distribution system as soon at it reaches or passes 10% of input to the grid.

Brian H | August 4, 2011

typo: "as soon as it reaches".

David70 | August 4, 2011

The wind turbines are already there. Hundreds along the Columbia between Washington and Oregon. Of course, a lot of times many are idle, even when the wind is blowing. Apparently not enough demand at those times. But you're right. It will be a very long time before it's paid off.

There are also many many in the Palm Springs area in California.

Brian H | August 5, 2011

Since they can't sell power profitably at market pricing without subsidy, the payoff time is "Never".

VolkerP | August 5, 2011


still disagree with you that most kinds of renewable energy sources are higher cost than coal, nuclear and natural gas plants. Iceland has attracted a lot of electricity-intensive industry for it's vast amounts of low-priced hydro power. Please don't discourage people from solar and wind - not one dollar spent on these is delaying the MPP research, right?

Brian H | August 6, 2011

Every dollar spent on them costs about $2 in lost opportunity elsewhere, and at least another $2 in downstream damage to the infrastructure. E.g.: Spain found ever green job created cost 2.2 elsewhere in the economy, and Scotland and Denmark found it was about 3.7. And that's just the start of it.

Brian H | August 6, 2011

typo: every green job

ChadS | August 26, 2011

And yet PSE in WA keeps building more because it's working out so well for them. Without subsidies! I guess you'd better call them and tell them that they're losing money, Brian...

Brian H | August 29, 2011

Because it's [stupidly] mandated, not because "it's working out so well". WA has a surplus of hydro already (which it doesn't count as "renewable"), and sometimes has to pay the wind farms to go off line because they're interfering with baseload. Other times, they're offline because the wind is too low or too high.

Wind cannot provide baseload, and when it exceeds about 10% of capacity, it rapidly causes more costs and problems with the baseload grid than it's worth.

hwye81k | August 31, 2011

TESLA, has anyone heard anything about this battery?

Brian H | August 31, 2011

well, that article is from last Feb. There were a number of new cars equipped with it to be built in March.
Here's another article from April:

Slightly higher energy density than the TM ESS, but about 1/10 as expensive! Imagine replacing the 300 mile battery for <$2,000.

Here's a current article, but without much new testing info:

hwye81k | September 1, 2011

Yes, I've seen those articles and more. I'm hoping that TESLA has too.

Timo | September 1, 2011

I'm a bit skeptical about that battery, first claim was over double the density, now it is only slightly better, so it might be that production cost is similar exaggeration if not direct lie. I'm hoping that my skepticism is not justified, but I don't believe that until I see something more than just claims.

ChadS | September 2, 2011

BrianH, some of your claims about PSE operations are incorrect. Some wind power (well, "renewable") is indeed mandated, but PSE is building more than that; that was my point. They are also targetting far more than 10% over baseload--and no, not because it's mandated. The stuff being thrown away is a BPA distribution problem that has nothing to do with wind "interfering with baseload"--and the wind farms are NOT paid for it. There is a lot of hydro in WA, but there is no "surplus" (at its spring peak it still doesn't provide everything)--and in fact the majority of PSE's power is NOT hydro.

I'm not pointing this out in order to claim that wind is a panacea. Wind clearly can't provide all of our power, and it doesn't work at all in some places (and some places where it could work it is mismanaged and doesn't). I am just reacting to "hope as little as possible comes from wind. It's a kludge, very expensive..." which, while true in some places, is not true everywhere. I agree with you that wind (like every technology) has some very over-eager proponents, but it doesn't help to dismiss it completely when there are places that it is working out. Hawaii looks to be another place where it makes a ton of sense.

Brian H | September 3, 2011

Wind varies. Between 0% and 500%+ of desired and usable levels.
It must therefore have 100% conventional backup.

Just use the backup, forget the wind. Much cheaper, and FAR easier to manage. And takes up 1% of the real estate and new transmission corridors.

jkirkebo | September 4, 2011

Hydro (with reservoirs) are what's most efficient to balance wind. When the wind blows, shut down the hydro and keep the water in the reservoirs. If it blows even more, use pumped storage to put water back into the reservoirs (~85% efficient). When the wind stops, start up the hydro (ramp time 0-100% <1 minute).

Norway buys lots of wind power from Denmark and export hydro power back. This works well for both countries.

Brian H | September 4, 2011

Capacity: one day's output. With a windless blocking high sitting on top of you for weeks (as happened to the UK last year) that will not even scratch the surface.