Hacker News Reader

Comment by mcv | Hacker News Reader

If it's incredibly hot, doesn't that also mean it's incredibly sunny? Sounds like solar would be the perfect solution here. In Australia, energy prices sometimes go negative because of the abundance of solar energy[0]. Admittedly not during the hottest days when everybody turns up the AC, but using solar to supply that energy still seems like the most logical approach.

Of course alternating between extremely high and negative prices due to the volatility in supply and demand should create plenty of opportunity for businesses that make money simply storing energy and evening out the fluctuation.

[0] https://reneweconomy.com.au/wind-solar-push-south-australia-...

View on HN

One problem is that solar's peak is too early in the day to match peak demand in most of the US (including Texas). Texas's electricity usage typically peaks around 4-6pm in the summer, while photovoltaic output peaks around 12-2pm. There is some storage, but you would need a lot more to line those curves up. A cost-effective way to shift the solar peak to about 4 hours later would definitely help on a lot of days.

For the curious, you can browse the demand curves day-by-day here: https://www.eia.gov/realtime_grid/#/summary/demand?end=20190...

Demand response can load shift cooling onto times when solar is more abundant. DR is criminally underused and will be one of the top ways we will utilize renewables going forward.

Hell, you can even use DR to shift electric water heating to free up scarce power during peak A/C load.

You can shift the solar peak at least by one hour when importing electricity from California is possible... In Europe there are a lot of international power cables that help redistribute power across the region and that seems to work quite well (fingers crossed).

The Texas power grid is isolated and cannot routinely import power. It can in very temporary and emergency situations, but they do not as a matter of course to avoid federal oversight.

So while the US Western grid can time shift one hour in the way you describe, Texas is isolated.

https://en.wikipedia.org/wiki/Texas_Interconnection

https://www.texastribune.org/2011/02/08/texplainer-why-does-...

Yup, because ... Texas. In the entire country there are 3 roughly independent systems: East, West, and Texas. Maybe they’ll reconsider given these incredibly high prices?

There have been instances where the power spot price has gone negative, too. They have not had major outages. I don't see why they would reconsider.

There's definitely an opinion in Texas that ERCOT is more reliable than the other two US grid managers, so it's popular to remain independent for that reason as well (besides the political reasons). Both the Eastern [1] and Western [2] Interconnections have had major grid-management incidents that took down large portions of the grid in cascading failures, and some analysts think further incidents are likely [3], while the Texas grid has never gone down.

There is some support for expanding the capacity of the asynchronous DC interconnects, though, which would let Texas import & export more electricity to the other two grids without tying itself to them directly. It already does a modest amount of that, but the current DC links don't have very high capacity.

[1] https://en.wikipedia.org/wiki/Northeast_blackout_of_2003

[2] https://www.nwcouncil.org/reports/columbia-river-history/bla...

[3] https://www.vox.com/2014/4/14/5604992/us-power-grid-vulnerab...

Yeah Texas just has rolling blackouts every other month. Totally reliable and not considered an outage how?

As far as I'm aware, there have been no rolling blackouts in Texas since the last major incident in winter 2011. Got a source?

No, but blackouts after storms are pretty common. Sometimes lasting for days.

IMO, that’s a big part of the reason why backup generators are so very popular here in Texas.

Certainly, that’s the reason I have seriously considered installing a 10-15KW natural gas powered backup generator. It wouldn’t be grid-tied or even grid-interactive in any way, but it wouldn’t take me long to go downstairs and throw the transfer switch.

What part of TX do you live in? My parents have lived in suburban Houston for 20 years, and their power has basically never gone out for more than short blips, not even during tropical storms or hurricanes. The power lines in their neighborhood are buried though. I imagine if you're getting frequent outages you must have above-ground lines that get downed by trees? That can be an issue, but doesn't have a whole lot to do with the state-level grid, just whether a given subdivision or city has chosen to spend the money to bury the last-mile lines or not.

California has far more rolling blackouts. When I lived in Texas, the only power outages I ever saw were during hurricanes or some random local event like a transformer going out.

Texas may be somewhat isolated from neighboring grids but it does have some ability to do some solar collection time shifting since the sun take over 50 minutes to traverse the entire state. Texas spans almost the entire central time zone; El Paso is in the Mountain time zone.

Nevertheless, Texas seems to be an ideal location for both wind and solar energy (and ironically, oil and gas too).

Texas is actually the largest producer of wind power, by far.

https://i.redd.it/zug13sa7e1521.jpg

But lags in solar.

It's possible in theory but California is a net importer pretty much all of the time.

You should be able to shift it by 45 minutes just by locating the solar generation in El Paso since it's 500 miles west of the major population centers and 700 miles from the eastern border.

Sunset in El Paso is at 7:51 today (mountain time). Sunset in Longview is 8:05 (central time).

El Paso also has more reliable sunshine than Austin and Dallas and Houston.

Even with 4 hours of battery storage solar should be substantially cheaper than gas peakers.

There also should be a huge amount of opportunity for demand side behaviour. For instance someone coming home at 6 could precool their house at 3 and save money, we just need to make that easier for consumers.

Pre-cooling your house at 3 in the Texas summer will last until about 3:30 at the latest.

That's particularly frustrating, because effective building techniques for keeping houses cool in hot climates have existed for millennia.

The problem is all those ancient tech items require long term thinking, such as shade trees, masonry, insulation, subterranean, etc... This sounds more like a cultural shift from cheap and fast stick-built: they're spoiled by that mentality but paying the price now.

Yeah, but I don't want to do this: The town where people live underground: Welcome to Coober Pedy [1]. :(

[1] https://www.youtube.com/watch?v=uFA-OB-nJ_I

No one said you have to go anywhere near that far.

Thicker, solid walls with more thermal mass that soak up the overnight cool to release throughout the day. Better insulation and air sealing. Trees, roof overhangs, and shade cladding to prevent direct sunlight from hitting the house.

"Just add a bigger AC" has been the approach for far too long, and it's not sustainable.

Overnight low of 81° F in Houston tonight, and it probably doesn't get that cool until 2 am or so. There's no overnight cool to soak up. It's also super humid just to cooling more challenging.

Sure, there will always be some nights in the height of summer where you want some AC. The goal isn't total elimination; it's reduction.

Running AC in the dark night to cool that thermal mass will benefit during the day. Using some energy, but substantially less and at non-peak times.

https://en.wikipedia.org/wiki/Climate_of_Houston indicates 81 is above the average... and again, thick walls aren't the only available approach. Any approach will involve a collection of different tweaks.

The production cost of peak solar is probably a tenth of the cost for a gas peaker. You'd really have thought in a rational market there would be enough value there for people either getting a distinct thermal storage option or good enough insulation to make precooling work.

Sounds like Texas needs better insulation standards for their housing.

There are cooling systems that freeze a tank of water at night when power is cheap and run the A/C system though it during the day. I’m not sure how effective it is though.

Here in Chicago we cool entire banks of skyscrapers using that method. It's very effective.

https://enwavechicago.com/how-district-cooling-works/system-...

It’s called an “ice bear”, and is actually very effective at the level of an individual home.

I don’t know why more people don’t install them here in Texas.

Pointing the panels more westward could fix this with only little losses in peak efficiency.

Seems silly but I’d imagine a lot of places could use the lower rates during peak solar to precool the buildings.

My house only heats up about two degrees per hour with the ac off. In theory I could precool by ten degrees and coast until the evening.

Do you live in east/mid Texas? That sounds like some super great insulation if you are only accumulating 2 degrees an hour during a hot humid Texas summer.

Similar latitude actually. My house is concrete block though so it tends to resist temperature change a bit.

Also keep in mind large buildings will resist temperature changes even better than houses due to their size. Less surface area per volume.

My house in Dallas only rose by about one degree per hour during the day when the A/C was off, even on the hottest days. It was pretty well-insulated, and the large trees around it helped.

> humid Texas summer

This is probably the warped idea of "humid" I got from growing up in South Florida, but Dallas was rarely ever humid.

Dallas is relatively humid, but not Florida, no.

It also doesn't have the constant sea breezes and incidental rain that most of Florida enjoys.

Houston is as humid as it gets though.

Build your houses out of stone rather than wood, and they'll have a 10+ tonne thermal mass. Just a few inches of insulation around that on all sides, and the temperature barely varies day to night.

Our house went from 78 to 85 in 48 hours with the AC completely off. With highs over 100. We have an old house with basic insulation upgrades.

I'd expect it is a combination of insulation and thermal mass.

Good insulation is a great investment for any home. Keeps your house cooler in summer, warmer in winter, dramatically reducing your energy needs in both cases.

Do any of the smart thermostats have this function (tying in with rates?)

Yes they do. Austin Energy has the Rush Hour rebate program that precools your house before peak demand. It’s compatible with several programmable thermostats. It’s an opt-in program and you can override your thermostat if you want. Austin’s rates don’t change with demand, but they are tiered and increase the more electricity you use per month.

My Nest is signed up with my local power provider. In the summer, they turn up my temp a couple degrees from 2-6pm a couple times a month.

There's a pre-cool from 1-2 where they drop the temp, and if I leave it alone for the entire four hours, I get $5 off my power bill that month.

My parents (who live in Texas) get a rebate like this from their electricity company, which surprised me by even having an option to do it manually without a Nest-style internet-connected thermostat. If it's a day that the company wants reduced peak usage, they text+email you that you can get a rebate if you reduce usage during a certain set of hours. They judge whether you did or not by measuring your usage during those hours against your historical usage on similar-temperature days, so you can get the credit just by turning up your thermostat manually for a few hours. (This does require a new-style meter that reports hourly usage.)

Time of use rates will even get the residents to do it (precool) themselves.

This is where stationary storage (via batteries) makes a lot more sense. In Australia, the same problem exists, but they have an giant ozone hole so even more UV hits the surface. Tesla deployed what (at the time) was the biggest battery storage project in the world. They built a 100MW/129MWh battery setup with their Powerpacks. In another part of Australia, they're deploying a "virtual power plant" by equipping up to 50,000 homes with individual home battery backups via Tesla's Powerwall product. The grid operator will be able to tap into each of these batteries to redirect power to the grid or shift load off of the grid whenever they see necessary. This is resulting in massively improved prices for electricity in South Australia where they had a lot of rolling blackouts (due to the heat) last summer.

The newly released Tesla Megapack https://www.tesla.com/megapack seems like a slam dunk for this sort of thing. You store that crazy amount of sunshine that is excess in batteries and use them that evening or the following morning. The economics of this approach are becoming more and more realistic due to better technology and better manufacturing.

The problem in urban Texas is that all the asphalt and concrete acts as a stationary heat battery. Counteracting that is a huge part of the problem.

They could use batteries and air conditioning to do so, but it's likely the return on investment would be higher figuring out how to provide more shade even in non residential areas.

It's a shame we can't convert that heat to electricity in some efficient way.

PV solar roadways are a total bust, but I do wonder about liquid loops under the asphalt. If you need a source of cold to create a temperature gradient vs the road, you could always pull from geothermal next to the roadway collector.

Oh we can convert high grade heat to electricity with amazing efficiency. Problem is, this is super low-grade (low temperature) heat. Darn you, thermodynamics.

Provide shade by putting up solar panels above the roads.

I‘m imagining that Texas has far wider roads than Europe. Would it be possible to plant street trees or other tall native vegetation to provide shade?

Houston does just that. Streets in Houston are lined with live oak, providing shade to much of the street.

Have you been outside the loop? There is no culture of shade protection in Texas the way there is in other not places like Australia or mideast. Massive concrete car parks with no shade structures in sight. No solar panels on the roofs of big box stores. It's pathetic.

Would it make sense to mix some TiO2 or other whiteners into the asphalt and concrete to increase their albedo?

Edit: Looks like LA is doing something like that https://www.mnn.com/earth-matters/climate-weather/blogs/how-...

Does Australia really have a giant ozone hole? It doesn't seem so according to the Australian Bureau of Meteorology[1]. Also see [2].

[1] http://www.bom.gov.au/uv/faq.shtml

[2] https://www.huffingtonpost.com.au/2016/06/30/australia-never...

I don't think [2] sounds like a reliable source. It presents a "hole" as a binary thing, which didn't extend to Australia. But even over Antarctica, it was a level of depletion, not literally a hole. There may be a point to be made about the numbers, but the whole article seems to be predicated on the assumption that either there is a hole or there isn't without defining it.

Actually, what makes even more sense, is to have a non-conversion storage.. aka.. just pre-freeze a large amount of coolant at solar peak and store that for the next day.

Doesn't it make sense then to send the power eastward while max generating it and then import power from the west later in the day when you need it? You could theoretically leverage the time differences and peak usage/generation. I understand there is losses in transmission but seems like a good juggling act

The Texas grid is not integrated with other states. I'm guessing breaking this "independence" is a tough political sell.

Power arbitrage between non-connected grids is a thing in the rest of the world. I'd be very surprised if nobody is doing it to/from Texas.

It would require DC/AC convertors, so there is a slight energy loss, and fairly significant upfront capital cost.

If the utility offers TOU billing, more solar installations will point west, shifting peak supply later in the day.

I looked at the chart you provided, and calculated the difference between 12-2pm and 4-6pm usage, which seems to be only 13%.

Is this really such a huge hurdle in itself? I'm curious, not trying to be dismissive.

I think Chillers with ice storage could be used to handle this somewhat. They already are used for that purpose to buy power off peak prices and then distribute the cooling when its needed thus de-coupling the supply and demand somewhat. This is mostly just armchair speculation on my part however.

If they required/subsidized better insulation, they could pre-cool the buildings to a bit lower than optimal temperature and have a huge thermal reserve. But it's Texas, so...

All new residential construction is required to meet International Residential Code energy efficiency provisions. The State Energy Conservation Office provides funding opportunities to support efficiency programs and loans for public institution retrofitting projects. There are many programs requiring, subsidizing, and incentivizing energy efficiency: https://programs.dsireusa.org/system/program

What do you mean by it's Texas?

Commercial solar plants in Texas take advantage of the market by aligning themselves a few degrees off the optimum for total power, so that they can produce better in the heavier demand/higher spot market rate time periods in the evening.

A few percentage points lower in total mWh produced but more profit.

So maybe people should cool their homes while away at work, assuming they’re insulated enough to then need less cooling after work.

There is less of a relation between heat and solar intensity (and therefore solar cell output) than one might think.

A cloud-less summer day in Texas offers the same potential solar energy no matter the temperature. And as you go higher latitudes, temperature drops far faster than irradiance. Which is why solar power in southern states is only moderately more productive than in New England or, for example, Germany.

There's almost always some mis-match between these two. The "duck curve" [0] is the best known example of this issue.

[0] https://en.wikipedia.org/wiki/Duck_curve

This is somewhat true but a side effect of PV is that its performance degrades with higher temperatures. It is possible to use panels output as thermometers they are so consistent in this behavior.

I'm not saying they will stop working but merely they lose out in some potential.

The efficiency of photovoltaic tends to decrease with heat above 25°C.

Storing energy is expensive.

Storing energy is expensive - but as you can see - the value that the market has put upon this instant dispatch energy is also high.

The solution is to simply build more natural gas peaker plants, not solar. If you read the article it pretty much shows that Texas overinvested in wind power and it just so happened that during this period of time the wind wasn't as strong. Also for Texas having an energy crisis like this is unacceptable because Texas is sitting on the largest natural gas fields in the country.

Texas is sitting on the largest natural gas fields in the country

These things take time to sort themselves out. E.g. right now Texas is flaring a lot of natural gas (which they get as a byproduct of oil production).

Looked at dispassionately it seems insane.

https://oilprice.com/Energy/Natural-Gas/Gas-Flaring-Running-...

It's a persistent fallacy to assume that sunniness and heat equal more solar power. Reality is that heat decreases the efficiency of solar, so, the hotter it is, the less efficient solar will be.

Reply on news.ycombinator.com