Tesla Supercharger network

[HankLloydRight]

Here though the SCs were about 110 miles apart. I charged to 170 for each leg as buffer. Each of the four stops were about 20 minutes. It worked out really well. We'd stop, plug in, hit the local coffee shop or grocery store to get a drink and use the rest room, and the car was pretty much at the range we needed by the time we got back. 11 hour trip and we were slow at lunch and at the Vacaville SC which had a bunch of stores. We could have easily made it in 10 hours. I cranked at 75-80mph most of the way.

This was also my recent experience with two east-coast to mid-west SC trips. I found that by driving between 70 and 80 most of the way, I was able to average about 330 wh/m running about 120% of rated miles. So if I needed to go 150 miles, I would charge to 140% of 150 (or about 210 miles) with more than enough buffer to get where I needed to go, and never had a problem with range. And most of my SC stops where between 15 and 45 minutes -- the longer ones where I got lunch or dinner. Very easy overall!

 

[scaesare]

Exactly. It's useless information. If it shows 4 of 6 stalls are being used, two could be ICEd and there could be 3 Teslas waiting for the 4 non-iced stalls or the to ICEs to leave. Also, you don't know how many other Tesla owners are about to arrive at a SC location, so even though there could be one or two stalls free, three or four cars could show up. You really don't know what the *actual* situation is until you arrive.

All that infoboard tells you is how many are charging now. Nothing about ICEd spots, cars waiting, inoperative stalls, etc.

I'm of the opinion it's interesting because, although there was no doubt that Tesla could track that information in that manner, we now know they are doing so, and that the Tesla Mothership can parcel that information out with such granularity.

Given that there has been discussion of future updates to the car's nav system to incorporate smart(er) trip planning, it's intriguing to speculate that they might make use of this data on the car's console for our benefit...

 

[HankLloydRight]

I'm of the opinion it's interesting because, although there was no doubt that Tesla could track that information in that manner, we now know they are doing so, and that the Tesla Mothership can parcel that information out with such granularity.

Given that there has been discussion of future updates to the car's nav system to incorporate smart(er) trip planning, it's intriguing to speculate that they might make use of this data on the car's console for our benefit...

But there are still to many unknowns, like ICEd spots or the number of cars waiting for an open spot. Without quite a bit of additional hardware and software at each stall (which I doubt will ever happen) and some system to track who's waiting, for how long, and how long they need to charge, it's going to be pretty hard to really know the status/availability of any SC location. Unless they somehow eliminate ICEing and create a new ticketing/line system for waiters, I just don't see the number of stalls currently being used as useful information for really anything other than knowing just that -- something "cool."

 

[AlMc]

But there are still to many unknowns, like ICEd spots or the number of cars waiting for an open spot. Without quite a bit of additional hardware and software at each stall (which I doubt will ever happen) and some system to track who's waiting, for how long, and how long they need to charge, it's going to be pretty hard to really know the status/availability of any SC location. Unless they somehow eliminate ICEing and create a new ticketing/line system for waiters, I just don't see the number of stalls currently being used as useful information for really anything other than knowing just that -- something "cool."

I still see the information as valuable. If you are approaching a SC and absolutely need a charge you are going to stop regardless of the information. If you don't need a charge but would like to top off to be able to skip the next SC you would probably pull in. If the SC you are approaching is 'full' and you can wait till the next SC or home for your charge you will bypass the full SC. Repeat that approach during your journey.
Yes, it can't see an ICED spot and there may be other people heading for the same SC so the information is not perfect but I do think it has some value beyond the 'cool' factor.

 

[ecarfan]

While it's true that Tesla's SC data cannot detect "ICEing" or know how many Teslas are waiting to charge, that can be inferred to some extent. If an SC shows "full" for a significant percentage of the time than likely there are cars waiting to charge as well. This info can tell Tesla that the SC needs expanding, and I'm sure there have already been instances where Tesla has done that (Gilroy, Hawthorne). And that info can be used to inform owners of busy stations, and in the future when stations will be far more plentiful than they are now, can be used to route owners to alternate stations.

 

[andrewket]

If there are three Superchargers, there should be 6 stalls.
The first three will have a higher priority than the second three. However, if a car with single digit miles remaining plugs into the second three, they will prioritize that car for a little while. The controller boards are set up as a Master and Slave system, each controlling one stall.

The numeration goes like "Supercharger / Node"
If you see a 3A, it means it's the Master on the third Supercharger. All "B" nodes are Slaves and get a lower priority when charging unless the aforementioned mileage scenario is present.

So, if there is an "A" stall available, plug into that one first. If there are only "B" nodes available, plug into the one's which Master node has been occupied the longest. You can usually tell by the fan noise coming from the front of the car or the back of the Supercharger (if you can discern the noise).

A plethora of Supercharger information was given to me recently, but I promised to not digitally share that information.
There are some "Easter Eggs" that Fremont puts in them that are pretty hilarious.

All of this is public knowledge and had been discussed. Now the Easter eggs.. Do share.

 

[Lump]

Dashboard of Barstow on 7-4-2014 at 12:45 pm PST

60 minute average charging time :cursing:

 

[dhanson865]

Dashboard of Barstow on 7-4-2014 at 12:45 pm PST

60 minute average charging time :cursing:

40 minite average at hawthorne with plenty of other chargers in the area, why does it surprise you or anger you that one in a more remote area might see people charging longer?

 

[gg_got_a_tesla]

Love this new dashboard! Hope some form of it comes to the car in 6.0 and to the Tesla website as well.

Wonder if the occupied stall icon can show the real color of the car that's plugged in?! I guess Tesla has the VIN since it was presumably conveyed to the supercharger at the time of plugging in. Fluff feature but, would be cool to see!

 

[Lump]

40 minite average at hawthorne with plenty of other chargers in the area, why does it surprise you or anger you that one in a more remote area might see people charging longer?

Add another hour waiting for a stall to open during peak times, almost 2 hours in the Chillies parking lot in Barstow isn't fun.

 

[scaesare]

But there are still to many unknowns, like ICEd spots or the number of cars waiting for an open spot. Without quite a bit of additional hardware and software at each stall (which I doubt will ever happen) and some system to track who's waiting, for how long, and how long they need to charge, it's going to be pretty hard to really know the status/availability of any SC location. Unless they somehow eliminate ICEing and create a new ticketing/line system for waiters, I just don't see the number of stalls currently being used as useful information for really anything other than knowing just that -- something "cool."

Well, just because that's all the new data that's exposed on that dashboard doesn't mean that's all that's being collected and can be used.

For instance, if they know that there are open stalls, then they can reasonably extrapolate that the number of cars waiting = 0.

Furthermore, if the idle time per stall is above or below some threshold, there's probably a good chance they can determine if there's "saturation" at a SC during a given period.

We also don't know if they are collecting stats on the cars... they could be grabbing the "miles since last charge" data to see how far people are travelling on average before hitting specific superchargers at specific locations... and there's reason to believe that they can determine if the previous charge was was ALSO at a supercharger[1]. We know they mark chaging waypoints in the NAV system, and the data conversations that have been sniffed thus far show that the VIN is passed to the supercharger during the handshake. This means that each individual car's charging history could be logged, including date, time, location, length of charge, starting and ending SOC's, avg Wh/mi utilization since last charge, and how busy a site was when your car was there.

Thus, with tens of thousands of data points, Tesla might be able to draw some pretty interesting conclusions about how individual sites are used... what charging patterns on major routes look like at specific times, how people are using the chargers, etc...

They may use some of this data internally for planning and capacity management. Some may find it's way in to future FW updates for car features. And who knows what other tricks they may have up their sleeve.

I don't see this as just "a pretty picture of a full or empty stall", I see this as "early indicators that Tesla has data collection capabilities that may enable some truly innovative mechanisms for managing the growth of their SC network as well as how we 'fuel up' in the future".



[1] Visible Tesla exposes this flag, as it's author, Joe Pasqua, saw this variable in the datastream passed on from the car and built it in to his app

 

[Bugeater]

And from the interval between cars charging and empty spots an educated guess about ICEing could be extrapolated!

AKA, if the interval between charges in busy spots was just long enough for one car to pull out and another to pull in and plug in, then adjacent spots not in use may reasonably be assumed to be ICEd.

 

[HankLloydRight]

You guys are all looking "pie in the sky" under a perfect set of assumptions and an awful lot of "if"s and "could"s. But in reality there are just too many unknown variables to accurately or reliably determine anything of actionable value here. Sure, in a perfect world there "could" be, but we are not even close to any of the scenarios mentioned above, if you want to live in a dream world with perfect data all the time, go ahead, I surely can't stop you.

 

[emupilot]

Dashboard of Barstow on 7-4-2014 at 12:45 pm PST

60 minute average charging time :cursing:

That's not surprising considering it's 90 kW and some people stopping there are making the long run to Kingman. Tesla has plans to expand it (4 additional stalls, presumably 135 kW) although I don't think there has been any news on the expansion recently.

 

[AlMc]

You guys are all looking "pie in the sky" under a perfect set of assumptions and an awful lot of "if"s and "could"s. But in reality there are just too many unknown variables to accurately or reliably determine anything of actionable value here. Sure, in a perfect world there "could" be, but we are not even close to any of the scenarios mentioned above, if you want to live in a dream world with perfect data all the time, go ahead, I surely can't stop you.

Sorry if I was part of the 'pie in the sky' and offended you in some way. Having that info is in no way perfect but having more information about the SC status would be helpful if there are a number of SCs within your driving range and you decided to go to one or skip one based on the information supplied.

Aside: Doubt this will be in the 6.0 anyway. I suspect TM will use the information more internally for planning future additional stalls or sites as there are more TM cars on the roads.

 

[scaesare]

You guys are all looking "pie in the sky" under a perfect set of assumptions and an awful lot of "if"s and "could"s. But in reality there are just too many unknown variables to accurately or reliably determine anything of actionable value here. Sure, in a perfect world there "could" be, but we are not even close to any of the scenarios mentioned above, if you want to live in a dream world with perfect data all the time, go ahead, I surely can't stop you.

No, all I'm saying is that "some data" is better than "no data".

And, although what's on that dashboard display is a couple of new pieces of data, we have some evidence that more is available to Tesla.

So no, I'm not living in a "dream world with perfect data all the time". What I am pointing out is the possability that we might move forward in the 21st century to some different models of charge management that are a step ahead of the one used by gas stations since the time they started replacing places to buy feed for your horse.

But if you are determined to believe that the tip we are currently seeing is all that there could possibly be of the iceberg, than "go ahead, I surely can't stop you."

 

[Mayhemm]

How about just putting a webcam on a pole at each site, accessible through your Tesla account?

Boom! All your requested info made available, plus a little added site security for Tesla.

 

[Benz]

Tesla Motors have placed a provisional Supercharger station in France (south of Lyon). I guess that they will be placed in the US as well. Any thoughts?

 

[emupilot]

Tesla Motors have placed a provisional Supercharger station in France (south of Lyon). I guess that they will be placed in the US as well. Any thoughts?

There are portable supercharger installations at Bethesda, Maryland and Lone Tree, Colorado. I think they also used one to expand Hawthorne, California. The Bethesda installation was theoretically temporary, although they didn't remove it after a permanent supercharger came on line 35 miles away.

 

[GSP]

The pedestals could be modified to be able to detect the presence of a vehicle. Vehicle not plugged in = ICEd. We've also discussed a reservation system incorporated or working with the nav. That would give the system knowledge of how many cars en route, ETA, etc.

I would love to go build this system. If anyone at Tesla is reading this, hire me

I think Tesla execs, when questioned about vandalism, have mentioned that at least some of the SuperCharger sites have video monitoring. It would be trivial to automatically detect ICEd stalls with modern video recognition systems. Tesla almost certainly has experience in both their factory and their autonomous driving team.

GSP