If I Ran The Zoo: Designing a charge connector for the future

[AudubonB]

...But there should be no exchange between cord and port. Too much potential for leaks and negative consequences.

I think that's a reasonable concern. However, you may wish to remember (or learn for the first time) that, when Tesla was demonstrating/toying with the idea of Model S battery swap stations, that included use of quick-disconnects for the vehicle's own coolant system between car and battery pack.

When you drill down to the space available in a cable charger....I agree: we're getting to pretty small sizes.

 

[KarenRei]

Nix the coolant connection.

Without external-provided coolant, charge rates will never get down to those approaching gasoline fill times; there is simply too much heat released during high powered charging, and it's unreasonable to expect efficiencies to rise to the degree needed to prevent that. While long charge times may not bother us, does bother a large portion of the population. See Audubon's post for a good alternative to coolant intermixing.

Coolant for the cord, if there is to be any

Again: if you want higher power charging, and you don't want the cord to be so massive that you can't handle it, this needs to be done.

It's easy to not "want" to handle coolant. But the simple facts are we're dealing with large amounts of heat already, and when you're talking about multiplying the power, you're talking about multiplying the heat. And it's even worse than linear, because as you increase the power and/or push harder during equalization, you lower the efficiency and turn even more of the power into heat. It may seem desirable to have the cooling system completely embedded in the car, but you can't put an industrial-scale chiller on the vehicle without negative consequences.

Power should be throttled at certain temperature readings, and some app feedback or other indication should be provided when power is being throttled by plug heat. Maybe that already exists. I don't really know.

Already exists.

 

[arnis]

And then a failed cell means 2,7% degradation instead of 1,3% degradation.

This literally describes how far are you from truth.
Failed brick means the pack is ruined. One failed cell in those 7104 actually ends with unacceptable results.

Yes, but it can only charge a 400V vehicle at 400V, and thus the charger is operating at half capacity. That extra 400V is worthless to essentially all real-world EVs.

And your water connector? Try to be critical over your own creations before all else. It's not nice.
24V lightbulbs semi-truck use are essentially worthless to all real-world cars. Aren't they.

 

[KarenRei]

This literally describes how far are you from truth.
Failed brick means the pack is ruined. One failed cell in those 7104 actually ends with unacceptable results.

Wrong. And it's cell, not brick (quote: "And then a failed cell means 2,7% degradation instead of 1,3% degradation."). A brick is an entirely different thing from a cell. If you reduce the number of cells in parallel, then the failure of a cell has a more significant impact toward degradation, because you have fewer alternative paths for current to flow through.

And your water connector?

First off, coolant is not water. Glycol is the most common (including in Tesla). Most EVs (excepting passively cooled ones) already have coolant systems that can be readily adapted to use a heat exchanger. They cannot use 800V without major changes.

Lastly, providing external cooling is something Tesla is actively working on. Tesla recognizes the need for it, whether you do or not.

 

[AudubonB]

====> I FAIL TO SEE any reason for unpleasantries, overt or implied, in this thread.<=====

 

[miimura]

I really like the idea of the direct refrigerant cooling in the i3 battery pack. I think they should do something similar to cool the Supercharger handle. You can use very small diameter high pressure tubing because the flow rate is very low. If you're worried about the potential refrigerant leakage, use CO2. Compressed and pre-cooled C02 flow is proportional to cooling demand at the handle and cooling is concentrated right at the handle by fluid expansion. Residual cool gas cools the cable on the way back to the central compressor.

Let the car worry about cooling its own battery pack.

 

[skitown]

Again: if you want higher power charging, and you don't want the cord to be so massive that you can't handle it, this needs to be done.

Anaconda-sized Snake Charger to the rescue.

 

[KarenRei]

Anaconda-sized Snake Charger to the rescue.

That snake looks neat but was never a very practical design Tesla's in-ground charger is much more reasonable.

Note that irrespective of mass, it provides coolant. Because reducing charge times means getting rid of a lot more heat.

 

[scaesare]

Interesting discussion, thanks for kicking it off. And it is unfortunate folks are getting a bit snippy here... after all we are just spitballing, and exchanging ideas is a good way to advance the idea... it would be nice if everybody played nicely.

I can't find anything concrete, but based on some pics of the connectors, and the pumps and lines in the car, I'd guess that the existing coolant lines are somewhere in the neighborhood of 1/2" ID. That's apparently sufficient for the flow necessary to handle today's ~120kW charging.

I wonder how difficult it would be to get the required cross-sectional area needed for significantly higher charge currents integrated in to the connector in a compact manner.

 

[electracity]

Tesla's in-ground charger is much more reasonable.

I don't see how a ground mounted charger could work in areas where it snows. Even in a garage the crud that comes off a car in winter would be a problem.

A sophisticated high voltage system with a possible cooling loop and crud are not a good match.

Even in a non-snow climate I would rather plug a car in rather than keep a ground mount unit clean.

 

[electracity]

Yes, but it can only charge a 400V vehicle at 400V, and thus the charger is operating at half capacity. That extra 400V is worthless to essentially all real-world EVs.

You could design a battery pack to switch the sub packs in the car between parallel and serial: 4 x 200v sub packs or 2 x 400v sub packs.

 

[KarenRei]

Looking at Tesla's patent diagrams, it looks like Tesla's battery coolant is already isolated from the rest of the system by a heat exchanger which services three separate loops:

It's easy to picture working a fourth loop, from the charger and back, into that exchanger. And that seems to be implied with this drawing from their in-ground patent:

I don't see how a ground mounted charger could work in areas where it snows. Even in a garage the crud that comes off a car in winter would be a problem.

I'm not picturing the problem (and I've lived in cold areas my whole life). Snow and ice are not going to resist the force of pneumatic or electric linear actuators, and electric heaters on the connector are nothing compared to the powers available at hand. Of course whether it's in-ground or not is not the relevant part - it's - when one is discussing auto-connecting charge port s- whether to use a three-axis positioning system, or a bunch of tiny actuators to form a snake. The former is much simpler, which is why 3d printers don't print with snake arms As for positioning of automated , it could be in the ground, to the side, in front, in back, even above - doesn't really make a difference.

That said, auto-connecting charge ports are a bit off-topic; provision of coolant (reiterating: necessary for higher charging powers, because heating increase faster than linear with charging rate, particularly at the end of charge) is an issue entirely independent of auto-connection of the charge port. I only raised Tesla's patent to show that they're actively and presently working on the concept.

For what it matters, here's the current size of the radiator unit (not counting the compressor and other components on the coolant supply line):

The general public wants to see EV charge times be 5-10x faster. I regularly chat with people outside of "EV Land", and the concept that a charge might take 30-40 minutes in good conditions to 80%, and much longer to 100%, is almost unthinkable to them. Yes, I of course know the counterarguments to that (you rarely need to use fast chargers, you need to stop anyway, etc, etc), and I dutifully cite them, but that doesn't assuage people who've been trained by a lifetime of gas station fillups and whose minds automatically create worst-case scenarios. Yet even halving charge times would require tripling the size of that cooling system if you're trying to keep cooling entirely onboard. It's just not practical. Just work another an independent coolant route into the heat exchanger from the charger and you can vastly reduce the size of that unit. And the mass per car. And the cost per car. And decrease the energy per mile the car has to spend due to weight. Which on its own means slightly faster charge times.

 

[electracity]

Today's EV batteries (20-100kWh) are not capable to charge considerably faster no matter the thermal management.

How fast could current cars charge with no thermal or amperage imitations? Considering just the cell C rate at optimal temperature.

 

[KarenRei]

How fast could current cars charge with no thermal or amperage imitations? Considering just the cell C rate at optimal temperature.

It's limited only by ion mobility. Which is an adjustable parameter, balanced off against other parameters such as energy density. It's possible to make li-ion cells whose ion mobility allows for charges in a matter of seconds. But ion mobility isn't the limiting factor, the limiting factor is removing tens of kilowatts of heat (the heat energy of dozens of portable space heaters, or a good-sized fire) from the car.

Once you get to the end of a charge and have to get to cell balancing, heat removal becomes even more critical, because cells that are at a higher state of charge will convert more of their incoming energy to heat while other cells still need more power.

 

[KarenRei]

To put it another way. Let's say you want to put the energy of a third of a tank of gasoline into a car, in the length of time it takes to put gasoline into a car. And let's say that you can manage to do it at 90% efficiency. The amount of heat you're releasing is still equivalent to dumping 10% of the gasoline under the car and lighting it on fire, with a fan fanning the flames so that it burns entirely in a raging firestorm during the time you're filling up the tank.

If you're moving a lot of energy, even a small percentage of waste heat is a lot of heat.

 

[electracity]

A critical piece of information that only Tesla knows is supercharging frequency. What does the distribution of "on the road" charging look like in 2017 for S/X owners?

I suspect that a surprising number of S/X owners have not used a supercharger this year.

I assume at this point that a robotic supercharging connection is for perception and show, not to address a real problem. The only real benefit is for people who are supercharging and wish to not exit the vehicle.

Obviously Tesla has planned for an independent external coolant loop. I'm not sure that they are sure that it will be implemented.

One thought on a robot charging arm, as opposed to floor charging: The car could do the sensing with HW2 and tell the robot arm that there is no person or object near the charge port. Floor charging requires a separate system to verify a safe environment.

 

[Skotty]

Without external-provided coolant, charge rates will never get down to those approaching gasoline fill times; there is simply too much heat released during high powered charging, and it's unreasonable to expect efficiencies to rise to the degree needed to prevent that. While long charge times may not bother us, does bother a large portion of the population. See Audubon's post for a good alternative to coolant intermixing.



Again: if you want higher power charging, and you don't want the cord to be so massive that you can't handle it, this needs to be done.

It's easy to not "want" to handle coolant. But the simple facts are we're dealing with large amounts of heat already, and when you're talking about multiplying the power, you're talking about multiplying the heat. And it's even worse than linear, because as you increase the power and/or push harder during equalization, you lower the efficiency and turn even more of the power into heat. It may seem desirable to have the cooling system completely embedded in the car, but you can't put an industrial-scale chiller on the vehicle without negative consequences.



Already exists.

Note that I wasn't saying no to coolant. I was saying no to coolant channels between cord and plug on the car (a feature which appeared was part of the concept plug -- correct me if this wasn't the case). Have the car side provide it's own cooling for the plug if needed, and have the charge station side provide it's own cooling for the cord if needed.

 

[gavine]

 

[Skotty]

A quick thought on ground charger. I don't like this idea due to maintenance issues. Generally you want to keep everything off the ground in a public lot. Dirt, grime, etc. would be a problem. Maybe it would be okay if using wireless, but it would still be a problem for lot service. Road crews won't want to mess with road that has stuff embedded in it. Tesla might would have to provide their own paving crews for the lots with Superchargers in them. Maybe if the stuff in the ground could be easily turned off and easily removed when road service is needed, but I think it would be problematic.

 

[arnis]

@gavine, thanks for finding something appropriate. I have a comment under that video, 4 months old.

How fast could current cars charge with no thermal or amperage imitations? Considering just the cell C rate at optimal temperature.

There will never be 100% efficient chemical battery. And the core (center part) of the cell will get too hot, even if you have
unlimited amount of cooling around the cell available.
Though Ioniq appears to have the fastest C-rate. 28kWh battery that charges at 2.5C until 80% SOC (70kW rate)
Imagine Tesla's 100kWh battery charging at speed up to 250kW. Something similar.

Without external-provided coolant, charge rates will never get down to those approaching gasoline fill times; there is simply too much heat released during high powered charging, and it's unreasonable to expect efficiencies to rise to the degree needed to prevent that. While long charge times may not bother us, does bother a large portion of the population.

Well, without going above 400V, charge rates will never get down to those approaching gasoline fill times.
Accidentally, video above explains numbers, spot on.
I'm working at ABB right now. I'm dealing with low voltage systems (400V AC). I have an excellent visual example, what we need to get 4000A running (though this device is 3-phase, imagine only 2, also DC is more demanding, but let's forget that).
Width of each Cu-bar is 10cm (4"), thickness 1cm (0,4"). Each handles 1000A, MAX, at no more than 35*C air with passive airflow.
I'm unable to lift 4 bars stacked on the table without taking it seriously. Copper is 9x heavier than water.
Yes, it is possible to run coolant, but this liquid is not Helium and actually flow must be significant. Two hoses inside the cable.
My good guess would be 3/4" hoses. I leave you with hometask: 50:50 ethylene glycol mix and normal maximum flow by coolant pump without turbulent flow. Try to extract any number of kW of heat you desire. Let's say 10kW extraction rate from vehicle. Choose your hose dimensions.
Yes. It is called science. No science, no real thing. Just speculation.

And this is why CCS standard (adopted by US and EU) is designed to go up to 1000V.
Nobody forces manufactures to stay at 400V. Though sky is not the limit, 1000V is.