Model 3 Supercharging Capable Discussion

[AZ Desert Driver]

OK - I think you confirmed that one needs 30 minutes. Less time if battery is not empty, less time if you don't "overfill" but allocate 1/2 hr to fill as an average time budget.

Can I also conclude that a typical Blink setup will fill at 33 kWh/hr so the cost at $0.59/kwh for a 30 min fill will be about $17.70?

 

[nwdiver]

If you sat your car at a gas station pump, pumping gas, while you ate a restaurant....everybody in the restaurant would nod in agreement as you scurry to move your car.

You go have diner while pumping gas?

Supercharging takes on average ~30 minutes... if you need more of a buffer it's going to take about an hour. 95% of the time it's irrelevant wether your car sits there for 30 minutes or 3 hours. For those remaining few occasions that higher throughput is needed Tesla has decided hiring a valet for $10/hr is an effective solution... I tend to agree. To apply a fee to 100% of use in an effort to increase availability 5% of the time is crazy. The supercharger pitch is that you stop... plug in... go get lunch... if you start adding the caveat that you need to run back the minute you've got enough juice or you get fined that's not gonna go over well.

This whole argument is based on the idea that a per minute fee will reduce the required number of superchargers by ~50%... which isn't supported by any empirical evidence and... to be blunt... is rather insane. Charging a per minute fee is unlikely to reduce the required size of the network and will add confusion to a technology that already confuses ICE drivers... not to mention piss off the 60, 70 and 'A' battery owners whose cars charge slower.... yeah... it gets really complicated.

What's really amusing is that the pay per use/minute model has failed... everywhere. Everywhere someone has tried to fund a charging network with funds collected by charged by the minute or the kWh they have failed. Tesla has succeeded by funding superchargers with vehicle sales. Call me crazy but if something is working is seems odd to shift to a model that has failed so spectacularly like Blink.

 

[diamond.g]

OK - I think you confirmed that one needs 30 minutes. Less time if battery is not empty, less time if you don't "overfill" but allocate 1/2 hr to fill as an average time budget.

Can I also conclude that a typical Blink setup will fill at 33 kWh/hr so the cost at $0.59/kwh for a 30 min fill will be about $17.70?

$19.40 as you are paying per kWh not for how long you charge.

 

[AZ Desert Driver]

OK - I think you confirmed that one needs 30 minutes. Less time if battery is not empty, less time if you don't "overfill" but allocate 1/2 hr to fill as an average time budget.

Can I also conclude that a typical Blink setup will fill at 33 kWh/hr so the cost at $0.59/kwh for a 30 min fill will be about $17.70?

Post #562 just gave me new data. If I understand his points.. Blink does NOT fill at 33 kWh/hr, so it will take longer. But because I'm buying power, not time, the final fill will still cost $17.70 for 33kW. Plus however you wish to value -Baubles bought at the merchant, my time, convenience and those other soft-costs.

 

[EcoHeliGuy]

The question is:

Does Blink charge for the Parking space or energy consumption?

By time implies the Parking space, by KW implies usage.

In one scenario your a parking provider, in the other your a regulated utility.

 

[diamond.g]

The question is:

Does Blink charge for the Parking space or energy consumption?

By time implies the Parking space, by KW implies usage.

In one scenario your a parking provider, in the other your a regulated utility.

EVgo is arguably worse as they charge a membership fee and a usage fee.

 

[AZ Desert Driver]

$19.40 as you are paying per kWh not for how long you charge.

yer right!! my calculator must have been mashed with a fat finger. 33*0.59=$19.47

 

[AZ Desert Driver]

So this ($19.47) fill up fee - my home 14-50 system that will cost $500 would be paid off with 25 visits to neighborhood Blinks. Just one month breakeven. Sounds like good economics to install home system. Glad I did not have to invest into the Blink capital.

For SC system- local recharge economics don't apply. Just expansion of vehicle usability- which is "priceless"

 

[nwdiver]

Sounds like $10/hr is the low end... I pity the poor fool that dozes off at a vacant supercharger at 2am for 5 hours... because it's 2am, there's no one else there and he needs to sit in his car for 30 minutes... sorry... that accidental nap just cost you $45.

 

[Ben W]

Supercharging takes on average ~30 minutes... if you need more of a buffer it's going to take about an hour. 95% of the time it's irrelevant wether your car sits there for 30 minutes or 3 hours. For those remaining few occasions that higher throughput is needed Tesla has decided hiring a valet for $10/hr is an effective solution... I tend to agree. To apply a fee to 100% of use in an effort to increase availability 5% of the time is crazy. The supercharger pitch is that you stop... plug in... go get lunch... if you start adding the caveat that you need to run back the minute you've got enough juice or you get fined that's not gonna go over well.

If it saves $2k off the upfront price of the car, I still think many people would happily opt into it. And even if you're not blocking other drivers, topping up still costs Tesla money for the electricity. Ultimately this shows up in a higher price for the car or reduced Tesla profits. An incentive to not charge if you don't need to will keep costs lower for everybody.

This whole argument is based on the idea that a per minute fee will reduce the required number of superchargers by ~50%... which isn't supported by any empirical evidence and... to be blunt... is rather insane. Charging a per minute fee is unlikely to reduce the required size of the network and will add confusion to a technology that already confuses ICE drivers... not to mention piss off the 60, 70 and 'A' battery owners whose cars charge slower.... yeah... it gets really complicated.

I never suggested that existing owners would switch to a pay-per-use model. I expect that all Model S/X's will remain forever on the unlimited model. So "60, 70, and 'A' battery owners" will not be pissed off. On the contrary, they would be happier that pay-per-use for Model 3 is helping prevent the flood of new cars from clogging up the system. Even if it helps 5%, it still helps.

What's really amusing is that the pay per use/minute model has failed... everywhere. Everywhere someone has tried to fund a charging network with funds collected by charged by the minute or the kWh they have failed. Tesla has succeeded by funding superchargers with vehicle sales. Call me crazy but if something is working is seems odd to shift to a model that has failed so spectacularly like Blink.

The Blink model has failed when it's for-profit. You know why? Because pay-per-use discourages non-essential use. This is EXACTLY the point. If Tesla implements such a system, it would not be to make a profit off the fees; it would be to reduce congestion and misuse. And for the exact same reasons that Blink has failed, it would be highly successful at accomplishing this.

 

[nwdiver]

If it saves $2k off the upfront price of the car, I still think many people would happily opt into it.

No.... it doesn't... expanding the network to support 300k cars will cost ~$500M... where does that come from? How much of a subsidy do you expect higher optioned cars to provide?

Lets say 50% of those 'happy opt' for pay per use... now you have 300k cars trying to use a network only big enough to support 150k and you've got a fraction of the funds required to expand it... oops.

The Blink model has failed when it's for-profit. You know why? Because pay-per-use discourages non-essential use.

No... Blink never even got that far... they were only trying to cover expenses and expand the network. You need an external source of revenue to make a fast charging network viable.... like vehicle sales.

 

[Ben W]

No.... it doesn't... expanding the network to support 300k cars will cost ~$500M... where does that come from?

If my assumption that pay-per-use will decrease congestion and thus require a smaller network is correct, then expanding the network to support 300k cars may cost only $300m under the pay-per-use structure, versus your $500m under the unlimited structure. Of this $300m, some (say $100m total) would be taken upfront out of the price of every car. (As it currently works for the S/X.) The other $200m would come from pay-per-use fees over time. The maintenance and electricity components of the SC costs don't need to be paid for by Tesla upfront; they are paid over time, so they can be collected over time.

The failure of Blink's business model has shown clearly that people use pay-per-use charging much less than expected. This hurts Blink (because it reduces their profit off the fees), but it would help Tesla (by reducing the capital-intensive build-out required to accommodate demand, which then reduces maintenance and electricity costs).

 

[CuriousG]

Something I don't think that has been brought up is: Would this be the time to reconsider the battery swap option? I know the initial battery swap program was a dismal failure but it might be worth taking a second look once the Model 3 becomes more mainstream. I'm just hoping we never see anything like the Tejon Ranch Christmas debacle.

For those that are new to this here's a video of it in action:

Battery Swap Event

The idea is approximately 3min per swap at a rate of $80. I'm sure in reality it's longer. You have to come back to retrieve your original battery which will be charged. So at a minimum it will be $160. At least that was the idea back in 2013. Battery prices are cheaper so I don't know how that may or will affect their original service price.

This could give give Model 3 owners a pay per use option, albeit one that would be more expensive than current gasoline prices.

 

[nwdiver]

If my assumption that pay-per-use will decrease congestion and thus require a smaller network is correct, then expanding the network to support 300k cars may cost only $300m under the pay-per-use structure, versus your $500m under the unlimited structure. Of this $300m, some (say $100m total) would be taken upfront out of the price of every car. (As it currently works for the S/X.) The other $200m would come from pay-per-use fees over time. The maintenance and electricity components of the SC costs don't need to be paid for by Tesla upfront; they are paid over time, so they can be collected over time.

The failure of Blink's business model has shown clearly that people use pay-per-use charging much less than expected. This hurts Blink (because it reduces their profit off the fees), but it would help Tesla (by reducing the capital-intensive build-out required to accommodate demand, which then reduces maintenance and electricity costs).

- Blink failed because they targeted local users; no one is going to pay $0.50/kWh when it's $0.10/kWh at home... I agree 100% that pay per use would decrease local use... I always have... my point is that local use is maybe ~5% of SC use... as you've stated transient use is unlikely to be impacted. Most people already move on from SCs quickly... there's no reason to think that a per minute fee will significantly impact SC requirements... especially now that Tesla is hiring Valets to increase throughput during peak hours.

- You can't collect $200M upfront over time... I'm referring to it as 'upfront' for a reason... you have to have SC there to pay to use... that's going to cost $200M upfront... not over time.

You can't get around the upfront cost... you can't fund building a future network with that future network. The division between capital and maintenance costs is still largely an unknown. The SC network could operate like a ponzi scheme... which sounds bad but ponzi schemes only fail when they stop growing. Basically as long as Tesla keeps growing then there's no reason $2k/car can't support 'Free Long Distance charging'... once the network is mature a use fee will make sense.

Here's some amusing Pros and Cons to charging $10/hr...

Pro:
- Cars get cheaper by ~5%

Con:
- Lose 'Free Long-Distance charging' for marketing
- Confusion; 'how do you convert $/hr to $/gal @ 40mpg ?!?!?' -confused Tesla shopper
- 'I accidentally fell asleep charging and now I owe $50!... there was no one even there!' -Irate Tesla owner
- 'Why am I paying ~$1 more per charge?' - Model 3 70kWh owner whose car charges slower than a 90.
- More roadside assistance towing costs... people already cut charging close to same time....

More confusion... should we take the Honda or the Tesla... it's an extra 50 miles to follow the superchargers and ~2 hours of charging that's $20... about the same as gas now since we can take a more direct route... and we save ~3 hours of extra travel and charging time....

 

[Ben W]

No... Blink never even got that far... they were only trying to cover expenses and expand the network. You need an external source of revenue to make a fast charging network viable.... like vehicle sales.

Blink had huge quality control issues, which Tesla seems to have avoided. Instead of having tens of thousands of solitary L2 chargers scattered all over the place, Tesla has many fewer L3 chargers in centralized locations; much easier to build out and maintain. As I said, some of the upfront capital costs will come out of vehicle sales no matter what. In a pay-per-use structure, some of the upfront costs may be borrowed against expected future revenues. (The razor model.) With the current unlimited structure, there are no future revenues; just future costs. SC stations in many out-of-the-way locations are considered an advertising expense, and are vastly underutilized; those will be able to handle quite a large Model 3 fleet without any additional capital.

 

[nwdiver]

In a pay-per-use structure, some of the upfront costs may be borrowed against expected future revenues. (The razor model.) With the current unlimited structure, there are no future revenues; just future costs. SC stations in many out-of-the-way locations are considered an advertising expense, and are vastly underutilized; those will be able to handle quite a large Model 3 fleet without any additional capital.

The assumption you're making that IMO is deeply flawed is the idea that a per minute fee would have a significant impact on network size requirements. You agree it would have no effect on transient use and local use is no where near 50%. It would be an enormous risk to shift to a time fee based model... if that were offered as an option and demand didn't decline as expected Tesla is stuck with an undersized network and no where near sufficient funds to correct that.

The financial functioning of the SC network would fit the definition of a Ponzi Scheme but in this case there is the potential for it to become self-sustaining in the future. The actual SCs require very little maintenance and there's demand response revenue that can be generated with the batteries. If Tesla sells 500k cars that's $1B to support the network. Investments made in solar and wind farms can ensure the future viability of the network. Power doesn't have to be generated at the SC... Wal-Mart builds wind farms then 'sells' the power to offset energy used at its stores... Tesla can do the same thing.

 

[Ben W]

- Blink failed because they targeted local users; no one is going to pay $0.50/kWh when it's $0.10/kWh at home... I agree 100% that pay per use would decrease local use... I always have... my point is that local use is maybe ~5% of SC use... as you've stated transient use is unlikely to be impacted. Most people already move on from SCs quickly... there's no reason to think that a per minute fee will significantly impact SC requirements... especially now that Tesla is hiring Valets to increase throughput during peak hours.

Local charging is MUCH higher than 5%. (Where did you get this number?) I would guess it's closer to 30%. When the Model 3 comes out, and a lot fewer owners have access to 220v charging at home, the percentage will almost certainly grow even further.

- You can't collect $200M upfront over time... I'm referring to it as 'upfront' for a reason... you have to have SC there to pay to use... that's going to cost $200M upfront... not over time.

You can't get around the upfront cost... you can't fund building a future network with that future network. The division between capital and maintenance costs is still largely an unknown. The SC network could operate like a ponzi scheme... which sounds bad but ponzi schemes only fail when they stop growing. Basically as long as Tesla keeps growing then there's no reason $2k/car can't support 'Free Long Distance charging'... once the network is mature a use fee will make sense.

You can borrow against future revenues. And you can still use some of the vehicle sale price towards the build-out. The smaller the network you expect to need, the less you have to pay/borrow up front.

Here's some amusing Pros and Cons to charging $10/hr...

Pro:
- Cars get cheaper by ~5%

Con:
- Lose 'Free Long-Distance charging' for marketing
- Confusion; 'how do you convert $/hr to $/gal @ 40mpg ?!?!?' -confused Tesla shopper
- 'I accidentally fell asleep charging and now I owe $50!... there was no one even there!' -Irate Tesla owner
- 'Why am I paying ~$1 more per charge?' - Model 3 70kWh owner whose car charges slower than a 90.
- More roadside assistance towing costs... people already cut charging close to same time....

More confusion... should we take the Honda or the Tesla... it's an extra 50 miles to follow the superchargers and ~2 hours of charging that's $20... about the same as gas now since we can take a more direct route... and we save ~3 hours of extra travel and charging time....

5% is a lot. Those who can't afford the 5% would buy an ICE instead. Those who can afford it will likely buy other high-profit options instead (e.g. premium audio). So by reducing the base SC-enabled price by 5%, Tesla gains a lot of customers and also gains a lot of high-profit option upgrades.

"Free Long-Distance charging" will still be effective marketing to differentiate the high-end Teslas from the midrange, and a reason to consider getting a Model S instead of a Model 3.

One-time confusion is solvable. Even at $0.25/minute, it's a lot cheaper than gas.

It's not necessarily a bad idea to waive the stall-hogging fee if no one is being blocked. Perhaps they should do this. Or else send you a text message or warning chime if you're stall-hogging, so you won't sleep through it. This is a very rare and contrived case though; how often have you actually fallen asleep at a supercharger?

The 70kWh driver paid $10k less for his car than the 90kWh driver. (More likely Model 3 will be 55kWh / 70kWh.) It takes a lot of $1's to make a dent in that difference. And for charging towards the bottom of the pack, both vehicles will probably be able to accept the same maximum charge rate. In most cases that will be enough to make it to the next charger. It's only toward the higher end of the pack that the charge rate starts to meaningfully differ.

The car is getting a lot smarter about preventing Brodering. You have to ignore a LOT of warnings to actually run out of juice. I don't expect there will be increased towing incidents.

 

[Ben W]

The assumption you're making that IMO is deeply flawed is the idea that a per minute fee would have a significant impact on network size requirements. You agree it would have no effect on transient use and local use is no where near 50%. It would be an enormous risk to shift to a time fee based model... if that were offered as an option and demand didn't decline as expected Tesla is stuck with an undersized network and no where near sufficient funds to correct that..

I never said it would have no effect on transient use. On the contrary, it will have a significant effect on how efficiently people use the chargers during long-distance travel, even if they travel the same number of miles. Instead of leaving the car plugged in for an hour while you have dinner, you'll charge for the 20 minutes you need and unplug. It's not clear that valets would have the discretion to stop your charging at 60%, even if there are people waiting. (And then some people might Broder their cars and blame the valets.) So even for long distance, the same number of chargers will be able to accommodate more cars per hour in a pay-per-use scheme.

Furthermore, many of the existing long-distance superchargers are vastly underutilized and will be able to accommodate a much larger Model 3 fleet without any increased capital costs. Most of the future build-out will be on high-traffic corridors or in cities, where local supercharging is much more of an issue.

Do you have any concrete figures on local charging %? Again, I expect this % to increase dramatically with the Model 3. The current S/X fleet is beside the point, because I expect it will stay on the unlimited model (which pays upfront for its own SC capital costs, etc.) Other than the fact that a significant part of the Model 3 SC infrastructure (the currently underutilized long-distance SC's) has already been paid for by Model S/X, the pay-per-use discussion is all about the Model 3.

 

[melindav]

IMO, the likelihood of local charging depends a lot more on SC location than mix of S/X vs 3 users. For me, the only SC within my typical route is 90 miles away in a small 'truckstop' style freeway exit town. I am sure it's use vastly different than a SC in southern California. It is serving the exact purpose as the system was designed for - long distance travel.
Assuming the SCs would be abused by Model 3 owners because a Model 3 owner can't afford to install a power source is ridiculous and insulting. If the 3 was a $10k car, maybe. But I would hope anyone considering spending $35-55k on a car can spring for an electrician to come out and add a plug AND cover the SC use fee if there is to be one.

 

[Jersey Shore Tom]

I just put together a spreadsheet, using Google Sheets, to try to estimate the lifetime cost to Tesla to provide Supercharging to Model 3 buyers on the same terms as current Model S&X owners. I came up with $787.60. I can't figure out how to upload it as an .xsxl so I am uploading it as a .pdf. If someone can give me instructions, I will upload it as a live spreadsheet so you can play with it.

I tried to cover all of the costs and use sourced numbers where I could find them. Feel free to critique my numbers, but please let us know if you have a source for your choice or if you simply disagree. Of course it ok to simply disagree, but the more sourced information we have, the more likely it is we are close to reality.