Tanktwo’s Refillable Battery Makes Us Think It’s April Fool’s Day Again, But It’s Not

MAY 9 2017 BY MARK KANE 35

Tanktwo’s refillable battery cell

If you ever thought that you knew all the battery storage systems for electric vehicles, including battery swap concepts (think defunct Better Place), Tanktwo will surprise you.

Its idea is so ridiculous, that it makes us curious.

Tanktwo proposes to use egg-shaped battery cells that would randomly fill the pack tank, and using a real-time algorithm, will connect with each other in various configurations (taking into consideration weak cells), building working strings.

You fill as much as you need, and on top of normal charging capability, you can refill the tank in three minutes with fresh eggs batteries.

Why we didn’t think of that earlier?

String Cell™

Tanktwo’s Cell Swap

The string battery solution is based on a distributed and modular design that consists of automatically self-organizing cells, which are called string cells.

The string cell is a modular and independent unit inside the string battery. A single string cell consists of plastic enclosure, an electromechanical cell, a number of contacts, and internal processing capabilities. Through the conductive material on the surface of the enclosure, string cells form contacts with one another. The terminals of the internal electrochemical cell can be connected to any of the contacts, controlled by the internal processing unit.

String Battery™

A string battery is a type of rechargeable battery that consists of many individual electrochemical cells placed inside a battery enclosure in random order.

Tanktwo’s String Battery

The individual cells and the enclosure form a dense, random electrical connection network that can be reduced to create an optimal cell connection circuit which typically consists of many series connected cell strings.

Cell Swap™

Cell swapping is a method of replacing electrochemical cells placed inside the battery enclosure. To enable cost efficient cell swapping, the individual cells have to support some type of automatic removal, installation and conveying methods. Commonly cell swapping is used to replenish the battery with cells that contain more electricity, have better qualities to store electricity or have better performance.

Tanktwo’s refillable battery cells

Categories: Battery Tech


Leave a Reply

35 Comments on "Tanktwo’s Refillable Battery Makes Us Think It’s April Fool’s Day Again, But It’s Not"

newest oldest most voted


They must be late with their April 1 story…

We’ve seen this idea promoted before — altho with cylindrical “tiny cells”, not egg-shaped ones. The previous incarnation only advanced as far as a very amateurish website; this one actually has a promotional video.

Apparently there is no idea for batteries so impractical and, frankly, stupid that there won’t be at least two attempts to promote it. 🙄

So I guess, according to this video, the idea is to have the depleted cells sucked out of the tank with a vacuum hose, before the fresh cells are squirted in?

Sure, what could go wrong there?


I’ll take the 3 minute battery swap instead.

I think the idea is very innovative, and extremely interesting. It may very well be that strings are just not going to be practical. But it is clear the potential advantages are huge. Packs as they are made today are no better than their weakest cell (or the set of cells connected in parallel, acting as one) so when you got to put 96 of them in series to get into the 400V territory where today’s EVs are (and must be if they are to use DCFC) you are almost guaranteed to have a significant amount of capacity in the best 90 cells in that stack that cannot be exploited because continuing to discharge the pack would force the weakest cell too low if you did, permanently destroying it and thus the whole pack. The things about these strings are that every cell is an actively managed “minipack” with a BMS onboard. Sounds like expensive, but this isn’t a hard task for a cheap microcontroller, so the clever electronics is nearly free, adding only cents per cell. Which means you can continue to draw power from those cells that still have any but not draw from those that are depleted.… Read more »

“…you are almost guaranteed to have a significant amount of capacity in the best 90 cells in that stack that cannot be exploited because continuing to discharge the pack would force the weakest cell too low if you did, permanently destroying it and thus the whole pack.”

Or, you know, the EV could have a properly designed BMS, one which will identify bad cells and cut them out of the circuit to prevent that very problem.

Definitely thinking out of the box!

Er… out of the carton.

Well.. I like to be opened minded. So even though this seems ridiculous.. SHOW ME a working prototype… then we’ll talk.

What, you’re not impressed by a transparent box full of blinking lights?

I’m shocked, SHOCKED I say! 😉


Almost there! Another couple mil and it’s super batteries for all! By the way, the small contact area between spheres guarantees low amperage. Ok for leds, but not so good for running large motors.

What happens when the car hits a bump and all the string cells shift inside the container? Wouldn’t there be some serious electrical arcing or heat (over time)?

Sure, this concept will not work in a car, as it will not be possible to get high currents over the contact points without high losses and thereby heat, as well you need high power switches inside these cells, which are big, unreliable and extremely expensive. The cost for smart cells is only the least of its problems.

As well there are better ways to handle the weekest cell problems in batteries like charge transfer from stronger cells.

Nevertheless i like out of the box thinking. 99 out of 100 ideas lead to nothing, but one will change the world.

And some will lead to a cascade cell rupture and runaway chemical fire. Let’s hope the NHTSA shuts them down before this ends up on the streets.

The out of the box thinking is admirable. But with most new ideas, there are kinks and hopefully these kinks that can be resolved safely.

Although I agree with the ethos, there’s a reason the contactors on battery cells are bolted into place. I’d rather have a well-engineered pack than some “pump” technology. Besides, you’re forgetting all about cell health.

I have no doubt that a small team of competent electrical and mechanical engineers could come up with a very long list of reasons why this idea is utterly and even ludicrously impractical.

I also have little doubt your objection would appear on that list.

Too many fundamental issues created to try to make the battery act more like a gas car.

1) Thermal system?
2) Low resistance electrical connections?
3) Working Voltage Range?
4) Mechanical stability during shock/vibration?
5) Infrastructure?
6) Cost?
7) Energy Density?

I too wondered about thermal management. I think they need to make the cells 1/10 the size, and add cooling. If they were in some kind of non-conductive liquid, and small enough, it would be easy to pump them in/out of a car, and maybe you could pump the liquid around them, while in the tank, to cool them.

At the end he said 300VDC, but with the technology, it seems like they could instantly change the voltage to whatever they wanted by re-routing the connections.

To be sold by the dozen.

And that’s no yolk.


Charged EV Magazine had an article about this a long time ago. Not even remotely new.

Yep- I saw that.

If one is sufficiently into this line of journalism, then I get it. Batteries are a (or the) main point of separation between electrics and gas-guzzlers. But still, battery technology and research development are not in itself equivalent to the EV takeover we all foresee.
Nor is the Tesla brand for that matter.
Context is everything, but can unfortunately be overlooked.

Because sometimes you just need a little eggstra juice.

Who knows? The Quant/Quantino ‘impossible’ cars were at Geneva [was it?] again, driven by motoring journos, Top gear had a go last time but nobody could figure out the con. Supposedly drove 1400 km on a single fill of their magic bio-fluid, etc, etc.
Myself I don’t believe batteries or supercaps are the ‘final solution’ due to weight + other factors.
It’s impressive that so many inventors are trying their hands [& spending cash] at off-the-wall ideas; someday, someone will come up with a better answer than batteries. Or, less likely a battery with 5x the storage capacity at 1/5th of the weight…

If the battery overheats, will the cells get fried?
If I drive on rough roads, will the cells become scrambled?
Is this idea real, or is it just a shell game?
I mean, are these batteries all they’re cracked up to be? Could somebody egg-splain?!

I’d like to believe… but I’m chicken.

It’s an original idea but it should have stayed on the beer mat it was thought up on. The energy density of this system must be terrible, if you want battery swapping it would be better to use a Tesla type system.

I saw this idea before, and it made me smile again.

Oh, the wonderfully wonky out-of-the-box thinking! It lightened up my day and so it was worth every penny poured into this effort.

The creative spark may hit someone else and spawn an idea that does make sense from an engineering standpoint.

Looks like Yates and the Navy may have an alternative way for in flight recharging.


Although it may be quite a challenge to have those ball batteries flowing through an aerial refuel type hose in both directions. Or then they would need to do a double hose, one for removing the balls and one for the new recharged ones.

In theory it could work, or at least it would be more easy as a mid air big bulk battery replacement.

Note that if money is no object they could just drop the empty ball batteries end have a simpler unique hose to just bring the new balls.

Here is a picture of what is used now for kerosene transfer:

And what Yates uses for recharging in mid air:

A ball hose would be the kerosene type hose to transfer charged ball batteries. This could perhaps go faster indeed.

They could also use pneumatic tube cylinders:


Each cylinder would be a battery and they would arrange in strings in sub tube sections. This would avoid the random aspect and be more similar to what Tesla is doing with 18650 cells. They would be in sub strings of 6 assembled per 74 tubes for a total of 16 modules making a pack.

It’s amazing how many people think using the word “algorithm” explains all technical issues about all technologies. But this may be a world record…