Fast Charging - What is the secret?

Pedrodemio

100 W
Joined
Jun 30, 2015
Messages
115
Location
Brazil
As an enthusiast for all things electric, I've always wondered what allows fast charging

Lets take the Model 3 as the prime example at the current date, capable of a 250 kW peak charging rate, with a 96s46p, taking into account that that peak only happens at low state of charge, we are taking about 15 A charge current per cell, in a lot of information around the web and on this forum, we see cells taking a beating on cycle life at much lower currents, probably due to keeping a high charge current up until the CV phase

Now, one fact is clear, they only can do that with a hot battery, if we take this video into account, it's around 43°C that full speed is reached, and the temperature is allowed to climb even higher. Unfortunately I haven't found the same data for a Supercharger V3 that can deliver the full 250 kW, maybe the full speed indeed need closer to 50°C, my guess would be explained due to Arrhenius equation in allowing the reactions to occur faster

https://www.youtube.com/watch?v=SVm3muFrJbA&feature=youtu.be

Now to the question, if we were to get some other high energy cells, i.e. Sanyo NCR18650GA, Samsung INR18650-35E and INR2170-50E and many others, have a proper temperature control system capable of both heating up the cell to these high temperatures to prepare for charging and also cooling them down fast after charge is complete to lower calendar degradation rate, and probably doing a bunch of tweaking on the actual charge curve profile and temperatures, would be possible to achieve this fast charge rate with little effect to calendar life on other cells or there is more to it?
 
Getting the cells pre-heated even to borderline damaging temps

does reduce the damage (lost cycles) inflicted by high rate charging.

I think it would be very hard to push these limits the way Tesla engineers can.

But there are use cases where it's worth trying, fast charging being essential to the use case.

Just don't kid yourself that there aren't high numbers of cycles getting sacrificed off the lifetime back end.

And actual quantified data, much less instructions, I bet are closely guarded proprietary info, in our domain extrapolation would involve a lot of guesswork and intuition.
 
The first e-bikes using super capacitors as energy storage start to appear on the market. They can't beat lithium considering energy density, but it's sure something to keep an eye on. Those super capacitors can also charge very fast. I wouldn't be suprised if we will see hybrid batteries in the future being partly super capacitor and partly lithium. The super capacitors are very good at peak currents and they could even partially charge the lithium cells and minimise their self discharge problem like that.
An air conditioner to control your battery temperature doesn't seem very practical on an ebike imho.
 
obcd said:
The first e-bikes using super capacitors as energy storage start to appear on the market. They can't beat lithium considering energy density, but it's sure something to keep an eye on. Those super capacitors can also charge very fast. I wouldn't be suprised if we will see hybrid batteries in the future being partly super capacitor and partly lithium. The super capacitors are very good at peak currents and they could even partially charge the lithium cells and minimise their self discharge problem like that.
An air conditioner to control your battery temperature doesn't seem very practical on an ebike imho.
John Deere claims the new electric batteries under development (possibly as soon as 2022) for their 8R series tractors can be fast charged in about the same amount of time it now takes to fill up with diesel fuel. Most farms have power line access on at least one side of every farm field whether its a 640 acre section or just 20 acre field. Imagine seeing field charge stations (lock&key w/access code) adjacent county road powerlines bordering fields for Fast Charging an eJD ... nothing runs like a Deere.
 
2022 is still 2 years away.
The e-bikes with super capacitors instead of lithium batteries are already on the market.
And a John Deere to go to work is a bit expensive.
 
john61ct said:
Getting the cells pre-heated even to borderline damaging temps

does reduce the damage (lost cycles) inflicted by high rate charging.

I think it would be very hard to push these limits the way Tesla engineers can.

But there are use cases where it's worth trying, fast charging being essential to the use case.

Just don't kid yourself that there aren't high numbers of cycles getting sacrificed off the lifetime back end.

And actual quantified data, much less instructions, I bet are closely guarded proprietary info, in our domain extrapolation would involve a lot of guesswork and intuition.

Really good points, I imagined the amount of cells and packs they have running on test equipment to get on the charge curves they use

One way would be to actually use Model 3 cells, albeit expensive, they are starting to appear on the market, and focus the effort on actually extracting a detailed charge curve taking temperature into account, I would need a Model 3 for that :lol:

obcd said:
The first e-bikes using super capacitors as energy storage start to appear on the market. They can't beat lithium considering energy density, but it's sure something to keep an eye on. Those super capacitors can also charge very fast. I wouldn't be suprised if we will see hybrid batteries in the future being partly super capacitor and partly lithium. The super capacitors are very good at peak currents and they could even partially charge the lithium cells and minimise their self discharge problem like that.
An air conditioner to control your battery temperature doesn't seem very practical on an ebike imho.

Do you have a link for that? I quick google I haven't found anything comercial

I don't see it happening soon, with batteries getting a higher energy density every year, there is a lot of catch up for the super capacitors to do

Even as a buffer they make no sense on the current state, once you have enough of them to actually increase your peak power output, you would be better off using a bigger battery that would give you that extra power and also more range

And yeah, on a eBike it makes no sense, but on a motorcycle it starts to get appealing, the offers we have today for them have way too low range for a relative high charge time when comparing with cars
 
Internal resistance of the battery, heat and the ability to remove it, and battery chemistry are the main factors that influence fast charging.
 
The lithium chemistry and electrode are important factor imho. Lithium titanate can be charge very quickly and has incredible cycle life. Lithium iron nanophosphate cells from A123 probably have a decently high charge rate. Otherwise... NMC high discharge cells can probably take a pretty decent charge rate too.

Matador
 
There was a recent thread titled

How to fullfil LTO's promised 10-minute recharging?

by a noob wanting to tour Europe without stopping for coffee.

Can't find it now, had some good LTO discussion.

Terrible energy density though, but seems good for public buses. . .
 
The "ostrichoo julius" seems to use super capacitors as battery.
Besides the fact they can charge in 30 minutes, they also seem to be less sensitive to temperature fluctuations.
They also lower the possibility to set your house on fire and can be recharged 2000+ times.
Their disadvantages at the moment are their lower energy density and the fact they are self discharging.
I can't find the originial youtube video at the moment. But searching on youtube for "ebike super capacitor" should point you in the correct direction.
 
Regular capacitors can charge somewhere on the order of tens to hundreds of microseconds. Imagine moving hundreds of Wh into your ebike that quickly. You would need a really trick power supply to say the least.
 
Personally I think the "capacitors as EV storage" idea deserves its own thread.

Ideally with links to proven products already on the market.

That way this thread can remain useful for those interested in discussing fast-charging **batteries**
 
fast charging at the level that tesla does is not relevant for home gamers. the temperature managment system is impossible to recreate and as other car companies alreayd have proven: impossible to equal.

if -as a home gamer- you need fast charging at the level that tesla does you need to reconsider our cell choices and battery size.
 
If you're sitting around waiting for the battery to charge, then you're not using your EV properly, or you're not using the proper EV. You should charge when it's convenient for you, so your only time spent charging is to connect and disconnect the plug, since inductive charging isn't efficient, and it won't be long at all before something robotic handles the connecting and disconnecting for you.
 
Those on 1000mile road trips unable to afford also owning / renting a second ICE powered vehicle presumably excepted?
 
The max stated charge rate is continuous and doesn’t take the varying ability to take a charge happily at different states of charge. If a cell is stated at 8amp max charge rate maybe could do 12 or something till 2/3 full or something. Never seen studies on how varying the current at different soc effects but I know that’s how the big companies charge their cars so fast.

In my experience throwing 3v and 4v 30q cells in parallel they didn’t even noticeably get warm. I forget the current that passed based on the resistance but it was way higher than the stated max charge rate. I forget all the numbers but I was impressed. At 3v I bet u could do a huge charge rate before any damage from heat. I think dendrites still can form though and destructive and dangerous in that way but not from the heat. I think.
 
John in CR said:
If you're sitting around waiting for the battery to charge, then you're not using your EV properly, or you're not using the proper EV. You should charge when it's convenient for you, so your only time spent charging is to connect and disconnect the plug, since inductive charging isn't efficient, and it won't be long at all before something robotic handles the connecting and disconnecting for you.
A few notes:

1) Inductive charging is pretty efficient. It's the cost that's going to drive ebikers away from it. (but may work well for EVs)
2) Robots are even more expensive.
3) Many of us (me included) choose charging times/regimes to maximize battery life, not convenience.
 
I’ve seen at best maybe 80% efficiency inductive charging. If charging a vehicle at any decent rate that could be a huge amount of heat produced n seems good for a slow rate only. Electricity is cheap but that heat needs to be removed or could snowball into bad news

Then again maybe could just heatsink the coils and no problem at higher current.
 
Hummina Shadeeba said:
Never seen studies on how varying the current at different soc effects
The chemistry will "accept" (actually pull / demand) at a **much** higher rate than what is conducive to longevity, right up to nearly Full, after the CC-to-CV transition.

Unless the time saving is critical, best to stay well under the vendor max rating even in T-shirt ambients.

Any noticeable warming means you're sacrificing cycles off the back end.

Harmful dendrite formation is well past that point.

The overall hit to longevity may not be **that** much in a use case with high **discharge** C-rates, unless we're talking at cooler cell temps

but IMO best to be conservative if you're trying to get best lifespan possible.
 
john61ct said:
The chemistry will "accept" (actually pull / demand) at a **much** higher rate than what is conducive to longevity, right up to nearly Full, after the CC-to-CV transition.



Harmful dendrite formation is well past that point.

but is it necessarily going to reduce cycles when charging at a high rate when at low soc? i forget tesla's regime but how detrimental is it?

I thought I read dendrite formation can occur without any noticeable heat or when the cell temp is controlled

dendrites seem the boogie man and how often has a cell shorted out from dendrites? when a pack fails it seems more often something else. if a dendrite branches from one side to the other and shorts the cell it sounds extremely dangerous but super rare.


charging regimes have a lot of variables but the standard way of fully charging and possibly leaving it like that for many hours isn't a good start if trying to increase cycles. i didn't search much but just a hunch that youd do just as well if not better fast charging to 80% and never going higher. maybe theres some decent comparisons around even on here but haven't seen them.
 
Hummina Shadeeba said:
but is it necessarily going to reduce cycles when charging at a high rate when at low soc?
Low or higher SoC, say 20-90% range does not matter.

It is cell **temperature** that makes the difference,

if you want to optimize longevity

avoid fast charging unless you have a compelling reason.

If you do then only do it when the cells are at least warm.

If **hot** then less damage / faster rates.

And yes, only charge prior to driving, do not sit at high SoC. Colder temps are better then :cool:

And yes stopping at below 4.0Vpc or even lower is also conducive to optimizing, longevity **if** your range is not an issue.

The **real ideal** is centering your SoC/DoD around the 3.7V (or whatever for the chemistry) resting midpoint, and using as little capacity as possible.

If you cycled 0.5V that way for shopping trips only, the bank would probably last thirty years and over a million miles.

 
john61ct said:
Those on 1000mile road trips unable to afford also owning / renting a second ICE powered vehicle presumably excepted?

is owning a useless second car worth the extra hour or so you spend on breaks/charging worth the cost for those once a year trips?

yes, a EV only suitable for a certain percentage of the population, that percentage is just a lowly 95%. and because its not suitable for 5% of the population nobody should buy a EV.

if you travel with wife/kids you will have to stop way more often because of pee breaks then a EV can run for. just combine the pee breaks with fast charging and you dont lose any time compaired to a ICE car. as if adding a few hours on a 2 dozen hour drive is such a big issue, you are on frigging holiday, its not a race. enjoy the trip.

ps: you can drive 1000 miles for like 20 bucks in electricity. i dont car how much money people have, saving hundreds of bucks on fuel is worth the added time.
 
JackFlorey said:
...Many of us (me included) choose charging times/regimes to maximize battery life, not convenience...

I do too, except that I've gotten lazy (now that I only charge to 4.05V/cell except for rare long trips) about using a timer to schedule top of charge to coincide with leaving in the morning. I only charge 2-3 times a week (often only partial charges) and use a pretty slow charger, so at most the battery sits at full for a few early am hours. With my primary pack of 44ah that I typically only discharge by 20-25ah, I have great convenience, never have range anxiety with over 100km of range at 80kph cruise, and use my pack super conservatively for long life, all the while enjoying performance that exceeds 90-95% of the smoke belching motos I run across. I love creating safe space well in front of all cars and motos which also demonstrates the performance potential of electrics. Then when if they catch up at the next light I let them know how economical it is to have an amusement park ride with no lines and just pennies a ride.
 
Matador said:
The lithium chemistry and electrode are important factor imho. Lithium titanate can be charge very quickly and has incredible cycle life. Lithium iron nanophosphate cells from A123 probably have a decently high charge rate. Otherwise... NMC high discharge cells can probably take a pretty decent charge rate too.

Matador
Found the "10min LTO" thread
https://endless-sphere.com/forums/viewtopic.php?f=14&t=104319
 
flippy said:
fast charging at the level that tesla does is not relevant for home gamers. the temperature managment system is impossible to recreate and as other car companies alreayd have proven: impossible to equal.

if -as a home gamer- you need fast charging at the level that tesla does you need to reconsider our cell choices and battery size.

I wouldn't say impossible, but definitely a huge engineering feat, doable. Oh, and also expensive to make a one off, it will be a significantly percentage of the pack cost

I've been digging a bit on a bunch of papers, lots of cool methods to build a charge curve that doesn't affect cycle life and and make less than one hour charge possible. Unfortunately I haven't found anything that specifically that takes battery conditioning (heating in this case) in account like Tesla does

The starting point would be select a few candidates cells, probably based on the Li-ion cells cycle ageing being done by Docware (I've said before, but this endeavor is amazing, thanks for undertaking it)

Now we need to build a heating and cooling jacket for each cells and tie that to a cycle tester. First do a non thermal managed cycling following various methods from the papers I've been looking, and for each charge profile, do it again with various temperature steps and followed by rapid cooling after the charge is complete

So even if we can narrow down to 4 cells model before doing anything, we will probably need 100 different cycling channels, not something my pocket can afford

Or narrow down to one cell and focus in what is the best we can get with it
 
Back
Top