Hacker News Reader

Comment by torginus | Hacker News Reader

Personally speaking, having just bought an Ioniq 5 and installing solar at home what I see as the near future improvement is adding V2L functionality, which I can hook up to the generator input of my solar inverter, essentially adding another 60kWh buffer to my grid storage.

Considering how expensive residential batteries are and how quickly EVs depreciate, I think soon it'll be cheaper to get a used EV as a cheap source of cells that accidentally happens to be able to drive itself around.

Imo V2G, and V2H is unnecessary and add too much complication, I think for the future, solar inverters already have the necessary hardware and certifications to be able to take power and safely connect to the grid - something that requires different hardware and standards compliance in basically every country (yes even within the EU).

View on HN

Residential batteries are not that expensive anymore, at least not all of them. That's a misconception I also held until a few years ago ;-)

My first 14.3 kWh pack cost about 2800$ DDP from China, delivered 03/2023. For that one I did calculate how long it took for amortization, which I projected at about 5 years.

The second, identical pack was delivered 08/2024 and cost 2000$ DDP. Since we got an EV that's drawing about 14kWh per day, I didn't bother doing the math and just ordered it.

These are 280Ah 16S 51.6V packs, based on the EVE LF280K. In an enclosure, with a BMS (Seplos, 200A) and a dedicated balancer. They are good for 6000 cycles at 140A or less [each]. Mind these were both part of small bulk orders - I think each time we ordered 6 to 8 of these, which reduced shipping costs.

Today you can already get 51.2 V, 16 kWh batteries complete with balancer and BMS from a European supplier for 1200 € (1400 US$): https://www.nkon.nl/novat/nkon-ess-eco-s-51-2v-16-1kwh-thuis...

And the nice thing is, prices will only get lower. They also have a 32.15kWh system for €3500,- which is insane.

Unfortunately, almost all three-phase [1] on-grid inverters that are on the market, especially hybrid inverters, only support batteries with much higher voltages, like 150 V or even more.

[1] Three-phase wiring in homes seems to be very rare in the US, but is extremely common in Europe.

The trick is to get an inverter with a 48V battery interface, that's not locked down to a specific manufacturer.

But you're right, most installers do high voltage batteries.

That looks quite a bit cheaper than the pricing I'm used to seeing, like less than half - not saying they're bad, but probably there's a DIY and risk factor involved compared to buying a more established brand like BYD or Huawei.

I'm just pointing out this is way lower than typical pricing

Maybe, but I think the premium charged by the main manufacturer's is unjustifiably high. Those battery boxes are relatively dumb: besides the cells, they just need some voltage and temperature sensors for monitoring, some power electronics for balancing, a microprocessor, and an enclosure with connectors. Unlike a grid-tied inverter, this is a really simple system, and there's no way a 500–700% price premium over the cells is reasonable.

My new batteries were about 250EUR/kWh - my 10KWh unit cost 2500 EUR - scaling it up to a decent used 5 year old EV price - you can have one for 15k with 60+ kWh batteries, so I'd say it's at a very similar price.

You wanna check how much V2G inverters cost first.

If you don't work from home it makes 0 sense either.

IMO we are long way from proving V2G usefulness. Micromanaging your car SOC every night to save $2 is nuts.

Do you mind citing providers/product names?

I would take that further and say that residential solar without batteries has been proven to be a bad solution. Solar with batteries allows utilities and consumers to schedule when power can be sent to the grid. California utilities consider solar without batteries a PITA, and incentive structures have changed to reflect that shift in policy.

https://www.canarymedia.com/articles/distributed-energy-reso...

https://enphase.com/blog/homeowners/understanding-nem-30-and...

> Imo V2G, and V2H is unnecessary and add too much complication

I believe that's more a function of auto makers (and charger cos) not trying to do much about it (and grid cos not caring), than a technical issue. The benefits (especially if V2G) are quite significant.

I doubt any car owner will say no to earning revenue from something that costs them almost nothing. The problem is more of "how do we get there at scale". (Disclaimer, I studied this topic in my thesis.)

EVs are probably not going to depreciate as much in the future. The depreciation mostly happened because new electric cars have become cheaper.

As an example, let's say a 2 year old car is only worth 80% of an identical brand new car. The old car was bought for $50k at new, leading to a expected depreciated value of $40k. But since manufacturing has become more efficient a brand new car of the same model can be had for the same $40k. Nobody would be willing to buy the 2 year old car at the same price. You'd probably have to charge only $32k [0]. But then it looks to you as if it has depreciated 36%.

The question is how much new electric cars will fall in price in the future. And if they continue to fall, at some point the dollar value of the depreciation will be too small to care about.[1]

0: $32k = 0.8 * $40k

1: E.g. if a new car is $10k, $3600 in depreciation over two years is annoying, but not a big deal.

In the US, V2L limits your ability to output power from the car to about 1500 W. It's not going to power your house as more than a stopgap, even if you do have supplementary house batteries. V2H/V2G justify their complexity by solving that problem, along with all the ancillary grid benefits.

Not sure if that's the case - however doing V2L requires the manufacturer to add an inverter to the car, and making that powerful probably adds extra cost most customers wouldn't pay. TI just looked it up and my Ioniq can only do about 2kW sustained - but since this charges the house battery, that's enough - idle load is just a couple hundred watts.

If the car charges the house battery, what charges the car?

If you have solar panels or time-of-use electrical rates, you charge the car when power is cheap/free, and spend stored power when the grid costs are high. During a protracted outage, maybe you drive the car to a fast charger.

You pointed out a significant limitation of my current setup - right now there are 2 plugs - one for discharging the car through a proprietary manufacturer's V2L adapter, and one for charging.

I'm planning to make a 'box' that can switch between the 2 functionalities on the same cable.

The whole setup is a bit clunky as it is right, now, but I'm kinda more surprised that it works at all, and how well the fundamentals work.

This whole thing was more of an experiment in 'no way you can do this' to actually doing it, but I think this is HUGE, and will transform the way people think about electric cars.

A typical house averages less than 1500W. And most of the higher usage overlaps the sun being out. So if you have supplemental house batteries to handle bursts then 1500W of V2L can go a very long way.

the average hides a lot of information. the largest peak load is often an electric stove, which is regularly greater than 1,500 kW.

Also, this idea that higher usage overlap with the sun being out is laughably wrong. Solar noon is between 11 AM and 2 PM. Very few people are home at that time. There is a reason that peak grid demand in almost every country is in the early evening.

> the largest peak load is often an electric stove, which is regularly greater than 1,500 kW.

Does that change anything about what I said? This is specifically about "if you do have supplementary house batteries".

> Also, this idea that higher usage overlap with the sun being out is laughably wrong.

The reason we have the duck curve is that insolation and demand largely overlap (especially when we're talking about the worst case part of the summer), but then for part of the evening they really don't overlap.

The peak use is evening, but there's a significant ramp up when the sun rises and the whole day is much higher than night. https://ars.els-cdn.com/content/image/1-s2.0-S03062619173137... (This isn't the US but finding household graphs in particular is annoying, and most of the US has more summer heat than denmark)

Anyway evening is one of those bursts where you use the supplementary battery to handle the rest of the load. Even 10% of the car's capacity, 6kWh, could cover almost all use above 1500W.

Everything you describe is true only in some places, likely California. In much of the rest of the world, electricity demand peaks in the evening, when the sun is low in the sky and continues well into the evening, when the sun isn’t out. Notice how even the Wikipedia page about the duck curve lists mainly California. Even in Australia and the UK, daylight hours and electricity demand mostly do not overlap.

>> The peak use is evening

> In much of the rest of the world, electricity demand peaks in the evening

Huh?

And again my chart showing that the entire day has much higher use than at night, then an extra high peak in evening, was Denmark.

> And most most of the higher usage overlaps the sun being out.

Aren’t most people at work / school when the sun is out?

Yes but high electricity use days correlate with air conditioning, and most people don't turn that off in the middle of the day.

If you're not worrying about A/C then 1.5kW goes an extra long way. Outside of cooking you'll rarely exceed it.

Very interesting - I did not know this was possible. A few questions:

1. Does the solar inverter do away with the need for a V2G or V2H unit?

2. What are the limitations vs a dedicated V2G/H unit?

3. Is generator input on your solar inverter a common feature across inverters?

1. It does. The only issue is that the car can only output about 2kW sustained (this is a model limitation). That's fine since I have batteries in the house.

2. Tbh not super familiar with V2G/V2H, other than it being super expensive for both the wall box and the car (only high end models tend to support it)/

3. No idea, but it's not a high end feature, I wouldn't count on any inverter to just have it, but if you're looking to buy one that does, I don't think you'll be breaking the bank.

Imo the future is for solar inverters to offer a dedicated DC car charger port, as once again all the hardware is already in there.

Thanks for the answers. I used to work for a EV smart charging company (Kaluza) that ran a V2G trial. V2G was financial success for the users, but I always thought the wall box was a potential blocker. I don't think the 2kW output is a big issue as the customer could still reduce there load when required, but the elimination of a wall box makes onboarding much easier.

As long as the inverter can also provide charging this definitely has some potential.

Reply on news.ycombinator.com