Future Charging for Model S 1-phase or 3-phase ?

[TEG]

TEG, don't forget that DC fast chargers do use 3-phase as input. In the US, it seems the slower "Level 2" EVSE units (such as the current SAE J1772 AC Level 2 chargers/EVSE units) are the ones which use single and/or split phase (I'm not not an expert, just following the public discussion of this subject)...

Yes, I think you understand the situation.

most american has the idea, their current grid mostly based on single-phase or even split-phase are sufficient for the future. the answer is NO! Even if you got a 100A fuse, the utility gives you an average of max. 10A consumption over the year.
While as long as your are the only one in the neighborhood its fine. But if half of your neighbors owns an electric car as well, everything changes. The local transformer is distributing the 3-phases to the homes, giving every street one of the phases. If to many cars start charging, the transformer will blast or simply switch off under overload. The utility has to setup bigger transformer and to introduce smart grid abilities to limit power-consumption during peak time.
The simplest is monitoring the voltage, the phase shift and to reduce the current, when a grid overload is indicated. With 3-phases, its much easier and the less used phases can be better balanced out...

We have heard Eberhard praise the benefits of 3-phase charging ad-nauseum, and it starts to feel like 'beating a dead horse', but not because the idea lacks technical merit, but more because it doesn't seem 'in the cards' for it to happen anytime soon in the USA. To be clear, the USA grid has plenty of 3 phase to industrial and large business locations. The problem is with the residential feeds. Eberhard is right about the phases getting split apart and fed to different clusters of houses. I know from my own experience here that many power companies will do the absolute minimum they can get away with on infrastructure upgrade expenditures. It would be nice to be proven wrong, but I just don't see 3 phases coming into residential neighborhoods like they have in many parts of Europe. Even if there are valid technical reasons to do upgrades, there are costs, regulations, tariffs, and political forces that slow or prevent such changes.

 

[TEG]

By the way, for what it is worth, the USA is rather slow to get higher speed internet to residential users too...
U.S. Broadband Speed Worst Among 29 Countries Surveyed - John Paczkowski - News - AllThingsD

 

[dpeilow]

I don't think Eberhard is saying that Model S should have 3 phase charging for residential use in the US, just that the car should the ability to make use of it. If the PEM contained in effect three chargers, they could be run on the 3 phases in Europe and paralleled up to run on split phase in the US.

But the point that he is trying to make is that in non-residential situations where DC fast chargers will be located, almost by definition 3 phase is available as it's needed to power the charger in the first place. That holds on both sides of the Atlantic. The open question is whether Tesla (and others) could make a lightweight charger of sufficient power.

But even if a common design isn't possible, the European market is big enough to develop a specific PEM version.


Anyway, Tesla know all this. I and others have sent them the information so if they want to act on it, they can. There's no point in discussing it ad nauseam.

 

[TEG]

Eberhard was clearly talking about residential use with comments like this:

...their current grid mostly based on single-phase or even split-phase are sufficient for the future. the answer is NO! Even if you got a 100A fuse, the utility gives you an average of max. 10A consumption over the year. While as long as your are the only one in the neighborhood its fine...

...Split-phase...100A fuse... 10A average... Those all describe older US homes...

But, yeah, high speed DC chargers like those with CHAdeMO are almost certainly all going to derive their DC from 3 phase 440/480V so if the vehicle had a native 440V 3-phase charger on-board you could skip the off-board DC step assuming a suitable 3 phase AC interconnect. (And as you know even that can be difficult to standardize.) For now, this would only be done at commercial locations.

None of the above negates J1772 as a viable standard for overnight residential charging. If Eberhard was suggesting that USA needs to bring 3-phase to residences for EVs to scale, I think he may be mistaken, and/or disappointed in the resistance to such a change. If Eberhard is simply suggesting that 3-phase charging could work in commercial locations... Well, yeah, but between CHAdeMO and SAE DC, the DC train is leaving the station soon.

But I am back to thinking that Eberhard is/was suggesting residential 3 phase by stating concerns over unbalanced grids and increased average (per house) consumption. Perhaps it isn't the ideal answer, but some of those concerns could be addressed by putting bigger transformers to neighborhoods.

 

[Eberhard]

my personal belief for charging in the public will be inductive charging. being quite save for vandalism but a little less effective with 5-10% loss. but standard are quite in the future. with this, any questions about DC, AC one-or 3-phase won't matter anymore or left for home charging only.

 

[TEG]

I suppose we would need some standards for inductive charging...
Probably some (RF?) protocol so that the inductive charging coil only energizes when a correct/compatible vehicle is parked over it.
Perhaps automatically bill for use too.
I wonder how much power could be safely transferred by such a system? Would it work for "fast charging?" (> 50kW)
So much for the V2G dreams ("Your car pack can help the grid in times of need.") that I don't think would work with one way inductive charging.

It would probably be more efficient if the vehicle could lower a receiver coil closer to the road surface when it detects that the car is parked over an inductive charging spot.

 

[Lloyd]

Were you talking about Level 2 charging, or DC fast charging (or both) ? Elon also mentioned the SAE standards (current and upcoming) in the Q&A section of the recent shareholders meeting, but only in so far as saying something like that they are good even though they could be improved.

They were not specific, and when I started to ask more, the informaiton stopped. The salesperson knew more, but was unable to provide more details.

 

[Lloyd]

Does anyone make an adapter for the Tesla HPC to J1772? I just bought an HPC and would like to be able to charge J1772 vehicles also.

Can this be done safely? Will the pilot signal get passed through, and is the HPC able to respond to this??

 

[TEG]

I have not seen or heard of anyone doing that (yet). I think the electrical signals are fine and "good to go", but the problem is the Tesla socket you would need.
Some Tesla HPCs have been permanently converted to J1772 by chopping off the Tesla plug and replacing it with a J1772. Then you could buy the Tesla J1772 to Tesla adapter cable and do it that way. But if you want to leave it "native Tesla", you would need to find a Tesla socket. I don't know if Tesla will sell you one. Maybe someday someone will convert their Roadster to native J1772 and have a spare Tesla socket to sell. (Tomsax did that temporarily, but I think he put his Tesla socket back on the car again.)

You want the "inverse" of this:

With "female" Tesla and "male" J1772.

 

[Norbert]

But the point that he is trying to make is that in non-residential situations where DC fast chargers will be located, almost by definition 3 phase is available as it's needed to power the charger in the first place. That holds on both sides of the Atlantic. The open question is whether Tesla (and others) could make a lightweight charger of sufficient power.

But even if a common design isn't possible, the European market is big enough to develop a specific PEM version.

There is a good chance you didn't mean it, but this appears to imply (in the context of the statements you refer to) that you are talking about a PEM version which would use 3-phase to replace DC charging. However that would be ignoring the fact that even the Mennekes standard uses DC for fast charging at the levels at which Model S, and future cars, are likely going to use it. 3-phase is not going to make DC obsolete, as was argued above, not even in Europe. (And neither are there any signs that inductive charging would come to that level). This feels like the after-effects of RWE marketing blown out of proportion. I'd rather RWE work on developing low-cost DC chargers (of various power levels), as building a DC network wouldn't really be that expensive in any case.

 

[Lloyd]

I have not seen or heard of anyone doing that (yet). I think the electrical signals are fine and "good to go", but the problem is the Tesla socket you would need.
Some Tesla HPCs have been permanently converted to J1772 by chopping off the Tesla plug and replacing it with a J1772. Then you could buy the Tesla J1772 to Tesla adapter cable and do it that way. But if you want to leave it "native Tesla", you would need to find a Tesla socket. I don't know if Tesla will sell you one. Maybe someday someone will convert their Roadster to native J1772 and have a spare Tesla socket to sell. (Tomsax did that temporarily, but I think he put his Tesla socket back on the car again.)

You want the "inverse" of this:

With "female" Tesla and "male" J1772.

Yes, I want the inverse of what Telsa sells. I would hate to plug someone's Volt or Leaf in and have it implode with 70 amps of power!

 

[scott451]

Does anyone make an adapter for the Tesla HPC to J1772? I just bought an HPC and would like to be able to charge J1772 vehicles also.

Can this be done safely? Will the pilot signal get passed through, and is the HPC able to respond to this??

Probably the easiest way of doing this (assuming you already have the J1772->Tesla) is to convert your HPC to J1772. The cable costs ~$350 from Avnet. I'll be doing the conversion shortly with my HPC. Now after the conversion you'll have a 70A J1772. The J1772 70A cord is much lighter than the HPC 70A cord.

 

[Tommy]

Probably the easiest way of doing this (assuming you already have the J1772->Tesla) is to convert your HPC to J1772. The cable costs ~$350 from Avnet. I'll be doing the conversion shortly with my HPC. Now after the conversion you'll have a 70A J1772. The J1772 70A cord is much lighter than the HPC 70A cord.

Link to Avnet's product page page:

Avnet Electronics Marketing - ITT Interconnct Releases SAE J1772 Connector Solution

 

[TEG]

Yes, I want the inverse of what Telsa sells. I would hate to plug someone's Volt or Leaf in and have it implode with 70 amps of power!

No need to worry about that. The HPC/EVSE would send a signal to the Leaf and say "you can have up to 70amps", and then the Leaf would start pulling 16A (which is the max the current charger in the Leaf can draw) and have no problem with it. Having higher current capability doesn't cause a problem. Some vehicles won't take full advantage of it though because they have 'wimpy' chargers.

 

[Norbert]

But I am back to thinking that Eberhard is/was suggesting residential 3 phase by stating concerns over unbalanced grids and increased average (per house) consumption. Perhaps it isn't the ideal answer, but some of those concerns could be addressed by putting bigger transformers to neighborhoods.

For residential areas where phases might get split apart, one important method to address this will be using smart grid capable EVSE. This, I think, is one of the reasons the DOE pushes smart grid EVSE. Together with time-of-day charging, these allow utilities (to some degree) to balance the load by controlling the time (within the off-peak period, during the night) at which EVs in a neighborhood start and stop charging, and to turn them off if really necessary to prevent a black out (in which they could not charge either). In my mind, EVs capable of a higher charging speed should give the utilities more flexibility in balancing the load, since they don't have to charge most of the night, and thus give more options in choosing the time to charge them, if you know what I mean.

This will reduce the number of cases where bigger transformers need to be installed. And perhaps even more importantly, to remind, charging in residential areas is mostly off-peak, at times when there is much less stress on the grid. Also, the utilities could surely further improve load balancing by having areas that can be switched to any of the 3 phases as needed to balance other areas. That shouldn't be difficult, even if requiring a change in how the areas are connected to the transformers. I guess.

 

[TEG]

Yeah, I follow you... Well, the part about selective phase switching seems like it might be a bit 'far fetched', but I don't really know for sure.

Yes, 'smart grid' EVSE with remote disconnect to reduce load or maintain balance as needed seems to be in the works.

 

[Norbert]

Well, the part about selective phase switching seems like it might be a bit 'far fetched', but I don't really know for sure.

OK, in case that doesn't work well for some reason, here another idea:

We've heard that the utilities in the US, even though they don't usually install 3-phase in residential areas, can still do so if asked for. So, for example, close to each transformer with which they might have difficulties in load balancing, they could add 3-phase service to a number of residences, and for the EVs in these residences, install special EVSE. These would work like other J1772 EVSE off a single phase, but in addition to other smart-grid abilities like remote-on-off, it would have remote-choose-which phase-to-run-off. So they would have a certain number of EVSE which they can assign to any phase, and use that to balance the load-per-phase. If something like that is really necessary, it should be an affordable thing(for the utilities) to do.

Problems solved?

 

[GSP]

"Problems solved?"

Maybe, but a three phase charger would be a much more elegant solution.

Perhaps use the DC fast charge port with an "slower" 5-15kW 3-phase off-board charger. The main (only?) benefit would be a balanced load for the utility.

GSP

 

[TEG]

Some quotes from a friend who works in the EV charger business in the USA:

I think the three phase balance issue may be bogus. Most neighborhoods I've seen only have single phase feeders from the substation out, so the balance is between large numbers of blocks of houses. The only three phase that is common on poles seems to be in industrial or mixed use areas. Changing houses to three phase would be tremendously expensive.
...
Maybe newer subdivisions are different...

 

[dpeilow]

There is a good chance you didn't mean it, but this appears to imply (in the context of the statements you refer to) that you are talking about a PEM version which would use 3-phase to replace DC charging. However that would be ignoring the fact that even the Mennekes standard uses DC for fast charging at the levels at which Model S, and future cars, are likely going to use it. 3-phase is not going to make DC obsolete, as was argued above, not even in Europe. (And neither are there any signs that inductive charging would come to that level). This feels like the after-effects of RWE marketing blown out of proportion. I'd rather RWE work on developing low-cost DC chargers (of various power levels), as building a DC network wouldn't really be that expensive in any case.

If I was being greedy, I'd want to be able to charge at 44kW on three phase, 50kW on CHAdeMO and 100kW on whatever Tesla chooses to get the advertised charge time. We are literally seeing all of them being rolled out. If I had to drop one it would be CHAdeMO.

And this is not RWE (or anyone else's) marketing blown out of proportion. We now have close to 5000 Mennekes charging posts announced for the coming year in the UK alone. The AC equipment is one or two orders of magnitude cheaper than DC.