Waiting on Tesla Battery Day News Today

[pj.mask]

The solar panels help, but are they enough?

I counted 12x24 solar panels for each of the 2 canopies. They look like the common-sized 320W panels. That's 12x24x2x320 = 184 kW. Good for supercharging 2 cars at 100 kW a piece in the midday sun, while there are 24 charging stalls.

But hey, in one of the photos, one of the canopies is completely shaded by a nearby building! Man, that's a bad location.

But let's suppose the panels were out in sunlight all day, if they built it out of town. Sunlight data from nrel.gov says that in Las Vegas, these nearly flat-mounted panels will produce as much as 34,600 kWh in the month of June, and as low as 11,700 kWh in the month of December. It could have been higher in December, if the panels were tilted up to match the low sun angle, but you can do that only with a solar farm, not a shade canopy.

If each refueling EV sucks up to fill, say 50% of its 70kWh battery, then these panels could serve only 11 cars per day in December. In June, 33 cars per day.

Yet, Tesla claims to be able to serve 1,500 cars per day with this V3 station. Where's the juice coming from?

I think the key here is the battery storage to smooth out power grid demand - not the solar. I didn’t claim and the article didn’t suggest anything about solar powering the super chargers - not sure where you got that impression.

This is the same plan for California as a whole - planning on battery storage. It sounds like the fires are often cause by overloading lines that get hot and start to hang down low causing fires. If they balance those loads with battery storage, maybe that won’t be an issue?

Plenty of opportunity to add solar power to the grid. Also want to note Tesla powerwall advertises 90% efficiency round trip so let’s not keep using the 80% numbers.

Tesla said car to grid won’t happen. Best solution is lower cost iron batteries - not connected EV.

 

[FIREarly]

Tesla gives free roadside assistance for 4 years/50,000 miles. If driving the other EV's your out of luck to my knowledge.

Most Tesla owners plan to drive their car beyond 50,000 miles.
1. How much does Tesla roadside assitance cost after 4 years/50,000 miles?
2. Does AAA have an EV plan?

 

[ERD50]

.... Also want to note Tesla powerwall advertises 90% efficiency round trip so let’s not keep using the 80% numbers. ...

Yes, but we were talking EVs as storage. I'm assuming they aren't as optimized (more constraints on size/weight for being mobile) as a PowerWall, I am open to better figures if you have them, but the point stands that time-shifting actually increases the amount of kWh that must be produced. Also, the PowerWall #'s appear to be a best case spec under specific conditions:

The Tesla Powerwall has a 92.5% round-trip efficiency when charged or discharged by a 400–450 V system at 2 kW with a temperature of 77 °F (25 °C) when the product is brand new.

Home chargers won't have 400V available, and lower voltage lead to higher IR loss, garages aren't kept at 77F, and batteries age, etc.

.... Tesla said car to grid won’t happen. Best solution is lower cost iron batteries - not connected EV.

Exactly. But I was responding to Zathras claim (emph mine) that "the EVs can actually help them with their grid issues".

It isn't a question of V2G not being supported by Tesla, it is more a question of not being supported by utilities. ...

Can you provide some background on that? The utilities do allow products like the Powerwall to be used to shift night/day energy, correct? So why wouldn't they allow it for EVs? I suspect this is more a chicken/egg situation than the utilities not supporting it - if no EVs are currently set up for it, how can the utilities support it?

And if it would help the grid as you claim, it would seem the utilities would be all for it. But wacky things happen, so I look forward to some more information regarding this.

... However, ignoring V2G, an intelligent use of charging of EVs can help the grid.
Just having smart chargers that can vary the rate of charging overnight to smooth out surges will help ease stress on the grid. ....

That's a bit of a tortured twist. EVs will put more stress (at least on the production side) on the grid, period. Smart charging can smooth that out to a degree, but it is smoothing out a problem that EVs created. It isn't a plus for EVs, it's just a slightly smaller negative.

... In other parts of the country we already use on demand for heating water heaters. This is much more basic, but does aid the utilities. Expanding on this idea can help. ...

Good example, but totally irrelevant to the EV issue. Those water heaters are already in operation, so using a smart grid or simple timer to heat/store the hot water during non-peak times does help reduce stress on the grid (with maybe some very slight increased loss for holding the water a bit hotter than needed at times - very minor with today's well insulated electric water heaters).

Add an EV to the grid and it still increases overall demand significantly, even with smart charging. It's just not the same as the water heater example.

So I still say your statement that "the EVs can actually help them with their grid issues" is totally false.

-ERD50

 

[NW-Bound]

I think the key here is the battery storage to smooth out power grid demand - not the solar. I didn’t claim and the article didn’t suggest anything about solar powering the super chargers - not sure where you got that impression.

This is the same plan for California as a whole - planning on battery storage. It sounds like the fires are often cause by overloading lines that get hot and start to hang down low causing fires. If they balance those loads with battery storage, maybe that won’t be an issue?

OK, fair enough.

I wholeheartedly agree that energy storage is sorely needed. The article said that both solar generation and storage batteries will be included in V3 charging stations, but I do not recall seeing any numbers.

There were no photos of the storage batteries for me to even venture a guess. And seeing the PV panels, I could not help computing how much those panels would generate.

But we can roughly estimate how much storage capacity is needed. Back at the claim of 1,500 EVs/day being served, and I assume that each EV fills up 1/2 of its 70-kW battery. That's 52,500 kWh/day.

How much storage is needed to average the juice flow over a 24-hr period? 1/2 or 1/3 of that 52,500 kWh? I don't know.

But let's look at what is available now. Tesla is selling Powerpack units that store 232 kWh a piece for $125,793. To store 1/3 of daily usage, we are talking 75 Powerpack units at a cost of $9.4 million.

The problem has a technical solution. Is it affordable? I don't know either.

PS. Note that each of the Powerpack units stores about 3x the 70-kWh battery of the EV. The price is more than 2x the EV price. Powerpacks are not cheap. They are still around $500/kWh.

 

[MRG]

OK, fair enough.

I wholeheartedly agree that energy storage is sorely needed. The article said that both solar generation and storage batteries will be included in V3 charging stations, but I do not recall seeing any numbers.

There were no photos of the storage batteries for me to even venture a guess. And seeing the PV panels, I could not help computing how much those panels would generate.

But we can roughly estimate how much storage capacity is needed. Back at the claim of 1,500 EVs/day being served, and I assume that each EV fills up 1/2 of its 70-kW battery. That's 52,500 kWh/day.

How much storage is needed to average the juice flow over a 24-hr period? 1/2 or 1/3 of that 52,500 kWh? I don't know.

But let's look at what is available now. Tesla is selling Powerpack units that store 232 kWh a piece for $125,793. To store 1/3 of daily usage, we are talking 75 Powerpack units at a cost of $9.4 million.

The problem has a technical solution. Is it affordable? I don't know either.

PS. Note that each of the Powerpack units stores about 3x the 70-kWh battery of the EV. The price is more than 2x the EV price. Powerpacks are not cheap. They are still around $500/kWh.

Tesla selling a power wall for 125k isn't their cost. While I agree storage is still expensive that's today.

I spent some of my career around data storage. I distinctly remember being constrained and doing unnatural things because of the cost of enterprise disk.
Thirty years ago we were looking at 50k per gigabyte for mainframe DASD, er disk. Today I have a TB of solid state in my vehicle for tunes and sentry cameras. Extrapolate that.[emoji854]

 

[ERD50]

Tesla selling a power wall for 125k isn't their cost. While I agree storage is still expensive that's today.

I spent some of my career around data storage. I distinctly remember being constrained and doing unnatural things because of the cost of enterprise disk.
Thirty years ago we were looking at 50k per gigabyte for mainframe DASD, er disk. Today I have a TB of solid state in my vehicle for tunes and sentry cameras. Extrapolate that.[emoji854]

It's been pointed out many times, solar panels, batteries, motors are not on the same trajectory as computers and disk drives. The comparison has no merit, completely different things.

From the National Academy of Sciences:

https://www.pnas.org/content/110/14/5273

Sadly, such batteries do not exist. There is no Moore’s Law for batteries. The reason there is a Moore’s Law for computer processors is that electrons are small and they do not take up space on a chip. Chip performance is limited by the lithography technology used to fabricate the chips; as lithography improves ever smaller features can be made on processors. Batteries are not like this. Ions, which transfer charge in batteries, are large, and they take up space, as do anodes, cathodes, and electrolytes. ...

Batteries have been around longer than disk drives. Why aren't we there yet?

-ERD50

 

[NW-Bound]

For my DIY solar system, I have been buying a lot of electronic parts directly from Chinese vendors, via Banggood and AliExpress. At first, I was shocked to find semiconductor prices so cheap compared to mechanical parts. A 32-bit ARM microcontroller costs less than $1. An optocoupler IC, a MOSFET, they all cost a couple of pennies, and at the low-volume retail level to boot. Yet, a copper terminal lug may cost a dime or a quarter. You would think a solid-state device requiring expensive production capital equipment should cost more than a few pennies.

A bit of reflection gave me the answer. A terminal lug requires a sizeable chunk of copper, and copper is expensive now. The solid-state devices require expensive capital equipment, but once that machine is bought they can crank out zillions of the devices, each requiring so little material because they are so small, plus it's all silicon or sand.

And so, things that require a big chunk of material for their operation cannot drop in prices while increasing their performance like electronic chips. A small wire can never carry more current than a big cable.

And batteries cannot be shrunken the same way as solid-state devices. If that were true, we would eventually pack the power of an EV battery into a D-size cell. No, make it a hearing aid cell. All of the world problems would be solved.

 

[ERD50]

... And batteries cannot be shrunken the same way as solid-state devices. If that were true, we would eventually pack the power of an EV battery into a D-size cell. No, make it a hearing aid cell. All of the world problems would be solved.

Well, not *all* our problems. We still need to produce the electricity to fill those batteries up with energy. ( I know you know this, I'm just pointing it out for some others that may forget about the production side of all this).

-ERD50

 

[NW-Bound]

Having had some hand-on experiences with my DIY solar system, I will say that solar power production is a lot easier problem to solve than the energy storage problem.

In my corner of the world, a common 320W solar panel will produce 577 kWh/year, using actual solar data from nrel.gov. This amount of electric power is worth $70 with the average price of 12c/kWh that I am paying. How much does this panel cost? At the retail level, for me, around $120. I can get less efficient panels, which are a lot cheaper in terms of $/watt, but they will take more room which I do not have.

And so, in my simplistic viewpoint, I can generate a lot of power using panels that will pay for themselves in a matter of a couple of years. But I would need more batteries to store it for night use, and for cloudy and rainy days. At this point, the battery cost dwarfs the cost of the panels. I suspect the same enonomics is true with grid-scale production by electric utilities.

And speaking of batteries, do people who watch Tesla Battery Day presentation remember what Musk said about how much battery would be needed? If not, I will present a summary shortly.

 

[pj.mask]

I would say better ones are on a chemistry/metallurgical time frame. Maybe some of that in manufacturing as well.

I don’t really understand numbers on the batteries required for supercharging - I’m sure Tesla is able to calculate grid supply and battery needs. I don’t think it is that high and they have commercial and consumer variants. It probably is just peak leveling ie only supply fraction of load during peak.

As to car efficiency, I’d imagine the cars should be as or more efficient (assuming you can run the inverter ‘backwards’ because they charge/use the battery a lot more. One person got ~95% using wall meter vs computer battery kWh. I wonder if the tab-less design improves efficiency?

 

[NW-Bound]

What Musk said about battery production is this:

1) In 2019, about 0.1 TWh of battery was produced for EVs. We need to go to 10 TWh/year, and sustain it for 15 years, in order to transition to all EVs including transportation and commercial vehicles. That's an increase of production of 100x.

2) In 2019, about 0.006 TWh of battery was produced for grid energy storage. In order to go all RE, first they said another 10 TWh/year of battery production is needed. Then, Musk said to allow for the entire world, it would be 20 to 25 TWh/year, sustained for 15 to 25 years.


The presentation did not go deeply into how the above estimates were made, obviously because that was outside the scope of the talk. However, I could make a rough check of the battery requirement for EVs.

World production of vehicles is about 92 million/year. Divide that into 10 TWh/year, and we have the average EV battery of 108 kWh. That's larger than the current average EV car battery, but then they were talking about including commercial vehicles and not just personal cars. Vehicles last about 15 years on the average, and that's why they said the production period would be 15 years to convert the world fleet over. It all sounds reasonable to me.

About the requirement for other energy usage, I think it is reasonable based on what I looked up earlier regarding all energy usage, industrial as well as residential, and how that compares to vehicle energy usage.

In short, according to Musk, an increase of battery production of 300x is needed, and I find that factor reasonable.

 

[Koolau]

Batteries have been around longer than disk drives. Why aren't we there yet?

-ERD50

Heh, heh, maybe more to the point, batteries have been around longer than the ICE! We're still making incremental improvements to the ICE, but nothing dramatic - and nothing dramatic on the horizon that I've heard of.

My gut tells me, it's the same with batteries. We might be able to improve "losses" (in and out) of batteries but I don't see potential dramatic improvements in storage density. There certainly seem some opportunities to bring costs down. Maybe we can increase the rate of charging or "fullness" of charge over time.

But, there are only so many combinations of anodic and cathodic chemistries available. I'm SWAGing here that MOST of them have been evaluated, at least on paper. Yes, you can make dramatic improvements of transistors 'cause when I was a kid, you got 6 in a good portable radio. Because of miniaturization, we get millions of the little buggers on a square-inch chip! I don't see ANY opportunity to make a similar improvement from the old 2 lb dry cells my Gramps used to attach to his phone to boost the signal from "Central." Yeah, they're much "better" now, but we're never going from 2 lbs to 2 grams with the same storage potential. The chemistry just isn't there.

By the way, I'm not an electrochemist though I used to (have to) play one at w*rk. Don't get me started on the 'joys' of polarography. Well, playing with the mercury was kinda fun. SO, YMMV.

 

[ERD50]

Heh, heh, maybe more to the point, batteries have been around longer than the ICE! We're still making incremental improvements to the ICE, but nothing dramatic - and nothing dramatic on the horizon that I've heard of.

My gut tells me, it's the same with batteries. ....

But, there are only so many combinations of anodic and cathodic chemistries available. I'm SWAGing here that MOST of them have been evaluated, at least on paper. ....

There have been some interesting ideas for ICE, especially as a hybrid, as that can allow the ICE to work in a "sweet spot". But it is incremental, and I expect Lithium improvements will outpace them - but you still need to produce the electrical energy to "fill" those batteries, regardless how small/light/cheap the batteries get.

A few years back, a poster here (a ChemE) took a SWAG at that. IIRC, you can make some theoretical assumptions based on the Periodic Table. But then you have to allow for something to hold it together, those combinations might not hold up over time or be safe. But even theoretical, it wasn't totally amazing, I don't think. It's not like we will see an EV with a lifetime charge in a battery the size of a micro-SD card.

But clearly, battery tech will make some good strides, increasing range and lower costs. More people will be buying EVs in the future. But we still need to generate that electricity.

-ERD50

 

[NW-Bound]

There will be improvements on battery technology, but not at an exponential rate like Moore's Law. And that's why I watched Musk's presentation to see what was being done.

I like that he did not promise huge improvements based on some way-out-there technologies. He was doing some tweaks on existing technology, expecting some improvements in the order of 16%, but hopefully a lot more on the cost cutting. Even there, the dry electrode process was still iffy, and Musk said a lot of work remained to be done.

And finally, he acknowledged the amount of battery that would be needed to go green. It's huge, as I summarized in an earlier post.

 

[Time2]

I don't see how we could have 30% electric vehicles and not have grid problems.

The Pacific Northwest Labs study looked at all of the capacity that’s sitting unused in the grid, a grid built for "...the one hour of the year when demand for electricity peaks and for every other hour of the year goes underutilized,"
I see sometime in late December, in the North East around 7:30, everyone has driven home through the ice and cold, plugged in their chargers and the furnaces have come on and the brownout starts.

 

[NW-Bound]

I don't see how we could have 30% electric vehicles and not have grid problems.

The Pacific Northwest Labs study looked at all of the capacity that’s sitting unused in the grid, a grid built for "...the one hour of the year when demand for electricity peaks and for every other hour of the year goes underutilized,"
I see sometime in late December, in the North East around 7:30, everyone has driven home through the ice and cold, plugged in their chargers and the furnaces have come on and the brownout starts.

Enter the massive bank of batteries that are distributed throughout the region, charged up during the day and ready to be discharged to provide heating and lighting to the shivering northeasterners at night, and juice for their EVs too so they can go to work in the morning.

Yes, lots and lots of batteries will have to be part of the solution, in order to store solar energy collected down where I live, the sunny Southwest where we have maybe, oh, 5 cloudy days each year. And of course that means lots and lots of transmission lines to bring the juice up north.

In the long run, the earth will inevitably run out of fossil fuel, and RE will be all that humanity has. Well, maybe nuclear power will be fashionable again when people are cold in the winter and hot in the summer.

I like RE a lot, but don't see anyone coming up with a proposal on how to pay for all that infrastructure. California is suffering from PSPS (Public Safety Power Shutoff) because of sagging power lines as we speak. And California is the richest state, and can't afford better transmission lines.

 

[NW-Bound]

People talked earlier about round-trip efficiency of lithium energy storage. I think it is appropriate to share my own experience here.

I am adding a new block of battery storage of 12 kWh to my solar system. It is divided into 28 modules, and I carefully measure each module capacity so that I can later check their degradation with age and usage.

I have a voltage and current logger which records the I and V every 10 seconds during charging and discharging. It gives me a data file to import into a spreadsheet for me to compute the power (P = V*I). The modules are 120 Ah nominal, and I use a relatively low current of 10 A, so that the charging and discharging take 12 hours each.

My spreadsheet showed the round trip power loss being 5.7%. That's decent I thought, but what could be expected with this type of cells? Then, just out of curiosity, I measured the voltage loss due to the wiring of the setup. I had 7 milliohms of resistance that was not accounted for. That alone caused a loss of 3.5%. The cells themselves are better than 2.2% at the C/12 rate.

But there's more. I adopted Tesla's method of attaching each cell to the bus using small wires that serve as fuses at the cell level. That caused roughly a 1% loss. The intrinsic cell loss could be down to 1%. Lithium cells are very, very good!

With a much better professional cell assembly, along with top-notch electronics for the charging circuit and the inverter, I think total system efficiency can easily get to 10% loss or better for a round trip.

But even if the loss is 20%, it is still quite good. Pumped hydro power storage has 20-30% loss by comparison.

California is in dire need of more batteries to store the excess solar energy it generates in early spring early in the day, and has to pay Arizona and other neighbor states to take some off its grid. Yes, California has to pay to give away power some time. Yet, it still has to import around 30% net over a year period. When you have solar power out the wazoo some time, you need a place to store it. You do not care if it is 50% efficient if it is cheap. Lithium batteries are very good, but they are still too costly.

If you cannot afford enough batteries to store for use at night, when will you be able to store for a rainy day or two?

I shared this article in LA Times in a past thread, and it is worth repeating here.

https://www.latimes.com/projects/la-fi-electricity-solar/

On 14 days during March, Arizona utilities got a gift from California: free solar power.

Well, actually better than free. California produced so much solar power on those days that it paid Arizona to take excess electricity its residents weren’t using to avoid overloading its own power lines.

It happened on eight days in January and nine in February as well. All told, those transactions helped save Arizona electricity customers millions of dollars this year, though grid operators declined to say exactly how much. And California also has paid other states to take power.