How does a Supercharger work?

[deckofficer]

I've done a little more research and it appears that on a few of the high use SC stations they are using their own 400 kWhr battery for peak shaving. Since Tesla is picking up the bill for the grid supplied electricity for the stations, they have been getting hit with the high usage of cars charging during peak rates. Simple math, if the SC station can keep power draw at a fixed 24 hour level, they won't get pinched during peak by using these batteries. Elon and his team are visionaries indeed.

 

[tom66]

The Tesla charger modules achieve a >0.99PF, so they are very good loads on the grid and 500kVA transformer can provide nearly 500kW AC load, 96% peak efficiency means ~480kW DC.

Pictures of the charger modules:
https://technology.ihs.com/Teardowns/binary/devices/2611/photos/95866/binary?size=
https://technology.ihs.com/Teardowns/binary/devices/2611/photos/96038/binary?size=
https://technology.ihs.com/Teardowns/binary/devices/2611/photos/97217/binary?size=
https://technology.ihs.com/Teardowns/binary/devices/2611/photos/97213/binary?size=
https://technology.ihs.com/Teardowns/binary/devices/2611/photos/96466/binary?size=

 

[Cosmacelf]

Man that Charger module looks complicated! I had no idea rectifying AC could be that involved.

 

[kennybobby]

Is an active WiFi necessary for supercharging?

Is it necessary that the 3G card be installed/active to plug in for supercharging?

Are supercharging sessions logged on the car and/or sent in to the mothership, or does the charging station have it's own link back home to send log data?

 

[ggr]

Is it necessary that the 3G card be installed/active to plug in for supercharging?

No.

Are supercharging sessions logged on the car and/or sent in to the mothership, or does the charging station have it's own link back home to send log data?

The superchargers certainly have their own link back. However I don't think the car automatically sends any "operations normal" data home; only faults or emergencies. Everything else has to be pulled by them, and they don't do it unless you have a reason to ask them to do it.

 

[andrewket]

Each 120kWh Supercharger cab should have an A and a B station. The B station is a slave to A. If there is a greater need in the A station, it gets the power from B. If two dead cars show up and plug into the same cabinet's A and B station, the A car will get full power until it starts to throttle. Once it throttles, the "leftover" power goes to the B side.

Not quite. A and B are equal. The priority will go to the vehicle that plugged in first. Also the cabinets have been rated at 135kW (not kWh) for about 20? months now.

 

[MacroP]

Man that Charger module looks complicated! I had no idea rectifying AC could be that involved.

It is complex because it a switching supply (with DC-DC step up) so not a simple rectifier. It looks to be a charger from the Model S car not a supercharger module.

 

[nwdiver]

It looks to be a charger from the Model S car not a supercharger module.

It's my understanding that a Supercharger is 12 of these stacked together...

 

[stevezzzz]

When the guy opens up a Supercharger cabinet and shows me how everything works, and he's been responsible for commissioning many of them, I tend to believe him

Fair enough. But I'm still waiting to hear from anyone who's ever had their existing Supercharger session throttled down by a car plugging into the same cabinet later. If it has happened, it's not been reported here on TMC. Not that I've seen, anyway.

 

[andrewket]

Fair enough. But I'm still waiting to hear from anyone who's ever had their existing Supercharger session throttled down by a car plugging into the same cabinet later. If it has happened, it's not been reported here on TMC. Not that I've seen, anyway.

I agree with Stevezzzz on this one. A field tech may not understand the specific details. In December I met a somewhat new field tech at an SC while he was performing maintenance. He had never driven a model S, had not seen a P85D, and was unaware that there have been 3 different versions of the cabinets. In the area he was working there were only the newest version. My point is I wouldn't expect a field tech, even one that has commissioned a location, to understand the specific details of how the software works.

 

[jerry33]

The superchargers certainly have their own link back. However I don't think the car automatically sends any "operations normal" data home; only faults or emergencies. Everything else has to be pulled by them, and they don't do it unless you have a reason to ask them to do it.

I'm not sure that's correct because I was told that they could pull logs on specific dates, and I doubt the car has enough memory to hold all the logs records it's produced during it's lifetime. I suspect the car sends logs (or has logs pulled) on some more-or-less regular schedule. They pull the logs that haven't been uploaded yet.

 

[GreenT]

Fair enough. But I'm still waiting to hear from anyone who's ever had their existing Supercharger session throttled down by a car plugging into the same cabinet later. If it has happened, it's not been reported here on TMC. Not that I've seen, anyway.

I have definitely had it happen. Used a LOT of SC's and if I am on an "A" bank, and someone else pulls in on A later they *and I* have gotten reduced amperage. Ditto for B, etc.
Is that what you mean?
This is battery SOC agnostic as far as I can tell.

 

[andrewket]

I have definitely had it happen. Used a LOT of SC's and if I am on an "A" bank, and someone else pulls in on A later they *and I* have gotten reduced amperage. Ditto for B, etc.
Is that what you mean?
This is battery SOC agnostic as far as I can tell.

I'm not sure I understand your note. The A's are not linked together. Each cabinet has exactly one A and one B. These are shared. If you're on 1A, and someone else plugs into 2A, it would not have any effect.

If you're on 1A, and someone plugs into 1B, 1A will continue to receive the maximum power for that car's SOC and conditions (pack temperature, etc). The car plugged into B will receiver whatever is remaining. Conversely, if you were plugged into 1B and then later someone plugged into 1A, the car plugged into 1A would receive whatever is remaining.

As the SOC of the first car increases, the power available to the second car increases. The worst case scenario is that two cars with very low SOC plug in within seconds of each other.

 

[Saghost]

I have definitely had it happen. Used a LOT of SC's and if I am on an "A" bank, and someone else pulls in on A later they *and I* have gotten reduced amperage. Ditto for B, etc.
Is that what you mean?
This is battery SOC agnostic as far as I can tell.

Do you have specific notes about times, locations, and SoC at the time?

From everything I've read, this is highly unexpected behavior - there's no linkage between your "A" connection and someone else on a different "A" short of the utility transformer (or the battery pack if the site has one.)

Unless the whole site has some sort of power limit that you're bumping into what someone does on another A should have no impact, so I'll be interested in any data you have...
Walter

 

[GreenT]

I'm not sure I understand your note. The A's are not linked together. Each cabinet has exactly one A and one B. These are shared. If you're on 1A, and someone else plugs into 2A, it would not have any effect.

If you're on 1A, and someone plugs into 1B, 1A will continue to receive the maximum power for that car's SOC and conditions (pack temperature, etc). The car plugged into B will receiver whatever is remaining. Conversely, if you were plugged into 1B and then later someone plugged into 1A, the car plugged into 1A would receive whatever is remaining.

As the SOC of the first car increases, the power available to the second car increases. The worst case scenario is that two cars with very low SOC plug in within seconds of each other.

Sorry, may have confused letters with numbers.
Saghost ... SoC would have been at about 50 range miles minimum left for me usually. Can't speak for the other owners. Times nope. Happened at various SC's up and down east coast and Ohio. Other owners agreed but we didn't take scientific notes. Just knew, when possible (usually) to not occupy adjacent lettered / numbered bays.

 

[ggr]

Fair enough. But I'm still waiting to hear from anyone who's ever had their existing Supercharger session throttled down by a car plugging into the same cabinet later. If it has happened, it's not been reported here on TMC. Not that I've seen, anyway.

The people who man the stores and put in superchargers almost never actually own Teslas or live with them on a day-to-day basis. And this behavior would be totally controlled by software; I think you could look at the hardware all day and not be able to tell how it decided to split the current. I think it works like Andrewket said above. Except for slowdown due to throttling, I've never seen my charge current decrease suddenly.

 

[kennybobby]

Are the bank of 12 chargers in the cabinet water-cooled as in the on-board chargers?

If no 3G is needed in order to supercharge, then the vehicle must have a memory register with a bit set or cleared to enable/disable supercharging? May or not be related to id by VIN?

So is the vehicle commanding the supercharger over the CAN bus to request voltage and current settings?

 

[Chris TX]

Are the bank of 12 chargers in the cabinet water-cooled as in the on-board chargers?

They are water cooled, just like the ones in the car. There is a radiator that is on the back near a very large fan.

 

[Saghost]

Are the bank of 12 chargers in the cabinet water-cooled as in the on-board chargers?

If no 3G is needed in order to supercharge, then the vehicle must have a memory register with a bit set or cleared to enable/disable supercharging? May or not be related to id by VIN?

So is the vehicle commanding the supercharger over the CAN bus to request voltage and current settings?

The car definitely has a setting that says whether it is allowed to Supercharge - we've seen screenshots of it in the diagnostic laptops. However, the first thing the car passes the Supercharger over the CANBus is the VIN - presumably so the Supercharger can authenticate the car over its own network connection.

I don't know the answers to your other questions. I'm assuming the Supercharger modules are liquid cooled like the car chargers, but I've never seen it stated one way or another.

Despite the efforts to monitor the SC communications bus, I don't think we know how the charger cycle is controlled. Clearly parameters from the car beyond the battery voltage are involved (slower charging when the car is hot,) but it isn't obvious if the car passes the charger those parameters over the network and the charger decides how to respond, or if the car decides the right response and tells the charger. I'm not sure it really matters which way they chose to do it, as long as it works reliably. (I'll speculate the car directs the sequence, since the Supercharger otherwise has to keep at least 3 profiles on hand - 60 kW, 85 kW limited, and 85 kW unlimited - and know by VIN or data passed from the car which one to use.)
Walter

 

[HankLloydRight]

They are water cooled, just like the ones in the car. There is a radiator that is on the back near a very large fan.

Actually, minor note.. they are oil cooled.

When I first got my car, I was at the Greenwich SC when a tech was there literally doing an 'oil change'.