Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Smart thermostats inadvertently strain electric power grids

news.cornell.edu

Post reply on HN
Print view View original

131–140 of 226 posts

Re: Smart thermostats inadvertently strain electric power grids

#131

Post by SketchySeaBeast »

pdonis wrote: post #84
yellow_lead wrote: post #73

Earlier quoted context omitted.

It would be a smaller usage at 6:30, and not necessarily a peak, since usage would probably surpass it right after

> It would be a smaller usage at 6:30 Why? Won't all the heaters that turned on from 5:30 through 6:30 be on?

This might be a regional variation for how people are imagining this - I agree with you, living in a cold environment my furnace takes 2-3 hours to get the house up to temp, even longer in extreme (-30) cold, and will still have prolonged on/off cycles after the house heats up so that it's pretty much guaranteed to still be running in case of a staggered start. Warmer climates might be able to get things heated up in less than that hour.

Re: Smart thermostats inadvertently strain electric power grids

#132

Post by naasking »

michaelt wrote: post #66
furyg3 wrote: post #37

Earlier quoted context omitted.

This doesn't make any sense, though, as nobody turns on a heater for 10 minutes. Even if there is drift of 10-15 mins, there would still be a peak usage at 6:15 when all of the 6AM 'starters' are sucking power from the grid.

Imagine a grid supplied by a single coal-fired power plant. When demand rises the voltage drops and the frequency reduces, and the power plant has to start adding coal to the fire faster to get the voltage and frequency back to normal. If the demand rises over the course of 10 minutes, you get a small drop in voltage and the plant ramps up gradually. If the demand rises by the same amount over the course of 3 seconds…

> If the demand rises by the same amount over the course of 3 seconds, you'll get a much bigger voltage drop and the power plant has much less time to speed things up.

More than that, if the voltage drops too low, like from a sudden large power draw, you could get a brownout or even a blackout in severe cases. Voltage level is monitored and wild swings indicate instability which trips various safety systems.

Re: Smart thermostats inadvertently strain electric power grids

#133

Post by dragonwriter »

rblatz wrote: post #107
ceejayoz wrote: post #59

Earlier quoted context omitted.

It does work, though. Nest's program pre-cools to get you through the hottest hours without having to kick on the AC, moving some of the region's demand from the 4-6pm slot to the 2-4pm slot. I find myself going "huh, it's getting a little warm" at about 5:45pm, at which point I check and remember it's a "Rush Hour" session.

Does precooking really work? I’d assume it would have to cool my house down below 65 to keep it below 80 all afternoon. 111+ days are brutal in certain rooms even with both ACs set to 76.

> Does precoo[l]ing really work?

Depends on the shape of the temperature graph for the day, the power of your AC system, your insulation, etc.

If your on a day that doesn't get below 90 and gets above 110 during the hottest part of the day and your AC is barely rated at the minimum recommended for your square footage and you have average insulation, no, and even regular AC use is going to be marginal and your AC is probably going to frequent service.

(And the things that make precooling work better also will make your AC work better and, for the most part, more energy and maintenance cost efficiently when you aren't precooling.)

If you live in a place where you frequently need AC, aside from decent insulation, getting an overrated, high-efficiency unit (which seem to be correlated for residential units, probably because both drive up cost, and there is a correlation between people looking for more capacity and people concerned with efficiency) is a serious quality of life improvement.

Re: Smart thermostats inadvertently strain electric power grids

#134

Post by naasking »

pdonis wrote: post #84
yellow_lead wrote: post #73

Earlier quoted context omitted.

It would be a smaller usage at 6:30, and not necessarily a peak, since usage would probably surpass it right after

> It would be a smaller usage at 6:30 Why? Won't all the heaters that turned on from 5:30 through 6:30 be on?

Even if they're all on by 6:30, that gives the power plant time to ramp up production to meet demand from 5:30-6:30.

If everything is synchronized and everyone in a region starts drawing lots of power at the exact same time, the line voltage on the grid will suddenly drop below safety thresholds and could trigger a brownout for that whole region. The point being, randomizing power draw over a period of time allows the grid more time to match demand.

Re: Smart thermostats inadvertently strain electric power grids

#135

Post by yencabulator »

hyperpape wrote: post #7
labrador wrote: post #2

It's a well known problem, so I'm surprised no one thought about it ahead of time https://en.wikipedia.org/wiki/TV_pickup

It does seem like there are a few details that are different, because the thermostats will put sustained load on the grid, whereas the TV pickup will not. That means that a couple of nice technical solutions like adding jitter to the startup times would solve the pickup problem, but not the increased load at 6:00 AM.

> adding jitter to the startup times would solve the pickup problem

There's minimal jitter in people's behavior when they are explicitly synchronized by a television broadcast, and none can be added within the constraints of broadcast television.

Re: Smart thermostats inadvertently strain electric power grids

#136

Post by ceejayoz »

rblatz wrote: post #107
ceejayoz wrote: post #59

Earlier quoted context omitted.

It does work, though. Nest's program pre-cools to get you through the hottest hours without having to kick on the AC, moving some of the region's demand from the 4-6pm slot to the 2-4pm slot. I find myself going "huh, it's getting a little warm" at about 5:45pm, at which point I check and remember it's a "Rush Hour" session.

Does precooking really work? I’d assume it would have to cool my house down below 65 to keep it below 80 all afternoon. 111+ days are brutal in certain rooms even with both ACs set to 76.

I would not expect this sort of solution to be super useful in a place that gets 111 degree days, where AC is literally keeping folks alive. You'd probably get more benefits from a housing code that requires extended roofs for shade, light colored exteriors, etc.

Re: Smart thermostats inadvertently strain electric power grids

#137

Post by tguvot »

reaperducer wrote: post #126
tguvot wrote: post #85

Earlier quoted context omitted.

i have walls insulated according to california build code. it's not really possible to add more insulation to them, unless you rip them off and do spray in insulation. insulated attic with radiant barrier and insulated floors. i just checked with IR camera, most of surfaces in the house are now 22C-24C. After a few days of 40+C walls ill go to 25-26. some walls (south side of house) even higher. There are few little…

it's not really possible to add more insulation to them, unless you rip them off and do spray in insulation. Actually, it's possible to inject foam insulation into walls without ripping them off. It goes in through small holes, or through utility boxes. Plenty of companies do that. I've even seen it advertised on TV in California.

I know.

But injectable foam insulation usually going into empty walls. I already have butted fiberglass in there. I don't think it's viable to make laparoscopic wall surgery to remove fiberglass . I did a quick google, and looks like you can do injectable foam over fiberglass. But I am not sure how much R it will add, but I am sure that for cost of of injectable foam + wall repair, I can get another hvac and run it for next 10 years...

I did consider recently to double walls on sunny side of the house with few layers of foil backed foam boards.

Re: Smart thermostats inadvertently strain electric power grids

#138

Post by giarc »

LazyMans wrote: post #89
taylodl wrote: post #71

I really don't like the strategy of super cooling in the morning. I prefer fresh air instead. I throw the windows open wide and let all the cool morning air in. By midmorning I'm usually starting to close those windows because the outside air is getting warmer. By mid afternoon I'm closing all those windows and turning on the AC and setting the temperature to 1 degree (Fahrenheit) cooler than the current room tempera…

I predict in certain areas, "free cooling" systems will become the standard. These are system which do just what you talk about automatically. When there is a need for cooling and exterior air is cool enough, supplement supply air with air from the outside. This way your AC system load is reduced, or if cool enough, the compressor doesn't need to run at all to remove heat from your home by replacing it with cooler ai…

I wish I had a system like that and have thought about it a lot. I live in Calgary and today will be 28C and it will go down to 12C at night. I wish my system could just run a fan at 7am and draw in a ton of cool air to cool the house. Instead my AC runs for 30 minutes or whatever. Seems like a waste.

Re: Smart thermostats inadvertently strain electric power grids

#139

Post by philwelch »

kennywinker wrote: post #112
saurik wrote: post #98

Earlier quoted context omitted.

Why would you expect someone who isn't willing to sacrifice their comfort for a $10/mo incentive to sacrifice their comfort for a $0/mo incentive? Deciding "I don't like this person's ethics" doesn't mean "this person is suddenly inconsistent": you need to check your knee-jerk responses :/.

Well, for one thing people will often do for free things they wouldn’t do for money. Imagine choosing between “i will pay you $10 to come to my party” and “i’m having a party, will you come?”.

Parties are supposed to be fun, so if someone offers to pay you to come to a party, that sends a signal that the party is not, in fact, going to be fun at all--otherwise, why would they pay you?

On the other hand, paper straws are a shitty product that are unsuitable for use, but if I was offered a significant discount on beverages that were served with them...well, I wouldn't take it, but I could imagine someone who was desperately broke considering taking it.

Re: Smart thermostats inadvertently strain electric power grids

#140

Post by rsync »

tguvot wrote: post #64
evanreichard wrote: post #11

Ecobee has eco+ which has "Community Energy Savings" [0]. It was just way too intrusive. It seemed to be a reactive system vs proactive. Technology Connections goes into this more [1], but I believe they should be providing a system that proactively cools when there is less demand _prior_ to an expected increase in demand. Effectively using your house as a "battery". When I had it enabled, I was _always_ too hot (in…

in reality it's hard (and probably expensive) to proactively cool house with AC during summer at hot/sunny regions. I live in a house with rather good insulation (the one mandated by code) in south bay, yet all the walls getting rather warm and even if you try to precool air with AC walls with bring temperature back quickly My solution was to install whole house fan which runs through the night when temperature drops…

Whole house fans are really great.

It’s cheating, a bit, to live in a place that drops 20-40 F at night … but even if we didn’t, the ability to turn every open window into an air conditioner is a great trick.

Post reply on HN