Tesla Supercharger network

[EcoHeliGuy]

Without marketing, you can see the rolling popularity of the model S, not only are there more in California, but every car seen sells another car so the more that show up in Portland the more that show in Seattle. And the more that show in California, it's a snow ball. The Model 3 will have the same effect of a different price point crowd.

What I'm interested in, is if the Model 3 gets a fairly larger margin of efficiency, and has a smaller capacity battery. How long will they stay at a supercharger?

Maybe battery capacity will matter more because the closer the pack is to full the slower the charge rate, if you have a smaller pack then more time might be spent gathering the needed final range. But if they have a fairly large battery option like say 70Kwh your stop at the supercharger potentially could be very short in comparison to the S/X

 

[AWDtsla]

as I am coming to the end of my trip from FLA to CO and back I've come to the conclusion that Tesla has overbuilt the SC network. I do not mean in locations I mean in number of units per location. Outside of CA and some other urban areas the capacity is way overbuilt. the newer SCs have up to 8 units. In many thousands of miles driven across the US I rarely (outside of CA or the NE corridor) encounter another Tesla at a SC. an example: Is there really a need to have 8 units installed in Trinidad CO? I think that unless tesla is planning for the future of thousands of model 3s overwhelming the system, they've overbuilt in many locations

The incremental cost of adding supercharger stalls is small. The limitation is still the location of superchargers however, so I would not dare call it "overbuilt"

 

[Cottonwood]

as I am coming to the end of my trip from FLA to CO and back I've come to the conclusion that Tesla has overbuilt the SC network. I do not mean in locations I mean in number of units per location. Outside of CA and some other urban areas the capacity is way overbuilt. the newer SCs have up to 8 units. In many thousands of miles driven across the US I rarely (outside of CA or the NE corridor) encounter another Tesla at a SC. an example: Is there really a need to have 8 units installed in Trinidad CO? I think that unless tesla is planning for the future of thousands of model 3s overwhelming the system, they've overbuilt in many locations

Except for the so called "Mobile Superchargers", the minimum size has been 4 stalls and 2 Supercharger cabinets. This provides redundancy for a failed Supercharger Cabinet which I have seen happen every once in a while. Suburb availability is a very good thing!

Once a 4-Stall site is being considered, as others have said, there is not much incremental cost to going to 6 or 8 stalls. That incremental cost is far less than coming back and upgrading later. Given that Tesla is planning to increase sales, and is hoping to do so in a major way with the Model 3, putting in 6 and 8 stall sites makes very good fiscal sense, even if they are not heavily used now.

Also, see the thread below that I started on true capacity of various size Supercharger Site sizes. For the same low 2% probability of finding a queue, an 8-Stall Supercharger has about 3.5x the capacity of a 4-Stall Supercharger. I bet the cost of an 8-Stall Supercharger is about 33% more than a 4-Stall Supercharger, so for 1.33x the investment, Tesla is getting 3.5x the service capacity; that is a pretty good deal if you believe "Build it and they will come!"

Capacity of Superchargers Using an Erlang-B Model

 

[dhanson865]

Except for the so called "Mobile Superchargers", the minimum size has been 4 stalls and 2 Supercharger cabinets. This provides redundancy for a failed Supercharger Cabinet which I have seen happen every once in a while. Suburb availability is a very good thing!

Once a 4-Stall site is being considered, as others have said, there is not much incremental cost to going to 6 or 8 stalls. That incremental cost is far less than coming back and upgrading later. Given that Tesla is planning to increase sales, and is hoping to do so in a major way with the Model 3, putting in 6 and 8 stall sites makes very good fiscal sense, even if they are not heavily used now.

Also, see the thread below that I started on true capacity of various size Supercharger Site sizes. For the same low 2% probability of finding a queue, an 8-Stall Supercharger has about 3.5x the capacity of a 4-Stall Supercharger. I bet the cost of an 8-Stall Supercharger is about 33% more than a 4-Stall Supercharger, so for 1.33x the investment, Tesla is getting 3.5x the service capacity; that is a pretty good deal if you believe "Build it and they will come!"

Capacity of Superchargers Using an Erlang-B Model

Is Madison, WI an exception to the rule? It's 3 stalls and 2 supercharger cabinets or is it a temp location?

 

[emupilot]

Is Madison, WI an exception to the rule? It's 3 stalls and 2 supercharger cabinets or is it a temp location?

Madison was going to be 6 stalls but the property owner was concerned about losing that many spaces and asked that it be cut back to 3 at the last minute.

 

[dhanson865]

Madison was going to be 6 stalls but the property owner was concerned about losing that many spaces and asked that it be cut back to 3 at the last minute.

so is it 3 stalls 3 cabinets or 3 stalls 2 cabinets?

 

[Cottonwood]

Terre Haute, IN under construction. Should be done in a couple of weeks.

5600 E, New Margaret Drive, Terre Haute, IN 47803

 

[cpa]

Also don't forget the reduced charge received by the second to charge in a pairing of stalls. It is entirely possible that many of these less-frequently used locations do get more activity during holidays and summer vacations. So, the chances of not being the second one to charge decrease substantially when there are six or eight charging stations instead of four. In fact, when I was charging in Topeka one Friday evening last May, I was the fourth car to arrive, and this was about 9PM. Topeka has eight stalls. Found the lone unpaired stall to get max rate.

I would suspect that this situation occurs more frequently than we think, and will only increase as more Teslas are on the road!

 

[wdolson]

At this point Tesla's standard supercharger kit might be 8 stalls. ie the part the customer doesn't see that takes power from the transformer, converts it to DC, and distributes it to the SC stalls might be made standard with 8 outputs to support 4 charging stands with 2 ports each. In some areas they may leave some outputs unused if they can't get 8 parking spaces to use, but that may be the default capacity and where they can, they build out that number of ports at a location. From a manufacturing point of view it makes sense to make everything as standardized as possible, even if some capacity isn't used in some locations, it's still cheaper to make one piece of hardware than several different ones that do the same job just different capacity.

 

[FlasherZ]

Ugh, don't make me get out my AT&T/Bell traffic engineering books...

 

[bmah]

Ugh, don't make me get out my AT&T/Bell traffic engineering books...

Hrm, do Teslas showing up at a Supercharger follow a Poisson arrival process?

 

[Cottonwood]

Ugh, don't make me get out my AT&T/Bell traffic engineering books...

Hrm, do Teslas showing up at a Supercharger follow a Poisson arrival process?

Why wouldn't the arrival rate be Poisson? Wikipedia: Poisson distribution

Of course the average arrival rate will change with hour of the day, day of the week, day of the year, etc. Other things that can distort the arrival rate are convoys of Teslas.

No need to look up any tables in books anymore. Here is a good, online, Erlang calculator: Erlang B Calculator

Do you think Anger Erlang ever thought about owning the domain, erlang.com, where that calculator lives?

 

[Drucifer]

Why wouldn't the arrival rate be Poisson? Wikipedia: Poisson distribution

I don't see what french fish have to do with anything

 

[jrreno]

I do agree that the rate of Supercharger installs seems to be slowing down, but Champaign is a good example of a site that has been in progress for a while, but has not been observed because of its relatively remote location compared to many Tesla owners.

See Supercharger - Champaign, IL - Page 6 for some recent pictures.

I bet that Terra Haute, IN is in a similar state of Construction, but no one has found it yet.

I keep hoping to hear you say the same about Erie, PA.

 

[jerry33]

I don't see what french fish have to do with anything

They help you slither into that SC charging spot.

 

[bmah]

Why wouldn't the arrival rate be Poisson? Wikipedia: Poisson distribution

Of course the average arrival rate will change with hour of the day, day of the week, day of the year, etc. Other things that can distort the arrival rate are convoys of Teslas

I had in mind that queueing effects, etc. from charger congestion and/or paired charging stalls would distort the statistical independence of arrivals at other Supercharger locations. (San Mateo Supercharger, I'm thinking of you.) Admittedly, my memory on some of this stuff is pretty rusty, and those effects might not be significant.

No need to look up any tables in books anymore. Here is a good, online, Erlang calculator: Erlang B Calculator

Do you think Anger Erlang ever thought about owning the domain, erlang.com, where that calculator lives?

Or erlang.org, for the Erlang langauge?

 

[Doug Martoccia]

as I am coming to the end of my trip from FLA to CO and back I've come to the conclusion that Tesla has overbuilt the SC network. I do not mean in locations I mean in number of units per location. Outside of CA and some other urban areas the capacity is way overbuilt. the newer SCs have up to 8 units. In many thousands of miles driven across the US I rarely (outside of CA or the NE corridor) encounter another Tesla at a SC. an example: Is there really a need to have 8 units installed in Trinidad CO? I think that unless tesla is planning for the future of thousands of model 3s overwhelming the system, they've overbuilt in many locations

Tesla is building out the SC network to support expected max usage (e.g., 3-5pm on the last day of a holiday weekend), not average usage. I would consider Tesla successful as the number of complaints on this forum related to excessive wait times is reasonable. Tesla has quickly fixed bottlenecks. For instance, Hawthorne, Gilroy, Harris Ranch, Barstow, and San Juan Capistrano all had excessive wait times in the past that have been reasonably mitigated. (I know - I've been caught twice in a long line) Any appearance of "overbuilding" is most likely a result of planning for peak usage going out for the next 18-24 months. Each year the number of Tesla's on the road is just about doubling - many of these idle stalls will get filled up by Labor Day 2017.

 

[kort677]

thanks for your response, however if you read my post I specifically said excluding CA. I just drove over 2k miles across the country and in all of my stops to charge I encountered all of one other car charging, and he was a local. I still think that the system is overbuilt but I now I grasp that the costs of adding the extra stalls is low, so I supposed it is a moot point that I've made.

 

[ecarfan]

I don't think the current SC network is overbuilt on the long distance semi-remote travel routes that Tesla currently has charging available for (and of course there are still some east-west routes in the middle of the US that need to be completed). Tesla is planning for 5 years from now when there will be maybe an order of magnitude more Teslas on the road then there are now. At that point many of the existing SCs can be expanded, and new SCs built where needed.

 

[mmccord]

Given a near doubling of Teslas on the road every year, if there are two cars there today, there will be 8 two years from now, and 16 in three years. Suddenly they don't seem so overbuilt...