-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
LagunaHJames
Member
I could not agree more. We are getting ready to head out on our 5th long road trip (about 3000 miles this time) and it would really help to know the status at our next destination in advance. BTW, I received an email last Friday to alert me to the situation at Hawthorne which was helpful because I was thinking about stopping in on the way back from a LA trip. I adjusted our route and drove more conservatively and made it home with 30 miles range.
vdiv
Chief Grump
It seems the usefulness of realtime occupancy will be limited due to the following:
-- supercharging is quick, just because the stalls are occupied now does not mean they will be in 15 minutes and vice versa;
-- there may be a queue of cars waiting;
-- must be combined with the stalls' status, otherwise they may appear available, but they may not be functioning;
-- when traveling, often we don't have a choice and have to use the supercharger, whether it is busy or not.
-- supercharging is quick, just because the stalls are occupied now does not mean they will be in 15 minutes and vice versa;
-- there may be a queue of cars waiting;
-- must be combined with the stalls' status, otherwise they may appear available, but they may not be functioning;
-- when traveling, often we don't have a choice and have to use the supercharger, whether it is busy or not.
I definitely agree that they not only need real time status, but also historical status of that day. Like if it's a Friday at 5pm or something.
Data they already have, but have chosen not to share it with us for some reason.
What they don't likely have, and what I'd like is whether it's been ICED. At the Dalles in Oregon it's common for most of the stalls to be iced. So it would be nice for Tesla to keep tabs on it, and to alert the hotel when there is a problem (when a significant number are iced). So they can start to put cones in remaining ones.
It's to the point where the Dalles might not even be usable till the next day if you get there too late. They have that weird 30 min parking before 8pm (or something like that), but then don't say anything for past 8. So a person in a ICE car parks right there.
Data they already have, but have chosen not to share it with us for some reason.
What they don't likely have, and what I'd like is whether it's been ICED. At the Dalles in Oregon it's common for most of the stalls to be iced. So it would be nice for Tesla to keep tabs on it, and to alert the hotel when there is a problem (when a significant number are iced). So they can start to put cones in remaining ones.
It's to the point where the Dalles might not even be usable till the next day if you get there too late. They have that weird 30 min parking before 8pm (or something like that), but then don't say anything for past 8. So a person in a ICE car parks right there.
They do this in Switzerland with their trains. Each train has a projected occupancy. Having this info would be enormously useful when trip planning for the SpC network.
I fail to see how occupancy status would help. For most of us (outside of California, and even some in California) we don't have a choice of stopping or not, we need to stop to charge to get where we're going. Even if we could detour to a different route, that decision would usually be made ahead of time and the occupancy at that moment will be different from when you get there. Finally, how do you know they'll be there like hours? Whether occupancy is reported or not, you don't know how long cars will be charging there.
That's not always the case. Today somebody wrote here about a Model S blocking a supercharger for days.
Under each car icon, it could show how long the car has been there. If you see some of the cars have been there for a few hours, don't expect them to be gone when you arrive there in 30 minutes. Each sensor could start a timer from the moment the stall is occupied and then stop the timer after it clears. This photo is from an app they use in Hong Kong. It shows how many stalls are occupied. Imagine under each car icon it would also show the time it has been there. The topic for that app is here. The data is entered manually when somebody arrives and leaves the station. Obviously that is not very good but it is better than nothing. However what is interesting is that, at one location they managed to automate this process.
It seems like a lot of work for little benefit. I'd rather Tesla's supercharger team devote its efforts to increasing the number of locations and expanding existing ones if they're overcrowded. Which is what they've been good at doing so far. Even if I knew a certain location was historically busy at certain times, if I need to charge there, then I need to charge there. Understand I'm from Texas where our routes are straight lines and you can't skip a station and the only time anyone has ever had to wait was the day when owners were driving to Austin to see Elon-- actually I'm happy to see someone else charging when I am.
Busy at certain days/times usually means lots of people are traveling then. If they're traveling, they're not likely to occupy a space for hours, they're charging and going on their way. At a 8 stall station, even if people are charging an average of 45 minutes that means on average a space will open up every six minutes. I'll wait my turn.
SW2Fiddler
We Are Cognitive Dissidents
Unless you're local, I don't really see how real time status is going to help you much - at the time you're making decisions, you have a couple hours drive before you reach the Supercharger, during which the status is almost certain to change (unless it was all empty and remains that way, in which case it isn't an issue.)
What you do need, as Apacheguy touched on, is a projected status - what Tesla expects the availability to be at the time you would get there.
As I've said a couple times, I'm pretty sure this is coming in a little while. Once they get the Navigation routing through Superchargers working well and the interface on it sorted out, It's only natural to have it pull Supercharger statuses for the upcoming chargers. If Tesla sets up a server where they keep all of that information, the car can easily make "reservations" for you when it does the check. (The stalls would presumably still be first come-first served, this is just recording on the server your likely period of occupancy.)
If all the cars do that, suddenly Tesla knows that even though Gilroy is empty right now, in two hours it'll be full or overfilled by cars arriving from three different directions. The system can then respond by changing the navigation plans of cars currently charging at Harris Ranch, Petaluma, and Corning to charge higher and bypass Gilroy in favor of one of the other bay area stations (and route any new navigation plans entered around the station until the blockage clears or all the options are loaded.)
It's a cheap way for Tesla to improve the utility they get from the same hardware investment, so there's no doubt in my mind that some variation of this will be coming - possibly biased by historical data for different days/times at a given station in addition to the Navigation group-think described above.
Walter
Which sometime is the case. For example, Sunday evening when thinking of a place for dinner, I can see 8 stalls out of 12 is avail at Rancho Cucamunga or check Blink/ChargePoint to see what other L2 are avail. That help me decide should I go there or not....
Or on my way to San Jose, the map day for me to stop at Tajon Ranch is optional and I can make it to Harrish Ranch with 10% left. While going, I can see if the place is 90% capacity or 20%, that will let me make wild guess to stop or not. If 90% there is a greater chance I'll have to wait average 50-70 minutes, including charge time. If 20% there is a good chance my wait is 20-30 mins for charge time.
So knowing in advance the availability of the stations help decide direction I would go.
I hate going there, happened three times now that I get there and it's 100% full and end up waiting an hour. If I know in advance it is full or close to it, I'll consider another time to go....
Or on my way to San Jose, the map day for me to stop at Tajon Ranch is optional and I can make it to Harrish Ranch with 10% left. While going, I can see if the place is 90% capacity or 20%, that will let me make wild guess to stop or not. If 90% there is a greater chance I'll have to wait average 50-70 minutes, including charge time. If 20% there is a good chance my wait is 20-30 mins for charge time.
So knowing in advance the availability of the stations help decide direction I would go.
I hate going there, happened three times now that I get there and it's 100% full and end up waiting an hour. If I know in advance it is full or close to it, I'll consider another time to go....
Cottonwood
Roadster#433, Model S#S37
This will help in the future when there is a lot denser set of Superchargers.
Today, an example of that future is at the triplet of Superchargers, Culver City, Hawthorne, and Redondo Beach, all within 15 minutes or so of each other. They have 12, 12, and 8 stalls respectively for a total of 32 stalls, and are all within 15 minutes of each other.
If I were driving from Oxnard to San Diego, or leaving LAX after the car sat for a while, I would like to know as I was approaching this triplet, which one had the most open stalls. In busy times, that would greatly reduce the chances of me finding all stalls full.
See Capacity of Superchargers Using an Erlang-B Model where I discuss real capacity with low blocking rates (finding all stalls full, and having to wait). The table there goes up to 12 stalls. For grins, I looked up the Erlang capacity for 32 stalls with the same blocking rate of 2%; it is 23.7 Erlangs or average stalls in use. For a 12-stall Supercharger, it is 6.60 and for an 8-stall Supercharger, it is 3.6. If we add up the three individual Supercharger Site Capacities, we get 2*6.6 + 3.6 or 16.8. Put all 32 together in one group, and you get an Erlang capacity of 23.7.
That is an increase in capacity of this triplet of Superchargers of 41% just by providing good info to drivers and directing them to the least used Supercharger!
Of course, that is best case result, assuming all drivers use the info, but it is a very good direction without having to build more Superchargers. As all those Model X's and Model 3's hit the road and a corresponding density of Supercharger Sites appear, efforts like this will be a very nice multiplier.
Today, an example of that future is at the triplet of Superchargers, Culver City, Hawthorne, and Redondo Beach, all within 15 minutes or so of each other. They have 12, 12, and 8 stalls respectively for a total of 32 stalls, and are all within 15 minutes of each other.
If I were driving from Oxnard to San Diego, or leaving LAX after the car sat for a while, I would like to know as I was approaching this triplet, which one had the most open stalls. In busy times, that would greatly reduce the chances of me finding all stalls full.
See Capacity of Superchargers Using an Erlang-B Model where I discuss real capacity with low blocking rates (finding all stalls full, and having to wait). The table there goes up to 12 stalls. For grins, I looked up the Erlang capacity for 32 stalls with the same blocking rate of 2%; it is 23.7 Erlangs or average stalls in use. For a 12-stall Supercharger, it is 6.60 and for an 8-stall Supercharger, it is 3.6. If we add up the three individual Supercharger Site Capacities, we get 2*6.6 + 3.6 or 16.8. Put all 32 together in one group, and you get an Erlang capacity of 23.7.
That is an increase in capacity of this triplet of Superchargers of 41% just by providing good info to drivers and directing them to the least used Supercharger!
Of course, that is best case result, assuming all drivers use the info, but it is a very good direction without having to build more Superchargers. As all those Model X's and Model 3's hit the road and a corresponding density of Supercharger Sites appear, efforts like this will be a very nice multiplier.
You don't necessarily need all the drivers to listen - just a large enough percentage to alleviate the blockages. That's why I'm thinking the Nav integration is a critical piece of doing it right - when you're on a road trip running under the car's Navigation, it can show up in the same fashion as a traffic alert - "predicted 20 minute delay at Hawthorne SpC, re-route to avoid?" - and if it's connected in well with Nav and Autopilot, it can be almost seamless, with the car changing course and calculating new charge and arrival times on the fly.
I suppose it could eventually get more sophisticated, too - "here are your three course/charging options, with the estimated times for each and a list of amenities at each SpC site." Though that might be too overwhelming for some people who'd rather just get a set of directions and go with as little hassle as possible.
Walter
Cottonwood
Roadster#433, Model S#S37
I agree completely! It will take a reasonable percentage of users looking for idle Superchargers; is that fraction 30% or 75%; that will take some interesting simulations to figure out.
Hopefully, in the future, Tesla can do better than the current Nav through Chargers implementation. They certainly have set a low bar to get over for the next version...
That's the wonderful thing about Tesla - you know a new software version is right around the corner and loadable with minimal hassle - which hopefully improves things.
What I've read about the current solution suggests it needs a bunch of work - but I have confidence they'll get it figured out eventually. They have most of the key pieces of a really good solution basically in place (the predicted energy consumption is huge and AFAIK unique for in-car systems) - they just need to do a little tuning of the algorithms.
Similar threads
- Question
