Transition to higher voltage batteries / DC fast charging

[n2mb_racing]

I've been wondering, how will Tesla (and other manufacturers) transition to higher voltage DC fast charging? I can only see two options:

1. Change the battery, motor and other car systems to 800V or higher.
Cons: You lose the ability to DC fast charge at all existing supercharger and Chademo installations that are only designed for up to 400-500V charging. All the chargers must be upgraded.
Pros: Car internal wiring is easier.

2. Keep the battery and motor systems at 400V. Use a DPDT contactor to break the pack in half and put it in series for DC fast charging at 800V, parallel for normal operation (400V).
Pros: You get backwards compatibility with all existing DC faster chargers and the ability to charge even faster at new high voltage stations.
Cons: added complexity on the car.

 

[widodh]

I think they will go for option 2. A bit more complex, but it does allow backwards compatibility.

 

[spottyq]

Even if you choose to go with option 2, you still have to at least make the HVAC able to operate at both 400 and 800V. (You need cooling/heating when charging fast) and probably also the DCDC converter (though you could probably get away with a beefed-up 12V battery.)

So basically, you have to convert everything HV except the motors…

 

[VT_EE]

I'm curious why Tesla didn't design to 800V to begin with. What are the drawbacks of the higher voltage that caused them to go with ~400V systems? Is it harder to design motors and inverters at the higher voltage?

 

[n2mb_racing]

I'm curious why Tesla didn't design to 800V to begin with. What are the drawbacks of the higher voltage that caused them to go with ~400V systems? Is it harder to design motors and inverters at the higher voltage?

Increased cost for insulation, connectors, accessories, etc. I'm not sure if the parts even existed at the time. I think 400V was somewhat agreed upon by EV parts manufacturers.

 

[RDoc]

Even if you choose to go with option 2, you still have to at least make the HVAC able to operate at both 400 and 800V. (You need cooling/heating when charging fast) and probably also the DCDC converter (though you could probably get away with a beefed-up 12V battery.)

So basically, you have to convert everything HV except the motors…

It seems to me to be a lot easier to just tap into the main battery at whatever voltage is needed, 400V for instance. The pack is made up of lots of low voltage cells arranged into groups that can be separately drawn from to get pretty much any voltage up to maximum needed.

 

[Jeff N]

Even if you choose to go with option 2, you still have to at least make the HVAC able to operate at both 400 and 800V. (You need cooling/heating when charging fast) and probably also the DCDC converter (though you could probably get away with a beefed-up 12V battery.)

So basically, you have to convert everything HV except the motors…

Or just add in an extra DC to DC converter that goes from 800V to 400V and use that to feed the HVAC and 12V DC to DC converter only during 800V charging. It probably only needs to be around 7 kW (the 12V one is probably around 2 kW). Since it would only be used for short periods of time during 800V charging perhaps it can be built cheaper if not targeting the highest efficiency? I don’t know.

Maybe the 12V and 400V DC converters can be co-packaged and share some components such as the packaging and cooling in order to reduce costs.

 

[spottyq]

It seems to me to be a lot easier to just tap into the main battery at whatever voltage is needed, 400V for instance. The pack is made up of lots of low voltage cells arranged into groups that can be separately drawn from to get pretty much any voltage up to maximum needed.

If you do that, cells will get out of balance. I guess, if you rarely do 800V charging sessions and maybe beef up the BMS's resistors, it could work out.

Do you know any product that uses a battery like you propose ? (Just curious; theoretically it is a great way to have the different voltage but I wonder if someone has been able to work around the technical problems.)

Or just add in an extra DC to DC converter that goes from 800V to 400V and use that to feed the HVAC and 12V DC to DC converter only during 800V charging. It probably only needs to be around 7 kW (the 12V one is probably around 2 kW). Since it would only be used for short periods of time during 800V charging perhaps it can be built cheaper if not targeting the highest efficiency? I don’t know.

Maybe the 12V and 400V DC converters can be co-packaged and share some components such as the packaging and cooling in order to reduce costs.

I guess that could work. It seems to me that it would add a lot of cost/complexity though.

 

[Jeff N]

If you do that, cells will get out of balance. I guess, if you rarely do 800V charging sessions and maybe beef up the BMS's resistors, it could work out.

Do you know any product that uses a battery like you propose ? (Just curious; theoretically it is a great way to have the different voltage but I wonder if someone has been able to work around the technical problems.)



I guess that could work. It seems to me that it would add a lot of cost/complexity though.

Sure, but at first it would only be used on expensive models pushing the boundaries on charging rates and battery capacity or design. It would be transitional until it becomes cheaper to redesign all of the car’s high voltage components to run natively on 800V and make those in substantial enough numbers that their cost becomes competitive with 400V parts.

 

[arg]

If you do that, cells will get out of balance. I guess, if you rarely do 800V charging sessions and maybe beef up the BMS's resistors, it could work out.

Those BMS resistors would end up dropping precisely the amount of energy you are drawing from the half-way tap (if you are estimating 7kW load, then the resistors would be dissipating 7kW for the duration of the charge, or a lower amount over a longer time). Theoretically possible but not sensible.

Probably in fact relatively straightforward to make the auxiliaries dual voltage (400/800V).

The contactors for rearranging the pack, and the issues of making the two halves exactly the same voltage (or charging them one at a time and having to deal with the balance issues resulting) makes the whole thing fairly unattractive.

If it were me, I'd want to see widespread installation of dual voltage charging equipment before launching an 800V vehicle (which could then be single voltage).

Tesla don't have any need for this for several years at least (for cars at any rate - the Semi could be another story, but it is unlikely to be sharing Superchargers with cars).

That does potentially put Tesla at a disadvantage, as they will have rolled out substantial Supercharger installations at 400V-only, while competitors haven't yet invested in public charging in any useful fashion and could go straight to 800V.

 

[Jeff N]

That does potentially put Tesla at a disadvantage, as they will have rolled out substantial Supercharger installations at 400V-only, while competitors haven't yet invested in public charging in any useful fashion and could go straight to 800V.

Only for today’s installed base of Superchargers. I suspect Tesla will quickly begin installing all their new Superchargers with equipment that is capable of 800V even if Tesla passenger vehicles aren’t utilizing it immediately.

For example, all of ChargePoint’s new generation DC charging hardware will support 800V charging even when installed in configurations that are only capable of ~60 kW.

 

[arg]

Only for today’s installed base of Superchargers. I suspect Tesla will quickly begin installing all their new Superchargers with equipment that is capable of 800V even if Tesla passenger vehicles aren’t utilizing it immediately.

There's a bit of a chicken-and-egg situation: Tesla's superchargers are cheap to build because they are based on the high-volume parts used in the cars, so they would only be in a position to do that easily if the cars were already on the horizon.

Certainly it would be wise for them to do so when they can (perhaps upgrade the wiring at least so as to make later upgrades of built supercharger sites easier). Not entirely clear that they've got the engineering or management resources to spare for something that doesn't have impact on today's urgent priorities.

 

[Jeff N]

There's a bit of a chicken-and-egg situation: Tesla's superchargers are cheap to build because they are based on the high-volume parts used in the cars, so they would only be in a position to do that easily if the cars were already on the horizon.

Certainly it would be wise for them to do so when they can (perhaps upgrade the wiring at least so as to make later upgrades of built supercharger sites easier). Not entirely clear that they've got the engineering or management resources to spare for something that doesn't have impact on today's urgent priorities.

It seems pretty likely that the new Tesla semi truck is going to utilize 800V charging so presumably they are already doing this work. Elon has suggested that Tesla will have some new Superchargers capable of substantially beyond 350 kW and this is presumably for the truck with a giant battery pack. Charging rates much above 350 kW are probably more practical at 800V than at 400V.