How Tesla increased charging speed beyond the simple power increase

JRP3 · Jun 1, 2013

I've been pondering the increased charge speed that cuts charging times in half even though the power increase from 90 to 120kW is obviously not a doubling. One possibility is pulse charging. I remember a talk by battery researcher Jay Whitacre, (founder of Aquion Energy), and he discussed how pulse charging might have benefits for lithium ion cells. When charging a cell the lithium ions need time to intercalate into the crystalline structure, higher currents forcing more ions in means some of them bounce around and cause a "traffic jam" of sorts. Think of taking a bunch of billiard balls and rolling them all at once to the same pocket, one or two might go in but the rest bounce back. If you take the same balls but shoot them rapidly one after another at the same hole they'll all go in quickly. I think pulse charging might allow similar behavior for the lithium ions and increase charging speed without artificially and prematurely raising voltage. Another possible benefit to pulse charging is a reduction of lithium plating if I remember correctly. Supercharging might actually be good for your pack.
Maybe.

RandyS · Jun 1, 2013

What I read in the transcript is that they changed the tapering curve when charging. Normally when the battery is first connected, the charge rate starts off higher and then tapers off as the charge progresses (and the battery heats up). It simply sounded like they left the charge rate higher and tapered if off later, thereby increasing the overall throughput...

JRP3 · Jun 1, 2013

I think more has to be involved. Using higher charge current will force the cell voltage higher in a shorter time period, high cell voltage damages the electrolyte. They have to be doing something more than simply holding the charge rate higher for a longer period. Something else must be allowing them to do so.

deonb · Jun 1, 2013

JRP3 said:
I think more has to be involved. Using higher charge current will force the cell voltage higher in a shorter time period, high cell voltage damages the electrolyte. They have to be doing something more than simply holding the charge rate higher for a longer period. Something else must be allowing them to do so.

I wonder if that something else would also allow them to charge a cell at the rate of filling up at a gas station...

jdevo2004 · Jun 1, 2013

Pretty simple really. An increase in charge power coupled with a decrease in distance between superchargers means that only half the time is needed to charge up between charges.

deonb · Jun 1, 2013

jdevo2004 said:
Pretty simple really. An increase in charge power coupled with a decrease in distance between superchargers means that only half the time is needed to charge up between charges.

No it's not that simple.

Look at cinergi's graph on:
http://www.teslamotorsclub.com/showthread.php/12078-Typical-Supercharging-rate/page6

He shows a charge graph from 20 to 220 that takes about 59 minutes. The Tesla published rate for 200 miles is now 30 minutes. That's direct doubling - not Tesla math.

smorgasbord · Jun 1, 2013

deonb said:
The Tesla published rate for 200 miles is now 30 minutes. That's direct doubling - not Tesla math.

It's Tesla Math because it's Tesla Ideal Miles.

I haven't been paying close attention, but it seems that Tesla has gone from 150 Ideal Miles in 30 minutes to 200 Ideal Miles in 30 minutes.

deonb · Jun 1, 2013

smorgasbord said:
It's Tesla Math because it's Tesla Ideal Miles.

I haven't been paying close attention, but it seems that Tesla has gone from 150 Ideal Miles in 30 minutes to 200 Ideal Miles in 30 minutes.

Ok, I'll buy that. Then cinergi's graph starts at ~25 ideal, adding 200 miles, we need to find 225 ideal. That becomes 193 actual. So original charging time to that point was 45 mins.

If all that happened was increase from 90kWh to 120kWh, then it would have been 33 mins. The newly rated one is 30. That difference is not significant.

Maybe 30 mins could not have been achieved at 120kWh, and the tango was needed to get it down to just be able to scale linearly. Or maybe the tangoed numbers aren't published yet. I guess we need to to have cinergi's graph updated for 120kWh to really be able to tell.

Good catch!

smorgasbord · Jun 1, 2013

225 ideal miles is not 193 real world miles.

How many real world miles does a fully range charged 85 kWh Model S go at 75MPH?

deonb · Jun 1, 2013

smorgasbord said:
225 ideal miles is not 193 real world miles.

How many real world miles does a fully range charged 85 kWh Model S go at 75MPH?

Sorry, I should have said EPA miles. (i.e. The miles that gets shown on the cinergi graph).

It's just a converting comparitor where we assume Tesla's 200 number is what you see on an "ideal range" setting, and cinergi's number is on the "epa range" setting. So just converting between two display units. Ideally we should have a direct battery charge % measurement, but sadly we don't.

Now whether or not you actually get 193 miles on a 193 mile display depends on your driving habbits... If you're Dave Metcalf, you would get you over 300 miles on that

.

AudubonB · Jun 1, 2013

Billiard balls - this post.
Air hockey - Mr Musk and his hyperloop.

Has Elon got some table game thing going? Should we be concentrating on ping pong ball aerodynamics next? :biggrin:

JRP3 · Jun 1, 2013

If we can get another graph from a 120kW charge session similar to what Cinergi did it would be pretty obvious what was happening. If the voltage curve stayed the same or spiked up some but not as aggressively as I'd imagine the current increase would be then they must be doing something extra. Or it is Tesla math and the charge speed really isn't increased as much as they are claiming then maybe they are just letting the voltage pop up higher sooner from the increased current and keep it there longer.

stopcrazypp · Jun 1, 2013

smorgasbord said:
It's Tesla Math because it's Tesla Ideal Miles.

I haven't been paying close attention, but it seems that Tesla has gone from 150 Ideal Miles in 30 minutes to 200 Ideal Miles in 30 minutes.

These numbers are consistent with the power rating increase from 90kW to 120kW:
200mi/150mi = 120kW/90kW = 4/3

gregincal · Jun 1, 2013

JRP3 said:
I think more has to be involved. Using higher charge current will force the cell voltage higher in a shorter time period, high cell voltage damages the electrolyte. They have to be doing something more than simply holding the charge rate higher for a longer period. Something else must be allowing them to do so.

I assumed they were just being conservative before, and now with a few months of real world testing of the superchargers decided they don't have to taper off as soon.

JRP3 · Jun 1, 2013

Yes I might be over thinking the whole thing and they've simply found that these cells don't have any significant problem with higher voltages for the duration of a super charge. If that's all it is then that might open the door to even faster charging using pulse charging, which could provide additional 30-40% speed increase, which might get us closer to the "faster than a gas refill" territory. It is a cell specific frequency tuning I believe that must match the chemistry and construction to get the pulse just right, but by avoiding localized heating and allowing time for ion diffusion seems like a promising concept. The same could be applied to regen braking, and on the discharge side as well.

Skotty · Jun 27, 2013

When I thought about this, I came up with another possibility. This is all guesswork, but it fits. First, assume early supercharging charged the entire battery all at once. If at the low end the Model S battery is actually capable of charging at a much higher rate than 120 kW, then it would be possible to not charge all cells simultaneously. You could charge them in groups (semi-random to avoid stressing part of the battery more than others over repeated supercharges). As you have to start tapering off the first group, you keep charging the first group at whatever rate you can, and then start pouring any remaining power into the next group. Repeat. This would allow some percentage of the battery to be fully charged before tapering off the overall power level being fed to the battery as a whole. You wouldn't have to decrease the charge rate until the last group reaches the taper point.

If not the case, then maybe it's done that from the start already, or maybe it's not feasible for some reason, or maybe I just gave Tesla another idea for faster charging. ;-)

TEG · Jun 27, 2013

I would think that full power charging is already stressing the C rate on the cells when doing the whole pack at once.
Charging different sections of the pack in separate groups doesn't really make sense to me.

jkirkebo · Jun 27, 2013

Skotty said:
When I thought about this, I came up with another possibility. This is all guesswork, but it fits. First, assume early supercharging charged the entire battery all at once. If at the low end the Model S battery is actually capable of charging at a much higher rate than 120 kW, then it would be possible to not charge all cells simultaneously. You could charge them in groups (semi-random to avoid stressing part of the battery more than others over repeated supercharges). As you have to start tapering off the first group, you keep charging the first group at whatever rate you can, and then start pouring any remaining power into the next group. Repeat. This would allow some percentage of the battery to be fully charged before tapering off the overall power level being fed to the battery as a whole. You wouldn't have to decrease the charge rate until the last group reaches the taper point.

If not the case, then maybe it's done that from the start already, or maybe it's not feasible for some reason, or maybe I just gave Tesla another idea for faster charging. ;-)

It's not feasible. If you terminated charging early you'd end up with a pack seriously out of balance.

JRP3 · Jun 28, 2013

Correct. You want to treat cells the same as much as possible.

I need to correct one of my earlier statements on pulse charging. The higher current pulse charging actually increases localized heating which helps speed ion diffusion, but because it's localized it does not raise total cell temperatures as much and does not have much effect on electrolyte stability. At least as I understand it.

arg · Jun 28, 2013

Note that there hasn't just been one change to this - they seem to have been tweaking the charge profile gradually over time. Reports such as http://www.teslamotorsclub.com/showthread.php/18251-A-record-Max-Supercharger-275-mph!?p=370282&viewfull=1#post370282 show that people are already getting higher currents in the middle range of the charge than reported on Cinergi's graph, even though they are not yet getting 120kW. And that graph itself showed higher currents than were being reported in the early days of the Supercharger roll-out.

How Tesla increased charging speed beyond the simple power increase

Hyperactive Member

Fan of Elon

Hyperactive Member

Active Member

Member

Active Member

Active Member

Active Member

Active Member

Active Member

One can NOT induce accuracy via precision!

Hyperactive Member

Well-Known Member

Active Member

Hyperactive Member

2014 S P85 | 2023 F-150L

Teslafanatic

Model S P85+ VIN 14420 EU

Hyperactive Member

Active Member

Similar threads