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

[KarenRei]

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.

Except that the rate of dissipation from the core to the outside is related to the temperature differential between the inside and the outside. So there's a world of difference between chilling your coolant to ambient vs. to just above freezing. So, unless you're planning to install a 10 ton industrial chiller in the vehicle.....

And this is, by the way, an entire separate, additional issue to the issue of getting the heat out of the pack coolant.

Imagine Tesla's 100kWh battery charging at speed up to 250kW. Something similar.

And giving off over double the heat.

Well, without going above 400V, charge rates will never get down to those approaching gasoline fill times.

Simply false, no matter how much you boldface it. But if you want boldface in your response: to increase power, you can increase either the voltage or the current. Since you have to provide coolant anyway for the battery pack to dissipate the vastly higher cooling needs, then you're already cooling the cable, and thus can provide vastly higher currents on the same amount of copper.

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.

Funny that you're acting as if you're the one running numbers when I'm the only person in this entire thread who's actually conducted calculations. And now that you need numbers run, rather than conducting them yourself, you insist that I do them - with a mocking tone at that. Well, fine, because it's a simple request. The specific heat of a 50% ethylene glycol mix is 1814 J/kg-°C at around 0°. Let's say that we want to limit the heat rise in the pack to 10°C, so the fluid can handle 18140J/kg, or 18,14J/g. 10kW is nothing in terms of pack heating, but let's go with your figures. A kW is a kJ/s, so 551g/s. The specific gravity at 0°C is around 1.1g/cm³, meaning a flow rate of around 501cm³ per second, or 0,000501m³/s. What's our limiting pressure drop? Let's plug in the numbers for 1cm x 4m tubing. For water that would be a pressure drop of 1609 mbar The pressure drop for a 50-50 glycol water mix at those temperatures is reported to be 45% that of water, or 724mbar. What's the problem? We're talking 345 grams of glycol in the entire 4m length. Want a lower pressure drop? Increase the diameter from 1cm; it's still a small mass. 2cm and the pressure drop decreases to 26mbar.

What problem exactly where you expecting? It's pretty silly that you expected a problem to begin with, given that the car itself already has to move these volumes of fluid through its radiator. The concept that the car could handle it but a charger couldn't makes no sense at all.

The fact that you'd bring up air cooled copper is precisely the problem. Air cooling is vastly slower than liquid cooling. Today you have an air-cooled cable, and a vehicle getting rid of its heat with a radiator, to the air. Rather than a liquid-cooled cable and getting rid of its heat via transfer to liquid. If you rely on air cooling your cable, you have to use a heavy cable per unit power. If you rely on a radiator to get rid of charging heat, you're limited to your coolant being ambient temperature and having a high ratio of radiator mass to heat loss rate.

 

[AudubonB]

===>This thread may need some coolant lines running through it to keep tempers from flaring.<===

Keep.
It.
Cool!

 

[arnis]

Well, those thousands of amps must not only transfer in 2m cable. They start at inverters and end at cells. Therefore lots of things to cool down. There is a plug at the battery. There is also crimping connection inside the stall. And all those coolant lines must run with the whole length. Or we have to increase conductors massively.
What about loss of efficiency due to 400V system? Do we strike those out? We must cool the cable/mating surfaces due to losses. And if we quadruple the current, there will be more than 4x losses compared gen2 SC.

Also, what about cable weight and diameter. Tesla tried cooled cable. It was thinner. But it only handled 300-350A (normal for SC). If we have todays SC cable copper and add a pair of hoses. What then?


And lastly, what about price. It will cost much more. Will have more maintenance. Therefore, with the same amount of money, we will have less SuperChargers/stalls. Therefore waiting lines, like at fuel stations.

 

[KarenRei]

Well, those thousands of amps must not only transfer in 2m cable. They start at inverters and end at cells. Therefore lots of things to cool down. There is a plug at the battery. There is also crimping connection inside the stall. And all those coolant lines must run with the whole length.

Tesla already cools power electronics. Have you never looked at the inverter design inside the vehicle? It's already liquid cooled. One presumes that the AC/DC converters inside the superchargers are as well, although I've never seen a teardown of one. They pretty much have to be, you can't air-cool away that much heat in that dense of a form factor.

Heck, even amateurs cool their CPUs with liquid cooling all of the time. You can pick up liquid cooled heat sinks on amazon. For cheap. This isn't rocket science.

What about loss of efficiency due to 400V system? Do we strike those out? We must cool the cable/mating surfaces due to losses. And if we quadruple the current, there will be more than 4x losses compared gen2 SC.

Except that it doesn't work that way. You don't just magic increased power into existence; it takes a more massive AC/DC converter (or more realistically, more in parallel), which means more copper that the current flows through. As a consequence, losses increase linear with power, not quadratically, and heating per unit volume/mass is constant, since volume/mass rises linear with power as well.

Also, what about cable weight and diameter. Tesla tried cooled cable. It was thinner. But it only handled 300-350A (normal for SC).

Yes, Tesla made today's cable thinner and lighter by cooling. Where do you find a problem with this? The effect becomes more pronounced the higher the powers you want to go into.

I'm not sure you're getting the difference in cooling rates between liquid cooling and air cooling. Do you weld? I do. One of the big annoyances is accidental burns because you finish welding something move onto something else, and then later touch the original piece of metal - but it had gotten so hot that even after sitting in the air for a minute or two it can still instantly burn you. Now instead of letting it air cool, what happens when you plunge it into a bucket of water for say five seconds? It comes out cool to the touch.

The difference in cooling rates between air cooling and liquid cooling is night and day. They're not even comparable.

And lastly, what about price.

Far less copper per unit power = far less cost per unit power. Now, of course, any higher power charger will cost more overall - charger costs lie mainly in the power electronics (not the cable, not cooling, etc). But a higher power charger also services vehicles faster and moves on to the next one, increasing throughput. Tesla has already clearly decided with V3 that high power chargers are the way to go.

 

[arnis]

Superchargers are made out of the same module design Tesla uses onboard the vehicles:

If you want to charge 4x faster we need 2 of those per STALL.
That is expense most of car buyers will not want to pay for.
The bigger the capacity on board the vehicle, the less we need those rapid chargers.

Yes, if DeltaT is huge (welded metal), plunge in water will result in rapid cooldown. But that bucket will get warm also.
We need to keep the copper chilly.
PS: refrigeration cycle requires more electricity than just cooling with air passing through the fins.
I believe Tesla is pumping around 2kW of energy into AC compressor to extract 5-6kW of heat.
Another expense that comes up with very rapid charging.

I'm imagining now about raising voltage by having negative terminal running at -400V DC and positive at +400V DC.
Theoretically it should be possible to do some clicking with relays and switch these chargers into different configuration.
This means same power, different voltage options.

The effect becomes more pronounced the higher the powers you want to go into. - Not power, current.

You forgot heat transfer problems with water cooling. Ideally, you run copper inside the coolant. But that won't work easily.

 

[KarenRei]

If you want to charge 4x faster we need 2 of those per STALL.

1. Fred Lambert‏Verified account @FredericLambert 24 Dec 2016
   Supercharger V3? Now I'm curious. What power out are we talking about? 350 kW?
   
   5 replies18 retweets84 likes
   

Elon Musk‏Verified account @elonmusk
Replying to @FredericLambert
A mere 350 kW ... what are you referring to, a children's toy?

--------------------------------------------------------------

Tesla is doing this, whether you like it or not. Whether they plan to use more units per stall or simply higher power electronics.

That is expense most of car buyers will not want to pay for.

To repeat (I shouldn't have to): the higher the power of the charger, the higher its throughput. Meaning fewer chargers needed. What about this concept do you find difficult?

The bigger the capacity on board the vehicle, the less we need those rapid chargers.

Until we've met the average consumer's demands, we will always need higher power chargers. You may think they're being unreasonable with their demands. They don't.

Yes, if DeltaT is huge (welded metal), plunge in water will result in rapid cooldown. But that bucket will get warm also.

*Sigh*. The rate of heat loss to liquid versus air will always be orders of magnitude faster. "Orders of magnitude faster than X" applies whether X is a large or small number.

We need to keep the copper chilly.

Which means cold coolant (which you keep in a reservoir, prechilled and ready for the vehicle) - as in Tesla's patent.

PS: refrigeration cycle requires more electricity than just cooling with air passing through the fins.

And? It's still a very small fraction of the total energy used by the charger. If you're talking, say, 85% net efficiency, so 15% of the energy used in charging delivered as heat, and then a COP of, say, 3, you're looking at 5% of the charge energy used in cooling. Whoop-de-doo.

I'm imagining now about raising voltage by having negative terminal running at -400V DC and positive at +400V DC.
Theoretically it should be possible to do some clicking with relays and switch these chargers into different configuration.
This means same power, different voltage options.

Yes, because cooling a cable is hard, but realtime rewiring of a battery pack is a trivial no-downsides task.

Can you cite an example of anyone pursuing this approach ever before? Because Tesla has been extensively working towards cooling, of both charge cables and of the onboard coolant.

You forgot heat transfer problems with water cooling. Ideally, you run copper inside the coolant. But that won't work easily.

Again, we're talking glycol mixes, not water. But ignoring that, no, you don't have to at all. The maximum temperature you can run a wire has nothing to do with the copper itself, it's the temperature that the insulation rises to (there are different limits for wires insulated with different materials). And the insulation touches the coolant.

 

[JRoque]

Hi all. Nice thread! I may have missed a word or two from the posts but figure I'd throw a stone and go hide.

- An 800V charger can charge a 400V battery. In fact, voltage is pretty much irrelevant if one regulates current. There are advantages with higher voltage, up to a point.

- Why do we need a 3 phase plug when it would be DC charging the battery? I'm sure I missed this point somewhere. Three phase AC on-board charger maybe?

- 3ph doesn't need a neutral line, and neither does split phase 240V.

- Ground could be carried on the connector sleeve. This will provide ground connection (and static discharge) before other pins make contact.

- Higher voltage = lower resistance = less heat but if coolant flow is needed, there shouldn't be too much of an issue adding a fluid connector that perfectly seals both ends. Having full system coolant flow would allow for much higher current charging since not only the cable heats up but the batteries do, too. The same coolant pump in the car and drive the fluid through the charger cable and unit. The external charger would be passive, with a radiator, full of coolant so the car doesn't lose any in the process. Coolant flow-through does complicate matters somewhat, however.

And now I run. Be be gentle with your reply. I'm new here and still fragile... and I also wear glasses

JR

 

[spottyq]

I have seen mentioned a few times that an 800V charger can charge a 400V battery.

While this is can certainly be true, a lot of current Chademo chargers cannot charge Zero's electric bicycles who have a nominal battery voltage of around 100V. (The charger needs to be able to go down to 90V.) It went as far as Zero not shipping any bike with a Chademo plug, to avoid user confusion.

However, they found that many charger were not capable of going lower than 200V, even though the Chademo spec requires them to go all the way down to 50V.

My point is that it may not be as straightforward as it seems to build a 800V charger that can also support 400V batteries. I'm sure they are technical solutions, but for whatever reasons, some manufacturers decided it was not worth the extra time and money spend on it.

Source : Charged EVs | Zero Motorcycles was forced to abandon DC fast charging option in 2013, Interoperability testing is needed

 

[KarenRei]

An 800V charger can charge a 400V battery, but at half its maximum power. The total power rating of a charger is its maximum voltage times its maximum current. If you double the voltage of a charger to double its power rating, but then charge a vehicle at the original voltage, you've gained nothing.

 

[GSP]

Karen and McRat - and any other cognoscenti:

1. Regarding 3-phase in the USA - can you tell me the reason 3ø charging is not possible as things exist today? Is there not a non-Frankenconnector that could permit 1ø 240V AC charging at a domicile, and 3ø 480V (etc) when at a SpC?
....

The Type 2 EV plug does this nicely, allowing EV drivers to use either 1-phase or three phase AC to charge. I think the Type 2 standard allows up to 64 Amps 3-phase, and at least that much current from single phase. However most utilities in European countries do not allow more than 16-32 Amp to be drawn from a single phase.

Tesla's modification of Type 2 also allows full supercharging DC current levels, without the clumsy addition of two large pins in the CCS Type 2 connector.

GSP

PS. To others asking about 800 V charging, one method to accomplish this is an on-vehicle DC-DC converter to buck the 800 V charger supply down to the battery voltage. Another is to use more expensive, lower volume, 600 V power electronics (1200 V rating) instead of the usual 300 V components (600 V rating).

 

[GSP]

An 800V charger can charge a 400V battery, but at half its maximum power. The total power rating of a charger is its maximum voltage times its maximum current. If you double the voltage of a charger to double its power rating, but then charge a vehicle at the original voltage, you've gained nothing.

Actually you have gained something: the ability to charge lower voltage EVs.

This is just as important as being able to charge higher voltage EV at a faster rate. I will be extremely surprised if the new 800 V DC chargers are not thourgly tested for compatibility with today's 300-400 V EVs before start of charger production.

GSP

PS. About using only three pins for three phase. I will be shocked if the protective earth pin is not required (I know, very bad pun there). That is four pins minimum. However if you actually want to drive to areas where the wye is used instead of delta, and charge there, you also need a neutral pin. It would be ludicrous to design a standardized EV plug that would not work in countries that require three phase power in the wye configuration. Of course, SAE did exactly that with their short-sighted Type 1 standard.

 

[KarenRei]

Actually you have gained something: the ability to charge lower voltage EVs.

That's not a "gain", that's something you already have. You're spending a lot more to gain high voltage capability, but not improving your charge rate on lower voltage EVs (essentially all EVs) at all.

 

[GSP]

Karen,

If you stick with what you have, and not increase either voltage or current, no cars can charge at a faster rate.

If you increase either voltage or current, existing cars will not be able to use more voltage or current to charge faster, only the new cars designed to do so. However, the charger will be much more useful if it can charge both existing and new cars.

I think charger manufacturers will make faster chargers, even if they cost more and only will be faster with cars designed to charge at higher rates. Doesn't that make sense?

GSP

 

[KarenRei]

Karen,

If you stick with what you have, and not increase either voltage or current, no cars can charge at a faster rate.

If you increase charger current and providing cooling in the cable:

1) Cable overheat problems go away, increasing charger throughput and reducing customer delay, particularly during busy times.
2) Multiple cars sharing a charger all charge at higher rates, since the charger itself supports higher currents - regardless of whether the cars do anything with coolant.
3) Where cars can't ramp up beyond today's charge rates on their own due to internal heating limitations, the replacement of the 3-loop heat exchanger in the vehicle (either as a retrofit or for new models) with a 4-loop one, with a coolant loop from the charge port feeding the fourth loop, is all that is needed to allow vastly higher rates of heat removal from the battery pack. This means both higher total heat removal throughput throughput (aka, you're not limited by the onboard radiators), and an increased delta-T (coolant temperature vs. cell core temperature) in the pack due to the ability to maintain the coolant at the lowest acceptable temperature and cycle it faster, meaning higher heat flow. This is particularly important for achieving higher charge rates during cell balancing, where more full cells turn incoming energy to heat while other cells still have more to charge.

Onboard systems will never be able to compare with the capabilities of offboard systems.

If you increase voltage:

1) Essentially none of today's cars benefit. At all.
2) You gain no charger-sharing benefit either.
3) You have to make future EVs higher voltage to get any benefit at all, which isn't just a simple retrofit / design alteration, but a fundamental change. It also means (beyond higher insulation requirements) either increasing the number of cells (higher part count, lower energy density, higher cost) or reducing how many cells are in parallel (greater degradation).

There's a reason why Tesla has been working towards provision of coolant.

 

[arnis]

CCS goes with higher current and voltage. It is already written down.
No change to plug dimensions. Water cooled it is (in case of maximum current solution).
Two different plugs is not acceptable (one with cooling, another without).
Plug that supports cooling but does not have it one one version, will also be frankenstyle.

Coolant system is likely solution for semis. Anything bigger than small cargo van.
Costing much more. Having charging stalls differently positioned and locations away from city centers.
And definitely not normal CCS sized coupling. Likely two different couplings. One for cooling,
one for charging.

Imagining hundreds and hundreds of SC location for example in California, all having MegaWatts of power
available, is wet dream. It is expensive. Very expensive. Even battery buffering versions.

Our inventions have changed our lifestyles. Electrification will be one of those. We will not charge
electric vehicles as fast as gasoline vehicles just because we did it before.
As I've noticed, during last decade, people visit gas stations in totally different way compared to 20 years ago.
It's not a place to refuel your car. It's a place you visit occasionally. Doing more than only that.
Though there are stations that offer nothing else. We can see the trend.
Electrification is a business. If one can make money on other things while you charge, it will happen.
Without business opportunity, there would be no chargers.

 

[KarenRei]

CCS goes with higher current and voltage. It is already written down.

Written down where? Here's some of the first 350kW CCS stations:

Efacec's First 350 kW CCS Combo DC Fast Chargers Already Up & Running

The HV175 is a high power charging solution, able to supply up to 920 V nominal and 1000 V maximum at a maximum current of 350 A

Read: only 350A. For an EV charging at an average of, say, 350V, that's a mere 122kW.

No change to plug dimensions. Water cooled it is (in case of maximum current solution).

They are not water cooled. Their solution to delivering more current is, and I quote, "Connecting more HV175 units to a mechanical connection allows higher currents as can be used by some heavy vehicles." So enjoy connecting half a dozen different cables to your car, basically. That will never be the standard.

Two different plugs is not acceptable (one with cooling, another without).

Nor is such a thing necessary.

But - for the record - most of our chargers over here are are multiplug. Just thought you might like to know that.

Coolant system is likely solution for semis. Anything bigger than small cargo van.Costing much more. Having charging stalls differently positioned and locations away from city centers.And definitely not normal CCS sized coupling. Likely two different couplings. One for cooling,

Yes, because if there's anything manufacturers like, particularly Tesla, it's duplicated infrastructure (with increased idle time), needlessly oversized hardware on vehicles and making owners do more work. Sure.

Imagining hundreds and hundreds of SC location for example in California, all having MegaWatts of power
available, is wet dream

First off, gross.

Secondly, EVs will not overthrow gasoline vehicles until charge powers significantly rise. It's great for us early adopters, but step out of the bubble some time and talk to normal people about charge times,.

Third, as I've mentioned about five different times, and which you've ignored each time and continued pretending like it was never said (really annoying, by the way): the faster a vehicle charges, the faster the charger's throughput. If a charger averages spending an hour per vehicle, it can charge no more than 24 vehicles per day. If a charger averages 10 minutes per vehicle, it can charge up to 144 vehicles per day. You need 1/6th as many chargers.

Oh, but it's better than that. Because at present you share one charger between up to two vehicles. A large portion of a charger's time is spent idle. The more stalls you have served by single, higher power chargers, the larger percentage of their time that they will statistically spend doing useful work rather than sitting idle - making the economics even better. The max power being high enough to supply even a semi, any vehicle which can consume it that quickly, while for lower power vehicles the power is distributed across a greater number of vehicles. Some fraction will also always be later in their charge and thus at reduced charging power. The charger spends most of its time closer to 100% utilization,

I'll reiterate: this is the direction Tesla is moving. Whether you like it or not. Tesla is going high powers. Enough to make 350kW look like childrens' toys. They're patenting providing the vehicles cooling. Etc, etc, etc. I'm sorry you don't like the future, but it is what it is.

 

[arnis]

Tesla has lot's of patents. Patent is not the same as "real deal".

https://www.phoenixcontact.com/asse...promotion/52007586_EN_HQ_E-Mobility_LoRes.pdf
Well here is very short brochure explaining what PhoenixContact offers.

It appears you forget one important thing about rapid charging:
it shall and never be used as the main way to charge electric vehicles on daily basis.
Tesla has reminded that all the time. Some people ignored these warning and now we have what we have.
The fact that "some of us do not have a way to charge EV's at home" is not an excuse.
And due to that fundamental requirement* of EV lifestyle, rapid charging within 5 minutes makes no sense.
Yes, 15 minutes would be nicer than 40.

*Which is "EV's shall be mostly charged at night (or whenever there is excessive energy available).
It doesn't matter in case of thousands of EV's. It matters when we get to tens of millions. The vast majority must act that way.

More than 90% of charging happens on Level2 today. That number will rise even more. According to average commute data,
it should be, maybe in a decade or two, higher than 98%. It is not today as a) EV ranges are not excellent yet; b) Level2 EVSE's are still rare.
And then, charging within 5 minutes instead of 15 minutes, makes very little sense.