Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

How pairing at Supercharging works

This site may earn commission on affiliate links.
Consider two stalls of supercharger in urban environment, almost full for several hours. In that case urban superchargers can offer 72kW to 50kW up to 85% battery, so almost running full capacity. On the other hand regular superchargers are like 36kW increments, so if two cars are sharing each takes 72kW (same as urban), but if one car started earlier it takes 108kW giving the other car only 36kW. The problem is that the first car is tapering off - during this time, for example at 100kW, the paired total output is only 136kW, rather than 144kW. Hope this makes sense.

Maybe I missed it, but where did the idea that Superchargers switch in 36 kW incerments come from?

I think you get 36 to start, but I'm pretty sure I've seen it switching in 12s thereafter one the rare occasions I was second car to someone who was still drawing a lot.

It makes a big difference - with individual chargers switching, I'm pretty sure the sharing model will remain as fast or faster in pretty much any case. With 36 kW blocks, I' m not at all sure of that because of the cases where the lead car is using just over a block, as suggested above.
 
Methinks that Tesla has put more thought into how folks will use urban charging differently than long-distance trip charging in the long run. The the current behaviors of how folks today use either type of supercharger may not match with the behaviors in the future.

We know that paired superchargers allocate capacity in quarters, so if there are two cars a 145kw charger will allocate 108kw-36kw, 72kw-72kw, or 36kw-108kw, depending on the situation. Really for the vast majority of cases, what we're simply judging is whether when the 2nd car arrives, the allocation behavior should remain at 108kw-36kw, or whether 72kw-72kw is better. But what is "better"?

Those are allocated capacity, not actual usable delivery, so anytime the 1st car is in a taper, 108kw->72kw, 72kw->36kw, 36kw->0, it is using less than allocated, and it is POSSIBLE that a 2nd paired car COULD use more of that bank of 36kw than the 1st paired car. We're all trying to judge whether it is more fair, or more efficient (in overall utiliazation, avg charge speed) whether to allocate 36kw to the car that arrived 2nd rather than 1st.

Consider this: in the regular (paired) superchargers, Tesla could probably trivially modify them to act more like urban chargers. I.e. once the 2nd car plugs in, use a 72kw-72kw split instead of 108kw-36kw - It's a very simple algorithmic change.Then it will act exactly like a urban supercharger, until one car unplugs. This alone would guarantee that paired chargers would always be as or more utilzied or efficient than urban chargers. - we only need to debate what fairness means in terms of wait time for each driver.

But Tesla didn't do that, and probably won't. Here's why. Regular superchargers were designed WHEN the problem was long-distance trips (how folks use them in urban areas today may be different). If the superchargers were designed with an ultimate spacing of 120 miles apart, then the first 40kwh dispensed is a matter of NEED, not WANT. You NEED that to get to the next charger or else you risk being stranded (or get into the range anxiety <20% SOC that should be for margin). Anything you charge up beyond that is more about WANT, rather than NEED. If Tesla assumes every car is on a trip, they are really trying to minimize time to get what you NEED, and then it's actually the taper, not the allocation, that's actually slowing you down for what you additionally WANT.

Urban supercharing is s diffrent scenario, and Tesla may be optimizing for some future where urban chargers will be spaced no more than say 20 miles apart, i.e. you're never more than 10 miles from the nearest one. But that also means that to go charge, you may have to drive 10 miles out of your way and 10 miles back, just to charge. That's like 20 minutes (at 60 mi/hr), more like 30-40, just in drivng time on top of 30 minutes of charging time.

Here's where I'm guessing Tesla has put the thought into the future - Instead of going about your days and waiting until you drop to 20%, and making a specific trip out of your way to charge, you'll do it when it's convenient, say when you happen to be passing by while going to the store one day. But that means it's opportunistic, and you can't always plan to be at 20-30% SOC, so more folks will be urban charging opportunistically at 40%, 50%, 70% SOC even. So more visits will not even be able to take advantage of 108kw, let alone 145kw, i.e. you're already in the taper. When you add that more batteries will also be cold-soaked, which can still be the case even if it's 60F outside but you've only driven 10 miles, even less chance of using allocation above 72kw. Predictability, if you go for a 20 minute charge twice a week on average, may also be important, i.e. you can predict about how much time it'll take to go 40%->60%, 40%->80%, or 55%->75%, before you even head to the supercharger, so you can judge whether it's worth your time today vs you might pass by one in another few days.

Now this is oriented for the true urban owner who lives in a condo or hi-rise or rental that has no provision for L2 charging. If you are urban and have L2 at home, Tesla is not designing urban charging for you. And if you have a long, urban commute where you need to supercharge every day on your commute, you're really trip charging on your daily that would fit more of the distance trip scenario, but there's less of you, so you're not the target design point for either scenario.

TL;DR: Urban charging is being designed for the future where it will be shorter, but more frequent opportunistic urban visits at higher SOC, more cold-soaked (not same as freezing) batteries. Less visits would take advantage of charge capacity above 72kw anyway, and now charge visit durations are more predictable too.
 
Re three phase:

I think I understand your argument. However, please actually try to test it. Prepare two almost empty Teslas and start charging at paired stalls. Then disconnect the first car and see what happens. In very crowded superchargers, you will see exactly what @David99 mentioned.

Re, power limit:
I agree it's a bad press, but it's unfortunately a fact.

Ok, just hold on while I put in my order for another Tesla. I would get a CPO but after hearing of Bjørn Nyland's travails, I think I'll order a model Y. I'll likely get it sooner.
 
  • Funny
Reactions: hiroshiy
Maybe I missed it, but where did the idea that Superchargers switch in 36 kW incerments come from?

I think you get 36 to start, but I'm pretty sure I've seen it switching in 12s thereafter one the rare occasions I was second car to someone who was still drawing a lot.

Tesla technician explaining it and measuring the power. See my original post here. I see it step up in 36 increments all the time. Only exception is when your SoC or battery condition isn't allowing a rate that is 36 kW higher to what you are charging at currently. Again, see the graph in my original post. It shoots up to 108 but then drops down instantly because I'm at too high of a SoC to be charging that fast. It never switches down in 36 increments of course.
 
  • Helpful
  • Like
Reactions: hiroshiy and Fiver
Maybe I missed it, but where did the idea that Superchargers switch in 36 kW incerments come from?

I think you get 36 to start, but I'm pretty sure I've seen it switching in 12s thereafter one the rare occasions I was second car to someone who was still drawing a lot.

It makes a big difference - with individual chargers switching, I'm pretty sure the sharing model will remain as fast or faster in pretty much any case. With 36 kW blocks, I' m not at all sure of that because of the cases where the lead car is using just over a block, as suggested above.

Yes, it would make a big difference.

Someone said the chargers in the unit are in units of 12, but that each charger only works on one of the three phases, so they have to tie three together always so they can keep balanced.

I don't know for a fact what Tesla does. What I do know is that before supercomputers/servers were really just PCs racked together, mainframe computers drew power from all three phases in a single power supply. I haven't seen anything definitive on what Tesla has actually done in their power supplies.
 
  • Disagree
Reactions: WdstockWizard
Yes, it would make a big difference.

Someone said the chargers in the unit are in units of 12, but that each charger only works on one of the three phases, so they have to tie three together always so they can keep balanced.

I don't know for a fact what Tesla does. What I do know is that before supercomputers/servers were really just PCs racked together, mainframe computers drew power from all three phases in a single power supply. I haven't seen anything definitive on what Tesla has actually done in their power supplies.

It'd be interesting to see the guts of a modern Supercharger (as opposed to a first generation unit.)

The original Model S charger was a single liquid cooled module able to handle up to 40A at 240V (~10kW). They made the first Superchargers by stringing nine, three on each phase, later upped to 12/4.

When they went to Europe that wouldn't work, though - most chargers are three phase and not happy if you put a big load on one phase. Tesla responded by designing new charger modules that are three in a box, one for each phase.

As I understand it, that's what all modern Teslas get, with the US versions rigged to run all the modules in parallel from the same connection.

Modern Superchargers use the newer charger modules. That would suggest Tesla has the option to wire them to each draw from all three phases or to each draw from a single phase - the EU charger arrangement or the US one.

I wish we had some logged data from second cars to demonstrate what's happening. I've only been in that situation a couple times and I'm pretty sure I had rates between 36 and 72 kW for large parts of the time, but I have no records of the events.
 
It'd be interesting to see the guts of a modern Supercharger (as opposed to a first generation unit.)

The original Model S charger was a single liquid cooled module able to handle up to 40A at 240V (~10kW). They made the first Superchargers by stringing nine, three on each phase, later upped to 12/4.

When they went to Europe that wouldn't work, though - most chargers are three phase and not happy if you put a big load on one phase. Tesla responded by designing new charger modules that are three in a box, one for each phase.

As I understand it, that's what all modern Teslas get, with the US versions rigged to run all the modules in parallel from the same connection.

Modern Superchargers use the newer charger modules. That would suggest Tesla has the option to wire them to each draw from all three phases or to each draw from a single phase - the EU charger arrangement or the US one.

I wish we had some logged data from second cars to demonstrate what's happening. I've only been in that situation a couple times and I'm pretty sure I had rates between 36 and 72 kW for large parts of the time, but I have no records of the events.
Even single phase country like the US and Japan, would all higher voltage feed be three phase? Most Japanese CHAdeMO chargers require 200V or 415V I three phase. In the US supercharger feed is 480V three phase isn't it?

I would think there are no country specific supercharger difference except plugs.
 
Even single phase country like the US and Japan, would all higher voltage feed be three phase? Most Japanese CHAdeMO chargers require 200V or 415V I three phase. In the US supercharger feed is 480V three phase isn't it?

I would think there are no country specific supercharger difference except plugs.

Yes, and I agree.

The question is whether the 12 charger modules in the cabinet are each connected to a single phase or to all three phases based on the sub modules they probably have.

If the cost of the switching hardware is significant, they might mix and match - two blocks of 3, then six to switch between cars as the charge rates change.
 
  • Like
Reactions: hiroshiy
I wish we had some logged data from second cars to demonstrate what's happening. I've only been in that situation a couple times and I'm pretty sure I had rates between 36 and 72 kW for large parts of the time, but I have no records of the events.

The graph I posted in my original post is from my car being the second car. The data is from the CAN bus.

When you say you had rated between 36 and 72, of course that will happen when the supercharger gives you 72 but your car can't take the full 72.

I can only speculate, but I believe the switching in sets of 3 has two reasons. #1 to keep the load of all three phases equal. #2, you need far less switches if you combine three and switch them as one. Possibly also less wiring.
 
Even single phase country like the US and Japan, would all higher voltage feed be three phase? Most Japanese CHAdeMO chargers require 200V or 415V I three phase. In the US supercharger feed is 480V three phase isn't it?

I would think there are no country specific supercharger difference except plugs.

Yes, in the US all distribution above some voltage, 480 I believe, is three phase. 240 volts is always* single phase. There is 208 volts from two wires of a three phase circuit that each give 120 volts to neutral and there is a funny tapped and grounded delta three phase system (high leg delta) that produces proper 240 volts with 120 volts to the tapped neutral. *(nearly always)

Often the final run from a distribution transformer will be 240 VAC to multiple houses. Commercial installations often are direct 480 volts three phase.

When using a single phase of three phase, a device will only see a bit over half the voltage to neutral. 480 volt three phase produces 277 volts relative to the neutral.
 
  • Like
  • Informative
Reactions: hiroshiy and Fiver
The graph I posted in my original post is from my car being the second car. The data is from the CAN bus.

When you say you had rated between 36 and 72, of course that will happen when the supercharger gives you 72 but your car can't take the full 72.

I can only speculate, but I believe the switching in sets of 3 has two reasons. #1 to keep the load of all three phases equal. #2, you need far less switches if you combine three and switch them as one. Possibly also less wiring.

The wiring will be essentially the same. Less switching, but less flexibility and more wasted capacity. My guess is that the switches are designed into the charging units anyway. Each charger unit has two outputs which are all paralleled in a given pair.

After reading one of the posts I can see something odd falling out because of historical reasons. If the chargers were originally designed for single phase inputs and then adopted to three phase input by ganging three units, that would make sense. Rather than design an whole new unit, software changes can be done to gang multiple units without having to recertify the electronics.

I would also bet there is no real difference in the electronics in the urban chargers and the other Superchargers. The urban units are just fixed at a 72/72 kW apportionment. Can anyone confirm or deny this?
 
I would also bet there is no real difference in the electronics in the urban chargers and the other Superchargers. The urban units are just fixed at a 72/72 kW apportionment. Can anyone confirm or deny this?

I can't confirm 100%, but all the reading and photo's I've seen appear to show that the Urban charger backend cabinets are essentially identical to the classic supercharger backends. I'm fairly confident they are the same for ease of manufacturing and just have a different firmware to only output the 72kW per stall regardless.

There might be some savings in wire. I have no idea what wire size you need to run up to 120kW vs 72kW (at ~400v). I'd assume it's essentially the same wire though as that's a ton of power run up to 100' away regardless of max output.
 
There might be some savings in wire. I have no idea what wire size you need to run up to 120kW vs 72kW (at ~400v). I'd assume it's essentially the same wire though as that's a ton of power run up to 100' away regardless of max output.

Nearly twice the current means you need nearly twice the wire cross-sectional area to get the same resistance/heating. The Urban charger cables you plug into the car are noticeably thinner/more flexible (though still much heavier than an HPWC cable.)
 
  • Informative
Reactions: Rocky_H and Fiver
Nearly twice the current means you need nearly twice the wire cross-sectional area to get the same resistance/heating. The Urban charger cables you plug into the car are noticeably thinner/more flexible (though still much heavier than an HPWC cable.)

The urban charger around my area have the same cable thickness as the normal superchargers. Maybe they are experimenting with different cables.
 
Nearly twice the current means you need nearly twice the wire cross-sectional area to get the same resistance/heating. The Urban charger cables you plug into the car are noticeably thinner/more flexible (though still much heavier than an HPWC cable.)

Yes, that is pretty obvious, but the question I have is whether this is significant considering the overall cost of the equipment and installation. I suppose the thinner cable at the "pump" is nice in the winter when they stiffen up.
 
Yes, that is pretty obvious, but the question I have is whether this is significant considering the overall cost of the equipment and installation. I suppose the thinner cable at the "pump" is nice in the winter when they stiffen up.

Whether it's a significant factor in the overall costs I don't know - I'd assume the copper is a fairly small part of the total cost for a supercharger installation. The post I was responding to said that since it's such a huge amount of current to begin with, they figured the wire size was essentially the same, which seemed unlikely to me.
 
Interesting. I've only been to one Urban type, up in the Chicago area, and it definitely seemed significantly slimmer to me.

Copper vs. aluminum?

The only urban charger I used the cable seemed to be the same 120 kW ones, I didn't measure, but one complaint was that the cable seemed to be twisted somewhat. I had to actually put quite a strong grip to turn the receptacle 15 degrees to insert the cable into the car, and then the same amount to hook it back into the charger holder. I don't think a senior lady would be able to do this. Was this just careless assembly of the cable during manufacture? or do the installers adjust this?
 
I don't know where you got the idea, but I do not believe there is any circumstance where the urban charger delivers more total power to cars than the sharing Superchargers, and there are a lot of circumstances where it delivers less.

Think about it: the sharing system always has the option of a fifty fifty split, and will use it when appropriate, but the urban system can never give a car more than 72 kW, even if there's no car on the other side or the other car is at the top of the taper and can't use the power.

The Urban sites I have been to are split 3 ways, A, B, C. Not 50-50. I have always gotten 72kw, but I had no idea the status of the neighboring cars (no one around)