undefined | Better HN

0 pointsrobomartin2y ago0 comments

OK, I get where you are coming from.

>> Are you saying that's the case for the cars and loads we have today? >> Or, are you saying we have enough for a full transition to EV's?

> Pretty much yes to the second

We all wish that were true, sadly, it isn't the case at all. Our total power production capacity today is about 1,200 GW. A full transition to EV's requires (using clean energy) approximately three times that much power being built, and that's just for cars. You simply cannot schedule your way around that problem.

Here's a super-simple analogy to illustrate the problem:

You are the only waiter at a busy restaurant. It has only one cook. It has enough tables for both of you to be extremely busy pretty much all day. Sure, there are some periods of lower activity, yet, you never find yourselves sitting around doing nothing. you are busy.

The owner buys the building next door and expands the restaurant to now have 7 times more tables. He doesn't hire additional cooks, he doesn't install new order-taking stations or technology and you are still the only waiter.

The cook can't keep up. You can't keep up. The one-and-only cash register in the place can't keep up. The system worked well with 1/7th the patrons. It is now completely broken with seven times more. You need a lot more cooks (power generation) and a lot more waiters (grid, distribution) as well as all of the tools and technologies to support you (software, sensors, telemetry, maintenance, etc.)

You just can't schedule your way around a problem like that.

BTW, this is a super-simplified analogy to attempt to explain the most fundamental underlying problem. Don't try to slice and dice it, it isn't designed to be an accurate portrayal of all factors that go into the issue at hand.

In most areas of the US electrical systems were designed with assumptions made back in the 1950's. On average, it was thought, approximately 2 kW per home was needed. The wires, transformers, distribution, substations, etc. feeding neighborhoods, towns and cities evolved from these assumptions.

An electric car charging at home will demand six to seven times that much power. Which means that everything, from the home's electrical panel all the way to the electrical substation is now in peril.

If you want a really good explanation of this, I'd suggest you read chapter 3 of this book that was just released by the Institute of Electrical and Electronic Engineers:

https://spectrum.ieee.org/files/52329/The%20EV%20Transition....

If you go to page 14, you can get the real-life story that I used to attempt to create an analogy. The author describes precisely this issue in the city of Palo Alto, CA. Conclusion: Palo Alto needs to almost rip-up and upgrade its entire infrastructure to be able to service full electrification.

Electric stoves and other electrification only make this more difficult.

> should be offering discounts for not doing so and allowing simple overrides at the charger

My take on this is that people will do what they have to do given their circumstances. People have very random daily schedules. Some work at night, some have work and college, others might have a couple of jobs, after school activities, etc. There's a lot of variability built into behavior. It's like herding cats. In the end, I don't think behavior will make a significant impact.

> I'm also super excited to see what happens with vehicle to load in the coming years

While I understand the idea Vehicle to Grid (V2G) from an engineering perspective. I can't understand why anyone would ever do that.

First, the most fundamental problem: Home electrical systems and the distribution grid are not ready for this. You would have to replace millions of home electrical panels and power meters just to get started. I did just that when I built my solar array, it was not cheap. Imagine buying an electric car and having to spend $10K to $20K to redo your home system to support it. In addition to this, the transformers and other components are not ready to deal with V2G. The control systems, data and telemetry at most US power systems is in the stone age.

Even if all of this was magically resolved at zero cost, there are fundamental questions about V2G that, I propose, will lead people to say "no, thanks".

First: Battery packs have limited charge/discharge cycles. If you use your car for V2G, your battery will fail much earlier than if you just use it to drive. Why pays for that? The warranty will not cover it, I can assure you of that. Also, think of the waste and recycling load that would impose.

Second: I just spent money to fill-up my battery. And now I am going to let the grid drain it? Are they going to pay me? OK. They have to pay me enough for me to recoup my investment and more than justify having to spend tens of thousands of dollars when the battery fails. That power will be insanely expensive. Who is going to buy it? Energy rates are already ridiculously high. This makes no sense to me.

Third: Range anxiety with electrics is a thing today and will be a thing in the future. Why drain your battery to the grid? This makes no sense from that perspective. It's like saying to your neighbors "Hey, I just filled-up my gasoline tank. While my car is parked, feel free to siphon off up to 50% of my tank.". Nobody is going to do that.

Experts who have looked at this in great detail seem to agree with the idea that this isn't going to happen for a very long time, if ever. There are too many problems with the practical realization of the concept. This is common in engineering, an idea might sound great, until you actually go face the realities of having to build it.

Again, if you are interested in this, I urge you to have a look at that book from the IEEE. It's only 49 pages long. If you are reasonably technical it will be easy to understand. Lots of good references tood.

Grid issues abound. On page 33 of that book there's a map showing the status of the grid in CA. The legend below it reads:

    Red lines indicate areas where the grid cannot
    accommodate additional load without any thermal or
    voltage violations. Grey hatched areas indicate regions
    where gaps in utility grid data exist. Colored lines,
    keyed in the legend, indicate the available circuit
    capacity in megawatts

It's worth a look.

0 comments

ender3413412y ago

Thanks for the detailed response! Will definitely give that IEEE pdf a read.

> First, the most fundamental problem: Home electrical systems and the distribution grid are not ready for this.

I'm not as familiar with code for new buildings in CA, I know that they require solar on them do they not do solar to grid as part of that? That wasn't something I'd really thought of, I'd assumed most of the cost for that usually comes from retrofitting more so than if it's during the initial setup but haven't actually looked to verify that.

> First: Battery packs have limited charge/discharge cycles. If you use your car for V2G, your battery will fail much earlier than if you just use it to drive. Why pays for that? The warranty will not cover it, I can assure you of that. Also, think of the waste and recycling load that would impose.

It depends pretty heavily on the manufacturer, but newer batteries on higher end cars have pretty amazing cycle lifetimes, watching car enthusiests on youtube it's not uncommon for people to buy used tesla batteries from totaled cars for custom conversions or custom home battery systems.

> Second: I just spent money to fill-up my battery. And now I am going to let the grid drain it? Are they going to pay me? OK. They have to pay me enough for me to recoup my investment and more than justify having to spend tens of thousands of dollars when the battery fails. That power will be insanely expensive. Who is going to buy it? Energy rates are already ridiculously high. This makes no sense to me.

I'd fully expect to be getting that cost back + some small amount for the inconvenience back as credits on my bill. (I would mostly be expecting this for handling particularly bad peaks and not so much an everyday thing)

> Third: Range anxiety with electrics is a thing today and will be a thing in the future. Why drain your battery to the grid? This makes no sense from that perspective. It's like saying to your neighbors "Hey, I just filled-up my gasoline tank. While my car is parked, feel free to siphon off up to 50% of my tank.". Nobody is going to do that.

I'm not sure if you have an EV but for me range anxiety went away pretty quickly after buying mine, it's pretty much only for long range road trips that I have to pay any attention to range at all, though convincing people of that before they own an EV is pretty tricky (and there are people who need to drive hundreds of miles a day that don't work great with current EVs)

> My take on this is that people will do what they have to do given their circumstances. People have very random daily schedules. Some work at night, some have work and college, others might have a couple of jobs, after school activities, etc. There's a lot of variability built into behavior. It's like herding cats. In the end, I don't think behavior will make a significant impact.

Those people that have different daily schedules probably actually help as they're less likely to be charging during those peak evening hours.

From what I've seen the systems are pretty automated and there's a lot less for people to manage than you might think, it's more like setting up a thermostat where there's some initial work to setup a schedule but then it's just plug your car in when you get home and leave it plugged in and they can schedule the load so it's ready when you plan to leave and then if you have need to leave earlier than usual it's not hard to override it (though I do wish more EVSEs had direct buttons on them to control that type of thing).

robomartinOP2y ago

I think the IEEE document covers quite a bit of your questions/comments.

My neighborhood is barely 25 years old. Meaning, it was dirt when I bought my house and watched as they built it.

Still, the grid wasn't ready. The house was not ready.

When I put in our 13 kW solar array we had to rip out the original breaker panel and replace it with one of higher capacity. Not because of the size of the array (which is larger than pretty much all installations around me by at least a factor of 2). No, it was simply because, 25 years ago, when plans were drawn for this entire development, the Building Department did not have any requirements on the books for the future addition of solar. The meter had to be replaced as well.

Just to make sure it is clear. What this mean was tearing-up the stucco all around the existing panel, making a larger hole, rewiring the new panel and meter and redoing the stucco. Of course, stucco is basically concrete, which means that side of the house lost its color, which required repainting the entire exterior of the house because the faded stucco color was not going to be easy to match at all. I don't remember the cost of that part of the job. I was not allowed to do it myself (I did the rest of the solar installation). I think it was in the order of $20K, maybe more.

I don't know what code might look like today for new developments. I would hope they include planning for solar, which should include power distribution, transformers, meters, breaker panel and other requirements.

Regarding battery issues in V2G applications, well, the IEEE document covers them well. It boils down to faster degradation due to lots more cycling, warranty issues (the EV manufacturer covers miles driven, not V2G cycling), etc.

On the subject of paying for the power you contribute. Well, we'll have to wait and see. I have no clue how they might plan to manage this. If I am going to use an EV for V2G, the utility better pay me a lot more than what it will cost to replace the battery pack. Which means that the users of my power will likely have to pay quite a bit more per kWh than today.

I generate a lot of excess energy. The way they pay you back for excess solar today is absolutely laughable. In fact, the fear now is that homeowners will have no incentive to install solar because what is known as "NEM 3.0" metering, well, basically screws them even worse than those of us who already have solar.

Here are the basics:

https://www.solar.com/learn/nem-3-0-proposal-and-impacts-for...

j / k navigate · click thread line to collapse