Hacker News Reader

Comment by mullingitover | Hacker News Reader

> Well, that battery might cost more than the entire house. And for what purpose? So I can disconnect from the electric grid for a week, and then reconnect again when the battery has run out? Or to power it with expensive solar panels?

I can't figure out if he's being willfully obtuse here.

It's to allow you to get off the grid entirely. If you generate enough solar to cover 100% of your usage, right now you're still slapped with a grid connection fee. The battery covers you overnight and makes grid connection totally unnecessary.

Oh but what about charging your car? Battery swap stations.

View on HN

He actually makes two assumptions both of which hurt his point, the first is that cost of the battery is that of a house (given that an 85kW Tesla costs $90K and you don't need the car with it, we'll guess the battery is less than $90K) so lets call it $75K. And the second is that stabilizing energy costs doesn't change anything else.

If your house is nominally electric (using gas for things which gas is really good at, making heat) and your gasoline bill is non-existent because your vehicle is electric. It means you can cost out power with a depreciation/replacement cost model which gives you much better forecasting ability. One of the reasons given for taking 'energy costs' out of the whole inflation/cpi metric was due to its volatility.

I tend to agree with his thesis that Apple will not buy Tesla but I don't think he makes a good argument against the battery claim.

Author here: to run a house for a week, one of Tesla's battery would not be enough; you would need a few. I guess I wasn't much clear on the post.

You don't need to run the house for a week for a battery to be useful.

The standard idea for this is for the houseowner to charge the battery during off-peak night hours and deplete it during peak hours thus helping the electric company to decrease the variation between peak and average load. This also has the effect of decreasing the amount of copper in a cable required to connect a house to the grid.

If enough people have batteries, it also helps grid reliability. Most grid failures are single point of failure so having a distributed source allows power to keep flowing while repair is being carried out.

There are a lot of advantages to having a big battery which could power a household for even a day.

In the short-term, this also allows homeowners to take advantage of variable-rate billing from power companies. Charging a bank of batteries in off-peak would be much cheaper than getting full current during the peak.

One problem is that eventually the cost savings is eroded by everyone jumping on the bandwagon to use the off-peak hours, which eventually become more expensive. Also, regarding your point about grid stability, if enough power consumers use these kinds of batteries, some sort of scheduling might be necessary to actually level out the grid's demand across all 24 hours.

Average household consumption is around 30-ish kWh / day, so an 85 kWh Tesla battery would give you almost 3 days. The question I have is why you would need to store more than a day's energy for your home, assuming you charge it with solar?

I suspect you need much less than this, and a 10-20 kWh home battery would be just fine to do the job.

Rain?

Rain is going to lower the solar output, but a properly sized installation will marginalize that.

If a person is going to make the "off grid" investment, managing a balance between the solar and battery investments is going to be necessary.

Batteries have cycle time and capacity limits, and their cost is high, but slowly improving. One probably should not assume a battery core will last 20 years, though some likely can.

The solar installation probably will go 20 years, and solar can deliver savings in both the generation of electricity, as well as light and heat. A properly sized installation is going to be designed to work, in lower grade weather conditions, and after it's lost some of it's efficiency, for a very long time.

Of the two, it makes the most sense to be lean on the battery investment. New tech, cost reductions, etc... will likely play out a few times during the life of the solar installation.

And that installation can be added to / upgraded somewhat more incrementally without having to give up earlier investments in many cases.

The battery, unless it is very modular, won't share these same advantages to the same degree.

And the house can be improved as well. Smart devices to manage power use, like the Nest thermostat, can be upgraded or trained, or just will deliver battery aware use profiles.

People can adapt a little. Just choosing to wear some clothes in the winter rather than maintain that comfort zone heat level saves quite a bit. LED lighting, thermal assist for warm water, use of more energy efficient materials, and the list goes on and on...

A couple lifestyle changes could very well see that battery last longer than expected.

That was Jason's assumption, not mine. :)

On my particular solar installation its production is lower in the winter than the summer (combination of temperature and Sun angle) 3 days would be fine for 99.9% of the year I expect and I would hardly ever draw from the grid.

Hi Simone, when I installed the solar system on my house (5.2kW, 2001) we seriously looked at going "off grid" (which means batteries). The problem at the time was that getting a few 10's of kWh of batteries was both quite expensive (you basically either built your own system or used an over priced data center UPS type system) and quite intense in its maintenance. Tesla is working both the reliability and the maintenance issues simultaneously.

That means that using their tech, with two of their battery packs (so that I can disconnect one for service/replacement while the other can run my house) would very much be an interesting "feature" for Solar City to offer. Typical peak power requirements of a house might hit 25kW but a car it can easily hit over 75kW (100 hp equivalent). So they have the inverter technology to build a whole house inverter.

The bottom line is they have built for a car, that which could be sold for houses, with the "hard bits" mostly solved. But that only becomes a useful thing if they excess battery capacity because I'm sure they can make far more money selling this tech wrapped up in a car than they can as a 'get your house off the grid' kit.

Hi Chuck, that's a good point - but I guess we're still a few years away from cheap enough batteries. :)

Depends on the house, but for the average Tesla owner I suspect you're right.

85 kWh would last me over half a month (I live in a smallish house with big windows, no A/C, LED lighting, and gas for heating/cooking/hot water)

Battery cost is 12k for the 85 kwh battery as of about a year ago, dropping ~7% a year.

Depreciation of the battery is going to be much less predictable than energy prices. Just because a simple formula is used in accounting standards, doesn't mean that the true economic value of the battery is anywhere close to the accounting figure. For example, what if the battery breaks and needs repair? That's build into the depreciation, and very unpredictable.

Not in aggregate. The law of large numbers guarantees that we can come up with a meaningful average cost of ownership. We just need more data.

Knowing averages doesn't help the individual homeowner who has to deal with the increased risk.

It does, though, in at least two ways:

1. There's little reason (a prior) to think that there would be much higher variance in the level of manufacturing defects than with any other consumer product, so knowing the average provides a pretty useful approximation for the quality of the product.

2. The individual homeowner should be purchasing homeowner's insurance, and the insurance companies use the average to approximate the risk, which allows them to shield the homeowner.

New products have risks that have been dealt with effectively in the past. There's no reason to think that Tesla's batteries will be any different.

You originally said:

I understood this to mean that you were saying that the risk to households caused by deprecation of batteries, was less than the risk caused by varying electricity prices.

You don't seem to be answering this point anymore. I'm not saying that batteries are especially risky. I'm saying that the risk from batteries isn't clearly less than that from electricity prices.

I don't think insurance can deal with this for the usual reasons that insurance involves moral hazard. But maybe batteries will not require any maintenance, or maintenance will be so standard, that moral hazard won't be an issue.

All of those reasons and really, no power outages anymore ever. Unless the grid is down for a long time, days, weeks you could be uninterrupted.

In the future it will probably seem ancient to not have this, and laughed at that we had power outages or rolling blackouts.

It would probably be great in the developing world and off the grid. It could enable housing to be away from the grid if the solar was good enough or even networks of battery grids and many other uses we cannot imagine yet.

This also solves the peak load problem, since you can pay for power at night and run off of your giant Tesla battery in the garage during the day.

Let's say that electricity was FREE during the night. What exactly is your electric bill today? $300 per month if you're a big user? If that battery costs $50,000 and reduces your eclectic bill to zero, then it pays for itself in a mere 12 years.

In other words: no.

It's a similar story to solar panels, except that it might work for a variety of people who live in northern climates or don't have a ton of roof space.

It also might be a good candidate for utilities, most of whom buy electricity at market rates, to forward deeply stored energy closer to the delivery point.

From jegutman elsewhere in the thread, batteries cost $12000, so even buying two would cut that to 6 years and I bet the batteries usefulness would last longer than that so it would definitely add value to a home and probably be a reasonable investment.

Why Battery cost $50K? More like $10K + patent fees at most.

You can see how batteries are made on some videos and most expensive part will be technology behind assembly. The high cost of Tesla cars is caused now because it is a new, unique product that needs huge amounts of live cash to fund the development etc. Its like with Apple where their phones could be priced 50% of what it is now, but there is a lot of money spent on marketing, future development and to build up cash reserves in case of market turmoil.

Unfortunately I don't think cities will deal with this well if there all of a sudden is a mass migration - they still will likely have costs/debt to pay off and divide that cost among everyone in cities. I'm not sure many cities are planning for this transition.

Reply on news.ycombinator.com