back to article Smart thermostat swarms are straining the US grid

Smart thermostats, those unassuming low-power gadgets designed to keep homes at comfortable temps, are having an impact far wider than most might have considered, according to recent data. A paper from Cornell University brings bad news for renewable energy enthusiasts – smart thermostats are secretly taxing the grid. Smart …

  1. Ali Dodd

    I do like em but everything has issues and "with most homeowners only seeing energy savings of 5-8 percent, as opposed to the 25-30 percent they're capable of."

    capable of? I'd read advertised as capable of, I doubt any smart thermostat would ever save you that much unless you previously never set a schedule..

    1. Persona Silver badge

      Mine saves me some money but only because I can turn the heating off when I'm on holiday without the prospect of coming back to a freezing cold house and worse, having the wife complaining to me about it.

      1. Ken Moorhouse Silver badge

        RE: I can turn the heating off when I'm on holiday

        You could have left notes as to how she could switch it on in your absence.

        (Ah the ambiguities of the English language).

    2. hoola Silver badge

      Ah but this is all marketing. Just like this obsession that Smart Meters somehow save people hundreds of pounds.

      None of these gadget that claim to be "Smart" are actually smart in any way. They are just connected to the internet, some websites, back end logging and an App. Now it is the last bit that for some reason is perceived to make it smart.

      I have a thermostat at home on my heating, when it gets cold, it turns on, when it warms up it turns off. I know what temperature it does that at & don't care if it is accurate, just if it is comfortable (17 degrees + a jumper).

      Unsurprisingly all this "Smart" techy shite is turning out to be rather less smart and in reality no better, if not worse than the long-lived things they have replaced.

      1. Richard Jones 1
        WTF?

        Even better, the dumb 'smart' you use does not rely on a soon-to-be cancelled remote server like, let me see, could it be Hive?

        1. Ali Dodd

          Nope I'm not stupid enough to buy a Centric/British Gas product, they will screw you every time.

          A 7 day timer is good. A timer that adjusts for temperature in every room and regulates the main heating is much better, plus it also stops heating automatically when I'm out of the house/on demand & starts when asked remotely. It's not 'smart' per se but it is a damn sight smarter than a fixed standard 7 day timer and a lot more useful.

          Had a Nest (quite dumb and limited) and now have a Tado. Which is as far as I'm aware their core business and not particularly likely to go into the sunset for a while. Not everything is tainted by money grabbing fools and arseholes like Hive (peleton, etc) blatantly is.

          1. Anonymous Coward
            Anonymous Coward

            Core business companies are the sort that get bought up by bigger companies wanting to exploit another market niche or set of customers. Then they often EOL the products - and possibly just re-badge their own incompatible offerings.

      2. 43300

        And the 'old' and 'basic' stuff carries on for years, just doing its job. Unlike the smart shite where you'll be lucky to get five years out of a lot of it.

      3. Missing Semicolon Silver badge

        "Smart Meters somehow save people hundreds of pounds"

        They will, when you notice that electricity is £5/kWh, and you turn everything off.

        That was always the point - and why they were insistent on the display. You can't roast people with surge pricing, unless they can see the price.

        The purpose of Smart Meters is to implement "demand management". No need for load-shedding when people do it themselves.

        1. cyberdemon Silver badge
          Devil

          Re: "Smart Meters somehow save people hundreds of pounds"

          They also include a handy remote-controlled disconnector, which will perform load-shedding if demand-management fails. i.e. if you don't pay your monthly (or weekly) bill, then the lights go off immediately.

          Probably, they will soon include a "feature" where the cash cow customer can set a limit on pricing i.e. if the price goes above £1/kWh then it will open the contactor.

          We will all be squeezed bone dry for these fat cats. Free-market capitalism at its finest. Mrs T would be proud.

      4. John Brown (no body) Silver badge

        "if not worse than the long-lived things they have replaced."

        I suspect that is the most salient point in the long run. A cheap electromechanical thermostat, maybe with a basic timer, probably lasts decades and more than "pays for itself" in carbon costs. Fancy modern "smart" thermostat, lucky if it lasts 5 years due to either electronics failure or the "smart" host going under or being deprecated and all the barely recyclable e-waste it generates over the decades the old style one will last.

      5. MachDiamond Silver badge

        "Just like this obsession that Smart Meters somehow save people hundreds of pounds."

        They weren't to save customer's money. They make more types of measurement than could be done with the old fashioned meter and can be read from a distance, even a large distance so the power company saves money and doesn't need close access to the premises to read those meters. A huge benefit for them is the ability to switch off the power and switch it back on after a person has paid the same hefty deposits but without sending a person to the property. The same savings are there when a property changes hands and there needs to be an accounting as of that date for proper billing.

      6. cray74

        "Unsurprisingly all this "Smart" techy shite is turning out to be rather less smart and in reality no better, if not worse than the long-lived things they have replaced."

        I was annoyed when the homebuilder stuffed my new house with smart everything - garage door opener to thermostat - but the gizmos have worked out well. My elderly father, who also lives there, has vision issues. The smart devices let me check the home's status and fix settings from my office desk. I've been able to change the air conditioning from my father's accidental selection of "emergency heat" to "cooling" on several summer mornings.

        But, no, the smart controls haven't radically altered my home or utility bills. They just work well for my particular home situation.

    3. Plest Silver badge
      Mushroom

      Great if you want to hand out free cash!

      The key here is "potential to save".

      You can have all the gadgets you like that allow you to save but unless you learn how to use them effectively, no one will ever get any savings. I do like the smartmeters and controls but when we had ours fitted we had a 20 year old knackered heating system ripped out and completely replaced, we knew that simply putting in smart gadgets on our old crap system would be a complete waste of time and money, so we waited 6 months, saved up and had the whole system replaced.

      Here's the kicker. You tweak and tune your smartmeter, that means you save money ( I'm saving about 10-15% a month now after 3-4 week tuning it! ) and guess what? The leccy company put the direct debit up despite me paying them less, so now they get to cream all that interest off the money I've saved that should be in my pocket! I wont' get that DD money back for at least 12 months and given inflation it's losing value every day!

      Why bother you have to ask yourself? To save a few polar bears, yeah OK but to save me money and then watch some greedy energy company hoard it and earn off my money! The energy business needs a good kick up the arse!

      1. Ben Tasker Silver badge

        Re: Great if you want to hand out free cash!

        > The leccy company put the direct debit up despite me paying them less, so now they get to cream all that interest off the money I've saved that should be in my pocket! I wont' get that DD money back for at least 12 months

        If you're like me and track your reads/usage, you can probably get the DD adjusted.

        I've had reasonable success with going back and saying "here's my usage over the last year/2years using your current pricing - you can see our monthly usage averages at x/month, so the DD should be reduced to that".

        1. John Brown (no body) Silver badge

          Re: Great if you want to hand out free cash!

          !I've had reasonable success with going back and saying "here's my usage over the last year/2years using your current pricing - you can see our monthly usage averages at x/month, so the DD should be reduced to that".

          I've never bothered with all that. I just phone them up and complain they put it up too much and every time the person on the other end agrees and puts it back down again. No muss, no fuss, no "proving" my usage. It's just the "computer" sets a rate and it's always in their favour. I suspect it will be a bit different this year though when the next price negotiation comes around :-(

          1. David Hicklin

            Re: Great if you want to hand out free cash!

            > The leccy company put the direct debit up despite me paying them less, so now they get to cream all that interest off the money I've saved that should be in my pocket! I wont' get that DD money back for at least 12 months

            Got the same here although my calculations show a decent credit building up by late summer - BUT...

            ..That assumes the prices **stay the same** and all indications are for a massive price increase (again) come the autumn price cap adjustment, so maybe they are trying to buffer us against them.

          2. 43300

            Re: Great if you want to hand out free cash!

            If you use a variable DD you just pay for what you've actually used each month based on the meater readings.

        2. MachDiamond Silver badge

          Re: Great if you want to hand out free cash!

          > The leccy company put the direct debit up despite me paying them less, so now they get to cream all that interest off the money I've saved that should be in my pocket! I wont' get that DD money back for at least 12 months<

          The DD is your problem. It could be an even bigger problem if there is an error in their billing and you get a bill for umpteen thousand and it comes out of your account. Or, your paycheck is a couple of days late and the autopays you have set up send your account into overdraft.

          I much prefer to get bills sent to me on paper and spend a short period of time some evening to make the payments after review. I've had issues in the past. Paying a bill a day late is far less expensive than even one overdraft fee.

      2. Jimmy2Cows Silver badge
        Mushroom

        Re: You tweak and tune your smartmeter, that means you save money

        <rant>

        Sorry but this is just utter bollocks.

        You might be tweaking and tuning your behaviour in response to what you see on your smart meter's usage display, but your smart meter itself can do absolutely nothing to reduce your energy consumption.

        All it does is log your usage and report it to your supplier on demand. You know, metering stuff. It doesn't have buttons you can push or settings you can fettle to magically reduce your bill. It may even increase conumption a tiny bit, if your meter is leaching off your mains to run itself.

        The only way a smart meter could actually save you money itself, is if it constantly checks supply prices and switches you to the cheapest supplier. It would have to do this frequently, say every 30 seconds or minute. That would actually be smart.

        Of course this is impossible since we'll all bound to whichever single provider we've chosen to buy electricty from, and they would never open themselves up to such competition.

        Smart meters are only a tool to get more money from the consumer. Think a future of per-minute spot pricing and surge pricing when demand is higher. Any consumer benefit is an accidental side effect.

        This is why I will never voluntarily have a smart meter in my home. I use what I need to use. I turn things off when I'm not using them. I don't leave shit switched on just for the sake of it. All my lighting is low power LED. I've halved the brightness on all TVs and monitors in the house (2 kids + working from home, so that's a lot of screens). Other than that I use what I need to use, and I don't need a "smart" meter to take those reduction steps.

        </rant>

        Icon = my favoured solution to the mountain of "free" smart meters that have collectively cost us over 12 BEELLION pounds in increased bills to pay for the fucking things.

        1. heyrick Silver badge

          Re: You tweak and tune your smartmeter, that means you save money

          "This is why I will never voluntarily have a smart meter in my home."

          The problem isn't really the smart meter (I have one, here in France we weren't given a choice). The problem is that you seem to have suppliers who can tell what your consumption is day by day, and yet still feel eligible to take whatever the hell they like by Direct Debit.

          Can you not cancel the DD and pay the bill online? Then they can't help themselves to whatever fiction they devise.

          1. Anonymous Coward
            Anonymous Coward

            Re: You tweak and tune your smartmeter, that means you save money

            If you don't pay by direct Debit then they usually add a premium to cover their payment handling costs.

            E.on electricity allow me to modify my DD in my online account. I can raise it to quite high levels - although I have never tested for any actual limit. However - it only allows a decrease up to 10% - no matter how much excess credit is building up.

            British Gas is similar - except you can't decrease it in the online account details - you have to get through to a person to negotiate a reduction.

            There is some UK regulation in the offing to restrict suppliers from holding onto customers' large credit balances.

            1. MachDiamond Silver badge

              Re: You tweak and tune your smartmeter, that means you save money

              "If you don't pay by direct Debit then they usually add a premium to cover their payment handling costs."

              I'd scream. I've made it a point to never have any utility on direct payment. One slip of the digit and your bank account is toast. Maybe in 60 days or so they'll finally agree that you couldn't possibly run up that much usage without a significant upgrade to the lines/pipe and will deign to adjust the billing, perhaps send somebody around the physically check the meter readout or verify it's working correctly.

              It would suck to have power shut off when a mistake has them sending a bill that's way out of bounds, but I'd rather that than to not be able to pay my other bills where there is no dispute.

      3. heyrick Silver badge

        Re: Great if you want to hand out free cash!

        "The leccy company put the direct debit up despite me paying them less"

        This doesn't make sense. A smart meter is supposed to report your usage to the electricity company, who can then bill you for the amount of energy used (plus service charge). No more, no less, and certainly not some figure plucked out of their arse. Because, otherwise, what the hell is the point of a smart meter?

        1. John Brown (no body) Silver badge

          Re: Great if you want to hand out free cash!

          The DD is supposed be the average monthly cost of your expected usage over the next 12 months or whatever, but is usually set at a level that leaves you with significant overpayment at the end. Theoretically, the smart meter should be giving them better data to set the DD at a more realistic level. But why would they when they can hold onto £millions in overpayments, especially across the summer and collect the interest on it.

          I've always found a quick phone call sorts out the high level DD proposed and get it back to a realistic level leaving me with a small overpayment come the next re-assessment. If it's over £50 I ask for it back and it's normally in my account before the next DD goes out.

      4. J. Cook Silver badge

        Re: Great if you want to hand out free cash!

        I do like the smartmeters and controls but when we had ours fitted we had a 20 year old knackered heating system ripped out and completely replaced, we knew that simply putting in smart gadgets on our old crap system would be a complete waste of time and money, so we waited 6 months, saved up and had the whole system replaced.

        THIS.

        A smart thermostat is only ever going to be as efficient as the house it's installed in and the equipment it's connected to.

        My house was built in 1961 and still has the original single pane, aluminum frame windows- Hence, the house leaks air live a sieve, and the winter heats goes right out through the glass in winter and back in during the summer. I still get pitched for solar systems on the roof, and my default answer is "I want to replace the windows before I consider solar" for various reasons.

        1. Anonymous Coward
          Anonymous Coward

          Re: Great if you want to hand out free cash!

          In 1985 my 1970s house had steel window frames. In the middle of a very cold spell PVC double glazing immediately made a difference. A few days later the hollow cavity was filled with rock wool insulation. Again a noticeable improvement - especially as one wall is the end of the house block.

        2. MachDiamond Silver badge

          Re: Great if you want to hand out free cash!

          "my default answer is "I want to replace the windows before I consider solar" for various reasons."

          When real audits are done, most "smart" homes are so much more expensive they have no return on the added cost to build given the normal life of those devices. Many of the "smart" gadgets are made by companies that don't survive and it's twice as expensive to replace them with something later on. It would have been much cheaper to put in the old manual gear in the first place.

          I'm in the same position with regards to windows. Unfortunately, I have a "manufactured" home with non-standard size windows so some reframing will be required. I think I'll be able to get to the big living room windows on the North side before winter comes around. I did some DIY solar heating panels on the south side that do really well when the sun is out and low in the sky. In summer I can use a swamp cooler to keep the house down to a good temperature. I don't have anything that a smart thermostat would be used with.

        3. cray74

          Re: Great if you want to hand out free cash!

          "I want to replace the windows before I consider solar" for various reasons.

          Definitely, insulation can work wonders for electricity bills and it's generally cheaper than solar panels.

          I have solar panels on my roof but the real electricity savings on my new home are heavy insulation and low-E double glazed windows. Without the benefit of solar panels, my electricity bill is just 1/3 what my parents paid in the late 1980s (same part of Florida, same size home, no inflation adjustment.) All the insulation I added to my home above the builder's minimum is about 1/4 the cost of the solar panels.

    4. John Brown (no body) Silver badge
      Thumb Up

      "capable of? I'd read advertised as capable of, I doubt any smart thermostat would ever save you that much unless you previously never set a schedule.."

      Two words. Up to. :-)

  2. Adrian 4 Silver badge

    If switch-on time is so predictable that they generate a surge, why not rev up ready for it, as has always been done for advertising breaks ?

    I strongly doubt the 40% market penetration, too.

    1. elsergiovolador Silver badge

      I would be more worried that probably majority of these thermostats are Chinese and connected to internet.

      Which means, in theory, someone in Beijing may have a big red button ready to turn the heating to the max at all US homes.

      1. M.V. Lipvig Silver badge

        Or, on-off-on-off-on-off-on-off-on-off-on-off-CLACK! and there goes the grid for a week.

    2. Anonymous Coward
      Anonymous Coward

      Knowing there will be a peak is different from having the ability to always have low carbon power to put in the grid at the peak.

      Back in the day. we didn't worry about bringing another coal fired power station online for the ad breaks in "The Price is Right", but nowadays, we'd rather not do that, especially for a short period (I do recognise that it's more likely gas nowadays for transient peaks).

      The paper is observing that if the peak was lower, by having a little more randomness in the demand, there would be fewer problems supplying the overall demand as it would be spread over a greater period of time.

      1. LDS Silver badge

        If the thermostat gets its time from NTP and most of them are set at the same hour they will create a spike. They could randomize the turn-on time a little to avoid that - probably old thermostat clocks drifted away enough to spread the load over some time.

        After all even servers set to auto-restart after a power failure allow for that, to avoid a sudden high load on your power system.

      2. John Brown (no body) Silver badge

        "The paper is observing that if the peak was lower, by having a little more randomness in the demand, there would be fewer problems supplying the overall demand as it would be spread over a greater period of time."

        That's true, but they are framing it like this is a revelation to the power companies. The power companies already know. They may not know precisely *why* it's happening, but they know it's happening and can account for it. On the other hand, as you say, it could be spread out more but that's really a public education thing, which the power companies might like to get involved in, but I suspect it would, in reality, end up with most[*] people thinking, "ah fuck it, I'll just add 'n' minutes to the timer". Most people will almost certainly add 1-5 mins, not actually making much of a difference. A real solution would be for "smart" home devices such as thermostats actually be smart and talk to others in the neighbourhood or the power company and negotiate a time around about when the user wants it on and come on sooner or later by some minutes.

        [*] "most" being of those who even give it a thought, that is. There'll be plenty who think it doesn't apply to them or don't care, or assume others will do something. And, being the USA, there'll be a certain percentage who say something along the lines of "Muh FREEDOM!!! BigGuv/BigCorp ain't telling ME when I can turn on muh heat!!!"

      3. MachDiamond Silver badge

        "The paper is observing that if the peak was lower, by having a little more randomness in the demand, there would be fewer problems supplying the overall demand as it would be spread over a greater period of time."

        I think that having EV's calculate when they need to be charged by and switch on at night during a reduced tariff period is the same sort of thinking rather than people just programming the time when the rates drop to begin the charge. Being able to send pricing along the lines could also allow the power company to vary off-peak tariff times so if there's nothing but reruns on that night and everybody has decided to go to bed early, EV's could start charging earlier and home batteries can recharge at reduced rates as well. It would be a good way to use power when it's abundant to level off demand when it isn't.

      4. David Hicklin

        "we didn't worry about bringing another coal fired power station online for the ad breaks"

        That's why the UK built the pumped storage power stations: Idle to max power in under 30 seconds (according to the tour guide)

    3. Mike 137 Silver badge

      "why not rev up ready for it"

      Because supply capacity is not the only problem. A few days ago at tea time, smoke started pouring from a street manhole in our area. It turned out that the combination of an overload and very hot weather had caused the insulation of the final distribution cable to melt and the cable had shorted, blowing the substation fuse. Nobody had any power for almost four hours while they fixed it. I spoke to the repair crew and was told that the cable was 'fairly new' (1980s vintage). So there's going to be a huge amount of upgrading needed all over the entire country if we move exclusively to electricty as an energy source.

      1. Anonymous Coward
        Anonymous Coward

        Re: "why not rev up ready for it"

        A few superhot days in a row and every cable will start melting everywhere all at once :D

        1. jmch Silver badge
          Trollface

          Re: "why not rev up ready for it"

          "A few superhot days in a row..."

          You talking about the UK or Florida??

          1. J. Cook Silver badge
            Joke

            Re: "why not rev up ready for it"

            both. And Arizona. And Texas.

          2. John Brown (no body) Silver badge

            Re: "why not rev up ready for it"

            "You talking about the UK or Florida??"

            All this panic over a few warm days in the UK, LOL Must be the kids, too young to remember 1976 :-)

          3. cray74

            Re: "why not rev up ready for it"

            You talking about the UK or Florida??

            UK or Texas? Texas can't handle cold or heat .

            Florida can handle a long run of hot days. Despite the abject stupidity in the Crystal River nuclear plant shutdown and the recent utility push to undermine solar panel owners, Florida has enough gas- and coal plants to meet extended summer demands. The grid is also relatively robust unlike, say, Texas and its experiment in deregulation. High CO2 and now less renewable friendly, but stable.

      2. Toe Knee

        Re: "why not rev up ready for it"

        @Mike 137

        I work in utilities, and underground infrastructure from the 80s is an ABSOLUTE NIGHTMARE. I have a hunch that “deregulation” played into it by ensuring the profit motive overrode any other consideration in the building stage. Only one man’s opinion, but a reasonable one.

        1. Neil Barnes Silver badge
          Holmes

          Re: "why not rev up ready for it"

          the profit motive overrode any other consideration

          Isn't that why the companies are in business? Surely no-one starts a company thinking 'wow, this is an amazing public good I can do' and even if they did, shareholders would be baying for profit in no time...

          There are very good reasons for regulation, and one of them is 'do it bloody right first time or fix it expensively later'.

          1. EricB123 Bronze badge

            Re: "why not rev up ready for it"

            Sort of like the century old PG&E wire hooks that wore out, starting a fire in California that killed around 84 people?

          2. Claverhouse Silver badge

            Re: "why not rev up ready for it"

            To be sure, but every Revolution --- Wars of the Three Kingdoms, American Independence, French + Napoleon, Bolshevik, Nazi to mention only a few --- is funded by extensive looted expropriation and continuous taxes to follow. The Nazis in particular, and the Fall of the Soviet Empire were inclusive of vat privatisations.

            .

            In that way, any privatisation can only then lower costs to consumers, by

            a/ paying lower wages

            b/ lowering standards, especially safety and increased work hours

            b/ paying lower costs for raw and other goods

            c/ paying amounts lesser than expected to executives and stockholders

            .

            It is up to the management of the liberated companies to decide which are its priorities.

          3. jmch Silver badge

            Re: "why not rev up ready for it"

            "There are very good reasons for regulation, and one of them is 'do it bloody right first time or fix it expensively later'."

            Yes. Problem is, sometimes 'do it bloody right first time' isn't enforced and by the time it comes to 'fix it expensively later'*, the original company isn't around anymore and the taxpayers have to foot the bill (again).

            *in many infrastructure projects 'later' could be 10, 20 years++ later for stuff that's supposed to last 40-50 years++

        2. hoola Silver badge

          Re: "why not rev up ready for it"

          Aluminium telephone wires anyone?

          There are many estates of the same vintage where these were used because copper was so expensive. Teh real problems come where they are crimped to copper.

          1. ITMA Bronze badge

            Re: "why not rev up ready for it"

            Aluminium phone cables have far more problems than that.

            Ask any BT Openreach engineer. BT tried this out a number of years ago and current BT engineers see them as "a bloody nightmare".

            Like that utterly shite CCA network cabling (allegedly Cat5e/Cat6) - CCA, Copper Clad Aluminum. Look at it the wrong way and it breaks.

            I have had cabling companies try to install that - because it is cheaper - and didn't like being told to rip it all out and replace it with pure copper.

        3. Roland6 Silver badge

          Re: "why not rev up ready for it"

          It was probably more accurately the rip the government and customer off motive.

          With the UK utilities there was this perception that regulations were excessive aka gold plated. Hence why the Tories (and the non-thinkers) bang on about reducing the burden of red tape and regulation, yet are unable to readily point to any gold-plated regulation...

          However, reducing regulation (and inspection) causes problems; one of the well reported big infrastructure concerns in the US is the number of bridges that have been built to minimum "standards" and are now potentially coming to an early end-of-life. However due to lack of regulation and inspection it is not known which bridges and thus how many are at risk of collapse...

    4. Anonymous Coward
      Anonymous Coward

      I think the issue is that timers are more far more accurate than commercial breaks. As in: at 6:00:00 exactly the thermostats switch on simultaneously.

      That's something you'd think a Smart Meter would have been designed to avoid... but I always assumed that a Smart Meter was mostly there as an excuse to put the people who used to read meters onto welfare, rather than anything that would give me a benefit.

      1. Richard 12 Silver badge

        Because it's easy

        It's really easy to make a thermostat that raises the target temp (thus turning the load on) within a tenth of a second of the given time.

        And it's really easy to set the same default in every single one you make.

        The fact that's stupid is irrelevant. These things are built to a cost. Adding randomness and ramps to smooth the load change over a few minutes would cost more.

        The costs of deliberately synchronising all your customer loads to within a second don't fall on you. They fall on the utility company. And the consequences fall on your customers and their neighbours.

        This is simply another example of why regulations are absolutely required.

        The "free market" will always be destructive when permitted to externalise the costs and consequences.

        1. Lon24 Silver badge

          Re: Because it's easy

          And so dumb it is almost inconceivable that the software engineers who set these up didn't consider the inevitable system effect and add in a little randomisation.

          Thank goodness Linux folks aren't that stupid. Anybody running a fleet of servers (well Debian anyway) will have noted that the daily crontabs are all set at slightly different times by default so spreading the load of any interactions.

          1. Jimmy2Cows Silver badge

            Re: Because it's easy

            Why is it the software engineers' fault? They will have written the code to the specs provided for them. Chances are they will have been contracted at the lowest rate possible, and just told "make this work". It's unlikely any of them would be utility scale electrical engineers, and so is unlikely they would have any awareness of the problems that could arise.

            If anything it is the ultilities' fault for not considering the potential effects on energy generation and the grid, which can probably be put down to two things:

            1) The wrong people were involved in planning these things. Suspect all management consultants and MBAs who only care about the business side - their business side. Nary an actual engineer among them. There was a lot of free money floating around (in the sense that it would come out of customers' bills later, so who cares how good it actually is).

            2) Misguided adherence to "agile" practices without understanding what the fuck that actually means, and so doing a half-arsed job and not listening to any concerns that might be raised along the way, as that would interfere with their delivery timeline and so wouldn't be "agile". In other words a bunch of twats that follow buzzwords and wouldn't know their arse from their elbow.

            1. Roland6 Silver badge

              Re: Because it's easy

              >If anything it is the ultilities' fault for not considering the potential effects on energy generation and the grid

              Do not see it as being the utility's fault vendors of IoT devices are unable to think and design accordingly. Although, I expect now it has been flagged, national grid will get the regulations changed so that IoT vendors have a higher bar...

              What has changed is the nature of the problem, many manufacturing plants, after a full shutdown, have to call the grid so as to get additional supply necessary to take the restart surge ie. small number of users with big loads. IoT is a large number of users with a relatively small load, that combines to produce a large load.

        2. TRT Silver badge

          Re: Because it's easy

          Makes me wonder about smart street lighting too.

          1. Jimmy2Cows Silver badge

            Re: smart street lighting

            At least they will come on at differnet times relative to whatever time zone, due to different sunset/sunrise times at different longitudes and latitudes. A bit more spread out than many millions of smart thermostats in the same time zone all kicking in at the exact same moment.

            Perhaps. There are a lot of street lights. Hmm.

            Though I think some have a solar battery which charges during the day, so the lamp only draws mains power when the battery dies, which will vary from lamp to lamp.

            1. TRT Silver badge

              Re: smart street lighting

              Ah! The OLD smart street lamps used to come on at dusk and go off at dawn and relied on photocells under varying degrees of transmissive glass or plastic, but the NEWER types are synchronised by local mesh networks and time synchronised to a single signal to turn on or off at set times. The come on at dusk, say, and go off at midnight before coming on again at 6am if it's still dark.

              The old ones you could watch the lights come on a few at a time over the space of around 5 to 10 minutes - the new ones just all come on or go off at the same time.

    5. teknopaul Silver badge

      Because it cost money to manage peaks and, since meters are "smart", it would be cheaper to have the smart meters smarter.

      Jenkins cron jobs support "h" so that lots of jobs don't start at exactly :00 seconds past the hour. H is a hash that gives you predictability per job without precision.

      Seems like a simple software fix.

      It's about time we started to recognise the environmental damage done by people who get out of bed on time every morning.

      Bob says, be good, be slack.

    6. Roland6 Silver badge

      The laugh about this is that this phenomenon is well known in systems operation circles; In setting up batch jobs, we always set things to run at a minute or so after the hour, half-hour etc. in an attempt to avoid on the hour (ie. nn:00:00.000) blips in system load and thus general system slowdown.

      So it is a little surprising it has taken so long for this to be recognised as being a problem with IoT stuff with very simple clock behaviour, yet highly accurate clocks - so the probably of everything in my house coming on at the same real-world hh:00:00.000. is increased as is the probably of a large number of other peoples home devices also switching on in the real-world at hh:00:00.000.

      Interestingly, the solution would seem to be a local hub with the intelligence to spread start times and communicate demand requirement to the supply network.

    7. Missing Semicolon Silver badge

      why not rev up ready for it, as has always been done for advertising breaks

      Because there is simply no dispatchable generating capacity. In future, such peaks will simply result in a power cut.

  3. elsergiovolador Silver badge

    Smart?

    These thermostats aren't "smart" they just try to do whatever you program them to do. That's not smart.

    That being said, just watch when everyone is going to plug their electric cars. Ah they won't! Nobody will be able to afford these bloody things.

    1. SkippyBing

      Re: Smart?

      I'd say they're smartish. Mine didn't require programming, I just set the temperature to what I wanted it to be for the first few weeks and it learnt what I liked when. It's certainly smarter than the 7 day timer I had before as it doesn't bother heating the house when no-one is in it.

    2. Anonymous Coward
      Anonymous Coward

      Re: Smart?

      Many years ago there was an expensive central heating system that included external temperature sensors. When installed it would gradually build-up a prediction of how each room would warm up depending on its thermal properties and the actual weather conditions.

      A thermostat setting was the target room temperature you wanted for a particular time of day. The system then varied the necessary advance heating each time depending on all the variables. It even adjusted a room offset to allow for slightly higher temperatures in the evening when people were likely to be sedentary.

  4. Jellied Eel Silver badge

    Pope is catholic, bears defecate in woods..

    And there's nothing new under the sun. Other than 'experts' wanting to take control away from customers. So here's a thing-

    https://www.schnap.com.au/timers/mechanical-timer.html

    That people of a certain age are probably familiar with, because it was the standard kind of timer control used pre-'Smartbollox'. But..

    ...show a large performance gap between potential energy savings and actual energy savings," Zhang and Lee wrote.

    Easily demonstrated just by looking at your electricity bill and wondering why, if energy supply has gotten 'smarter' and 'greener', it's also got a LOT more expensive. Hence why there's so much focus on demand management, because the supply side is a complete shambles due to regulatory capture by the ecofreaks.

    But sure, let's throw a few billion at some new grid management company that could randomise control over people's thermostats, or energy in general. Alternatively, politicians could start asking why the 'renewables' lobby has failed to deliver any cost savings, or reliability.. All it's done over the last couple of decades is increase costs, complexity and reduce reliability.

    It'll be interesting to see if the Texas electricity grid collapses. Having 'renewables' responsible for power failures both winter and summer would be quite the achievement. Especially as Texas is considered ordering it's industry to curtail demand.. ie stop producing fuels and stuff, so increasing shortages, prices, and inflation.

    1. Alan Brown Silver badge

      Re: Pope is catholic, bears defecate in woods..

      "the supply side is a complete shambles due to regulatory capture by the ecofreaks."

      The amount of money being poured into "renewables" would make a VERY big hole in the carbon problem if it had gone into molten salt nuclear reactor R&D instead (and then building MSRs)

      I tend to agree with Lester (RIP) that once the bugs are ironed out and MSR+thorium (LFTR) tech started being widely deployed, nuclear power has a good chance of being the "cheap solution" we were expecting in the 1950s, possibly to the point of rendering wind+solar redundant+expensive boondoggles

      1. Jellied Eel Silver badge

        Re: Pope is catholic, bears defecate in woods..

        Yup. Sadly we have neo-luddites deciding our energy policy. When they're not supergluing themselves to stuff, presumbably not realising the glue is a product of their hated 'big oil'.

        T'other advantage of LFTRs is the can recycle old nuclear waste, which would remove the stock whine about waste being deadly for hundreds of thousands of years.

        Downside is I think despite there being a shedload of thorium sitting in mine tailing piles around the world, it's still considered radioactive waste rather than valuable fuel.

      2. skierpage

        Re: nuclear counterfactuals are irrelevant

        "The amount of money being poured into "renewables" would make a VERY big hole in the carbon problem if it had gone into molten salt nuclear reactor R&D instead (and then building MSRs)"

        Renewables have reduced carbon burning. Obviously, when the wind is blowing or the sun is shining, fewer tons of fossil fuel are shoveled into thermal plants. The percentage of electricity generation by coal has gone down, the percentage of electricity generated by renewables has gone up. In the USA gas plants still generate the majority of electricity, But since renewables are by far the majority of new generation, gas will drop too. Yeah yeah we all know renewables are intermittent, and gas plants continue to step in when renewables aren't generating. But renewables plus storage are becoming more popular, so the gas plants will fire up less.

        It's no use crying over spilt milk "we could have built nuclear like crazy and it would have (somehow) got cheap and we would have no worryies about intermittency." It didn't happen! In the world we live in, new nuclear produces expensive electricity 24 hours a day, and for much of that time it's undercut by renewables. It's not surprising electric utilities aren't interested. Nuscale is now saying it will make hydrogen at times when its electricity isn't wanted, which raises the question why not plug some electrolyzers into the renewable grid to intermittently make hydrogen, without the expensive nuclear plant.

    2. codejunky Silver badge

      Re: Pope is catholic, bears defecate in woods..

      @Jellied Eel

      It sounds like Germany is having a seriously rough time with the 'energy issues' (bad policy) at the moment. I just hope the rest of the world wakes up and starts trying to undo the damage of the green madness. Not likely and wont be cheap.

      1. Jellied Eel Silver badge

        Re: Pope is catholic, bears defecate in woods..

        Sadly, the EU doesn't really show much signs of changing policy to become pro-nuclear. Mainly because a lot of those politicians rely on Green votes, or support in their coalition governments. So currently the EU is still proposing wasting a few trillion more on 'renewables'.

        It would be far, far better for everyone if 'renewables' were cancelled and money was invested in LFTRs, SMRs or even just improvements in coal power.

        1. jmch Silver badge

          Re: Pope is catholic, bears defecate in woods..

          EU recently introducing some subsidy or other for low-carbon energy sources. Greens are protesting against proposed amendments that include nuclear and gas as such, since that would promote use of both. I can't fathom why they would lump the two together. Gas is convenient AND produces less CO2 per energy than oil/coal BUT it still produces CO2 and the main source in the EU is Russia, so I completely understand wanting to shut down any subsidies to it, even if for geopolitical rather than environmental reasons.

          But the EU should really be subsidising nuclear as much as any other renewable.

      2. Anonymous Coward
        Anonymous Coward

        Re: Pope is catholic, bears defecate in woods..

        Tufton Street Calling

    3. jmch Silver badge

      Re: Pope is catholic, bears defecate in woods..

      "Easily demonstrated just by looking at your electricity bill and wondering why, if energy supply has gotten 'smarter' and 'greener', it's also got a LOT more expensive. "

      'Smarter' and 'greener', even in the theory of smart metering, is that less units are consumed. It says nothing about cost. Having said that, as long as you can still heat your home comfortably, consuming N units at £2 per unit is better than consuming 2N units at £1 per unit - same cost but still less energy use.

      "It'll be interesting to see if the Texas electricity grid collapses."

      Not sure exactly what the Texas situation is, but while relying heavily on wind in the winter is problematic, sunshine statewide during the summer is extremely strong ( they're on the same latitude as North Africa!!) And unlike wind, it's much more predictable - sunrise/sunset times are known to the minute and cloud cover can usually be forecast relatively accurately a day or two in advance.

      Of course as you say they really do need to upgrade their grid!!

      1. codejunky Silver badge

        Re: Pope is catholic, bears defecate in woods..

        @jmch

        "'Smarter' and 'greener', even in the theory of smart metering, is that less units are consumed. It says nothing about cost"

        I am in the UK and the whole effort of selling this garbage is to lower your bills. Smart meters are sold on the concept of saving money. Hell do a quick google search and you will find in the google summary before even entering a page how it can save you money.

        It is the same lies that sold wind farms and solar as saving money, do you remember people talking about all this free energy? The shift in generation which has of course caused the energy bills to shoot up and sup[ply to be threatened in various developed countries.

    4. Spazturtle

      Re: Pope is catholic, bears defecate in woods..

      It was the gas network failing that causes the Texas electricity grid to collapse in the winter, all the gas power plants were sitting idle due to the gas network not being able to operate below a certain temperature. If you want a robust power gird then you need to use nuclear for your baseload.

      1. Tom66

        Re: Pope is catholic, bears defecate in woods..

        Or have batteries and diesel generators standing by to 'kick-start' the grid when things go wrong.

      2. Jellied Eel Silver badge

        Re: Pope is catholic, bears defecate in woods..

        It was the gas network failing that causes the Texas electricity grid to collapse in the winter,

        It was a bit more complicated. Pre-Obama, the Texas gas network used to use, well, gas to generate electricity to power the gas pumping/distribution network. Burning gas is un-Green, so politics forced them to stop doing that, and use 'clean' energy instead. So then when the wind (and temperatures) dropped, so did the electricity supply, gas couldn't be pumped, and Texas narrowly avoided it's grid collapsing and a 'black start'.

        Currently the problem's much the same, and same as UK, ie high pressure, high temperatures, and so not much wind-

        https://gridwatch.co.uk/Wind

        minimum: 0.543 GW maximum: 12.751 GW average: 5.115 GW

        for the month to date.

        1. khjohansen

          Re: Re: Pope is catholic, bears defecate in woods..

          "Pre-Obama"

          Governors Rick Perry + Greg Abbott has overseen an extensive DE-REGULATION

          - the Texas powergrid is NOT subject to federal oversight

          https://cz.boell.org/en/2021/03/19/texas-power-grid-failure-causes-lessons-learned-and-implications-eu-energy-market

  5. HCV

    Click

    PG&E creates incentives for subscribers to limit power usage from 4 to 9 PM, and I'd been wondering what was happening at 9 PM when everyone's air conditioners click on in synchrony. Now I know.

    1. Tom66

      Re: Click

      At 9pm the load for air con will be less, because the outside temperature will be lower. So this is still a good idea. Also encourages EV charging, cooking, and any other high power stuff outside of peak times if possible.

      1. Peter2 Silver badge

        Re: Click

        Also encourages EV charging, cooking, and any other high power stuff outside of peak times if possible.

        Because naturally everybody is willing to wait until after 9PM to start cooking their dinner.

        1. Tom66

          Re: Click

          "if possible"

        2. DuncanLarge Silver badge

          Re: Click

          > Because naturally everybody is willing to wait until after 9PM to start cooking their dinner

          9pm wow that is early. I sometimes manage that.

          Most of the time it is 10 or 10:30pm before I'm even preheating the oven.

  6. VoiceOfTruth

    So a "smart" thermostat

    is a time switch.

    I suppose today that a Teasmade would be labelled a "smart hot drink machine". Except it would need a cloud connection and a recurring subscription fee, else you don't get any more tea or time service.

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

    1. Wally Dug

      Re: So a "smart" thermostat

      Doesn't a Teasmade generate steam and therefore its own cloud?

      1. Anonymous Coward
        Anonymous Coward

        Re: So a "smart" thermostat

        Technically, if you can see a cloud, it is not steam, but condensation. Steam is invisible. Superheated high pressure steam is invisible and deadly.

        1. Toe Knee

          Re: So a "smart" thermostat

          A Steam Machine, if you will? Valve is in the tea business now, apparently :)

          1. ITMA Bronze badge
            Devil

            Re: So a "smart" thermostat

            Probably get sued by a certain games company for copyright/trademark infringement

  7. Peter Galbavy

    I think Economy 7 "solved" this problem decades ago. Surely they can't be that dumb... no, wait, yes they can.

    1. Anonymous Coward
      Anonymous Coward

      They still have this in France. Signalling (down the main) is used. And set so that everyone in the same neighbourhood/region doesn't have their water heater (or whatever else you want) switch on at exactly the same time. It is not rocket science.

  8. Lost in Cyberspace

    After looking at all the smart thermostats...

    I replaced mine with a dumb on that had 7-day schedules and -finally- separate controls for hot water and heating. Common sense applied to the schedule.

    Saved me plenty and no downtime when a service goes offline.

    1. Neil Barnes Silver badge
      Flame

      Re: After looking at all the smart thermostats...

      Same here: I had a seven day thermostat for years at my last house (hot water was on-demand from a combiboiler, so independent of whether the heating was on).

      Maybe twice a year I would come home to a cold house after a holiday in winter, and what do you know, in a few hours it was warm again. Amazing!

      1. 42656e4d203239 Bronze badge
        Flame

        My 7 day, not IoT, thermostat has 6 heating periods during the day with independent set points and it also has a holiday mode - I tell it when I am back from holiday - date and time (to the minute, if I can be bothered), hit the go button and lo and behold - the heating & hot water stay off for the duration and pop back on an hour or two before I get back.

        Why do I need an IoT thing to control that? I don't - even if I were delayed the actual difference in the yearly bill wouldn't be that significant. In so many instances "Smart" means "Ill thought out, waste of time and money"

    2. jmch Silver badge

      Re: After looking at all the smart thermostats...

      Absolutely no reason why a 'smart' thermostat should be connected to the internet / cloud.

      I understand having a cleaner interface - one of the places I lived in had an individual control for every room, on/off at certain times depending on the day of the week, 'vacation' low or off settings etc, but even setting it up once and forever was annoyingly tedious.

      Is it beyond the wit of man to have a system that can do all that, with a usable interface??

      1. SkippyBing

        Re: After looking at all the smart thermostats...

        Connecting it to the internet lets you check the settings while you're out. It's amazing how cold a teenager can get with the windows open, wearing only a t-shirt, in winter. Once they learn where the thermostat is you need to make sure you can turn it back off before they bankrupt you.

        With the app on your phone it can also tell if you're on the way home and pre-warm the house (which was handy when I had an erratic schedule). Not vital but it did stop me heating an empty building and make me feel like I was living in the 21st century, not the 1900s like some luddite who had to turn things on themselves.

  9. Anonymous Coward
    Anonymous Coward

    Nothing new - just a bigger scale

    About 30 years ago, a client of mine decided to stop staff (about 10 people) adjusting the controls on the panel heaters in the open-plan office. Something needed to be done as people didn't really understand even the simple controls there and would turn the thermostats up in an attempt to warm the office quicker (which, whilst not shortening the time taken to get to a comfortable temperature, had the effect of ensuring the heating overshot and windows would have to be opened to cool it down). Also, heating would be left on over the weekend. The manager's solution was to seal all the heater controls at a the right temperature setting and get a timeswitch installed on the main supply board. That way, heating would be off over the weekend and would come on in time to heat the office on Monday morning.

    Except it didn't, as the effect of all the heaters coming on at the same time caused a demand surge that immediately blew the main breaker!

    1. SImon Hobson Silver badge

      Re: Nothing new - just a bigger scale

      ... as people didn't really understand even the simple controls there and would turn the thermostats up in an attempt to warm the office quicker ...

      Yup, been there. In a previous job the office heating etc. came under my remit. At one time, there was this office with a quite decent reversible aircon unit - quite a high capacity for the space. Regardless of how often I (carefully and slowly) told people the basic principle of a T H E R M O S T A T, it would rarely go more than a day or two before someone fiddled. The moment anyone was either a little too hot or a little too cold - which is actually all the time, and both at once if you have an office full of women (sorry ladies, just an observation) - it would get fiddled with.

      Except, whacking the controls up to full, or down to min does affect the rate of heat-up or cool-down when you've a big A/C unit with variable speeds fans etc. So it would always end up as : someone's a bit cold in the morning, so set it to 30˚C; not long after, everyone's too hot so it gets turned off; then by early afternoon it's hotter still so turned on and set to 18˚C; then when everyone's suitable frozen it gets turned off again. But of course, having been turned off overnight instead of left on it's timer, the office will be cold in the morning meaning that the 30˚C period starts as soon as the first person gets in.

      And then they moan at me when the unit's set at 30˚C and they're hot; or they're cold when it's set at 18˚C.

      When we got the whole-building, chilled water system in, I arranged for electronic controls with no user access. Amazingly nearly everyone accepted that.

      1. Dimmer

        Re: Nothing new - just a bigger scale

        I have a kin that installs units. A recent customer requested a thermostat for each office. The customer did not want it connected to anything, just make it appear it was working.

        Solved the problem with half hot and the other cold.

        Now if it would just fix the problem of the heaters under the desk plugged into the ups...,,

      2. Peter2 Silver badge
  10. Duncan Macdonald Silver badge

    Randomized time offset ?

    New overnight EV chargers in the UK are required to have a random offset from the specified start time to reduce the surge - it is a lot easier to cope with a demand that ramps up over 10 minutes than one that ramps up over 1 second.

    "Smart" thermostats could employ the same feature. (Old fashioned mechanical time switches provided this feature by the inaccuracy of users setting the time.)

    1. Jellied Eel Silver badge

      Re: Randomized time offset ?

      New overnight EV chargers in the UK are required to have a random offset from the specified start time to reduce the surge

      Or create surges. Fun with GPS and NTP spoofing. Required gadgets may currently be found abandoned in Ukraine, or for sale on a DarkWeb near you..

    2. The man with a spanner

      Re: Randomized time offset ?

      Smart thingies rellying on time (especially synced viathe internet) are an obvious problem.

      Answers:

      Force people to set it themselves (we all have different requirements). Sometimes the right default is no default.

      Random factor added +/- 5 mins

      Modify start up if temp in a range

      Any more options

    3. Solviva Bronze badge

      Re: Randomized time offset ?

      Random or pseudo-random? Do they all have the same seed.....

      1. Duncan Macdonald Silver badge

        Re: Randomized time offset ?

        The EV chargers are supposed to have independent offsets - two easy methods would be to either use the network address or the number of operating hours as the seed for the pseudo random number generator.

        1. Richard 12 Silver badge

          Re: Randomized time offset ?

          Honestly, capturing the number of seconds past the hour when first turned on after initial installation would very likely be sufficient for retrofitted units.

          Or just ask the user to push a button each time they plug it in.

          Humans are a pretty good source of entropy.

          That said, pretty much everything bigger than an M4 class microcontroller has a built-in hardware entropy source. And the smaller ones have ADCs which can be used.

      2. Tom66

        Re: Randomized time offset ?

        EV chargers use a PWM signal to set the charge current from the car. The duty cycle of this will vary somewhat randomly in a small region, that's more than sufficient entropy. Or you could use the 50Hz carrier counts from last power off, or the variation in charging current, or plug temperature or all sorts of other things to get entropy.

        All you need is say 5 bits of true entropy to give a delay of 0-15.5 minutes and you've probably got enough of a distribution to make it work.

    4. rhcp

      Re: Randomized time offset ?

      Why not have the thermostats aim for the target temperature at the target time. Then the turn-on time would be dependent on the local conditions (temperature, set temp and room size) & would be randomised.

      A real 'smart' thermostat should be able to handle this with minimal learning cycles.

      1. SkippyBing

        Re: Randomized time offset ?

        Nest definitely does that, my boiler kicks in earlier the colder it is to hit the target temperature at whatever time it thinks I want it. Second order effect was to make me get more loft insulation as I was getting woken up too early as it fired up in December!!

        1. DomDF

          Re: Randomized time offset ?

          My Honeywell (or whoever they are now) does this too. It's not exactly what I'd call a "smart" feature.

  11. Nate Amsden Silver badge

    fully manual

    I learned the central AC in my apartment should only be run fully manually. Allowing the thermostat to keep temperature increases the chances the AC will fail. A few years ago I had a lot of problems with it and maintenance couldn't find the cause(they replaced some parts to no avail, I gave up trying to get it fixed). The AC compressor would fail to kick in(not an expert just going by the lack of noise), and instead ice would build up inside. It would then take many hours(sometimes waited 24h) for the ice to melt before the AC would work again. I'd estimate there is a 5% chance of this failure happening on any given startup. Turning it on manually every time I can tell when this failure happens(if the compressor doesn't kick on within 20 seconds), then I turn it off, wait a bit and try again. When the temp gets to a level I'm comfortable with(I have many temperature sensors in different places) I turn the AC off again. I never use the heat function. I think this year I have caught it fail to start 3-5 times.

    But to these smart thermostats, I suspect it's likely if there were no defaults the customers would set them similar to what they are set to now. It's clear the $$ savings isn't worth it to them, otherwise they'd change it themselves. Some power providers have incentive based electricity plans where the price can vary depending on time of day or whatever but I think most people just have the more basic plan that charges based on overall usage (probably in tiers) for the billing period.

    1. SImon Hobson Silver badge

      Re: fully manual

      You won't get ice build up if the compressor fails to run. More likely the unit is icing up, and then the compressor trips (or in a modern unit, just runs slowly) due to lack of load.

      The icing up will be due to one of two things : poor control system (any half decent modern system has automatic de-icing these days, but some cheaper ones won't); or wrong choice of system for the conditions.

      As to conditions, if your air is drier than the system is designed for, then it will ice up - at which point I suspect a few people will be scratching their heads and having a WTF? moment at the suggestion. But I'll explain ...

      When in cooling mode (you can switch everything around for heating mode), liquid refrigerant is passed into the avaporator coil in your indoor unit - via a valve or capillary tube. As it passes at a controlled rate into a space at lower pressure, it evaporates - and thus takes in it's latent heat. To supply this latent heat, heat needs to be conducted in from the air being passed over the coil by the fan - thus cooling the air. As I think most people realise, as you cool air, water will condense out of it once you hit the dewpoint which depends on the humidity.

      Now, the heat taken out of the air comes from two effects. The first is simple cooling of the mass of air, the other is from the latent heat as water condenses out. If the air is dry then the first is the main mechanism, if the air is very wet then the latter is dominant. So with wet air, a lot of the heat comes from condensing the water out - and so the off-coil temperature (i.e. the temperature of the air as it leaves the heat exchanger) is significantly higher than with dry air where the air must be cooled to a much lower temperature to extract the same amount of heat. And as a result, the surface of the metal pipes in the heat exchanger is higher with wet air than with dry air. And so with very dry air, the metal surface is colder, and you are more likely to have what water does condense out freeze in place rather than drain out.

      A good control system will detect the significant drop in temperature (which affects the refrigerant vapour pressure) and invoke de-icing - in the first place stop the compressor and keep the fan running so that the ambient air will defrost it. Or with a reversible system, it can reverse the heat flow and actively heat the evaporator coil to thaw it out. But many cheaper system will just keep pumping away regardless and the evaporator coil will get more and more iced up (and the ice even more supercooled) until eventually no air can flow. Ideally you need to be able to run the fans without the compressor so that any air that can get through will fairly quickly warm up the coil and thaw out the ice.

      The last time I had to explain this was to a friend who had responsibility for his employer's server room. They had A/C installed, it didn't work well (kept freezing up), and they were "sceptical" of the installer's suggestion that the air was too dry - in reality they'd been sold a completely unsuitable system as it couldn't cope with dry air.

      1. M.V. Lipvig Silver badge

        Re: fully manual

        You forgot one other problem that is far more likely than all that for a residential system - a tiny leak. An air conditioner will also freeze up when the freon* is low. Maintenance guy fiddles with it, tops it up, tiny leak keeps leaking, then it freezes up again. I just went through this on my home system, had to replace the evaporator as the leak was very tiny, and apparently there.

        *I know, but the word freon is like the word Coke. "Hey, grab me a Cole. What kind? I dunno, orange I guess." Univeraal, not quite correct, but everyone knows what you mean.

        1. SImon Hobson Silver badge

          Re: fully manual

          Hmm, I'm trying to think of the mechanism for that being an explanation.

          My experience when systems have lost some gas is that they reduce in cooling ability - but don't freeze up (seen it several times). In normal operation, hot compressed gas enters the condenser, is cooled, and at some point it starts condensing (at which point, it doesn't change temperature any more unless part of the condenser contains only liquid, where it can cool some more).

          As refrigerant is lost, the quantity of liquid will reduce - and initially little will change WRT performance unless the system is already up against its limit of condensing gas due to the outdoor temperature. There would be a slight reduction in the part of the condenser which only contains liquid, and that would only make a small difference to the temperature of the liquid line back to the indoor unit. As long as there is still some liquid phase, then the system pressures would be barely affected.

          At some point, there will no longer be enough refrigerant, and the liquid line will start to contain a mix of liquid and vapour. Here's the clever bit (one of the few bits I recall from doing this at university [cough] decades ago), while the volume flow rate through an orifice is higher for vapour than for liquid, the mass flow rate is lower. Thus a simple orifice or capillary tube can be self regulating - if there's vapour in the line, the mass flow rate reduces, so allowing the compressor to "catch up" (if it has the gas to do so).

          So the mass flow rate has reduced, therefore so will the heat input needed to evaporate the liquid also reduce - hence the system will suck less heat out of the air, and the coil should run warmer.

          [lightbulb moment]And now I see a mechanism for what you've observed. If the system is an old style on-off constant speed compressor, it may (depends on compressor design) draw down a lower pressure on the vapour side of the system. At a lower pressure, the liquid will boil off at a lower temperature, and part of the evaporator coil will therefore be at a lower temperature.

          How that manifests will depend heavily on the design of the coil. In some designs it will allow that part to freeze up with the result that the refrigerant will need to move a little further along to boil off - and so the coil will slowly freeze up from one end. In other designs, while a little bit may freeze up, the already boiled off (and warmed up elsewhere in the coil) vapour will come back along a different loop and control the freezing.

          A modern inverter design shouldn't have this problem at all as the compressor should run slower to maintain the same low pressure as the mass flow rate reduces - hence no change in the boiling point of the refrigerant.

          Doesn't change the fact that fi the compressor fails to run, there's no way the system is going to freeze up because of that - it's far more likely that the compressor not running is a result of whatever else has caused the system to ice up.

          .

          I might add that A/C service engineers are (like service engineers in any other field) not averse to providing "interesting" ideas for why something isn't working. I once had a call from a client at a previous employer where the A/C in their server room wasn't working properly - it had all the symptoms of a lack of gas (turned on, didn't really cool much, then tripped). Bear in mind that it had been working fine for 2-3 years before the service engineer told the customer that "it can't possibly work in that room because it's not insulated and the heat input through the ceiling is over-powering the unit". There was actually insulation in the ceiling (I'd had to go through it when running network cables), and they'd been out to e builder's merchant and bought a roll of glass fibre to add to it. After a few days of this "it's not working, it used to work", "it can't possibly work because ..." to and fro I called the company and spoke to their service manager. I described the setup, the fact that it had been working for a couple of years, and the symptoms - and he instantly agreed that the diagnosis "didn't sound right". An hour later he called back to say they'd found the problem (a stuck reversing valve) and fixed it.

    2. Richard Jones 1

      Re: fully manual

      That has not been my experience. Our system has been running for about 6 years when I guiltily decided to try to get it serviced for the first time*. It cools in summer and heats in winter, and is left to its own devices. It did ice up once, surprisingly during a cold, almost 100% humidity day. A brief rest and a wash down of the heat exchanger and 30 minutes later it was back to life again. The airflow might have been affected by the clutter of air carried muck, leaves, insects, dust, etc. that had stuck to the heat exchanger.

      *Three different dates were agreed with three different service suppliers, only the third came along, the others just disappeared.

  12. Anonymous Coward
    Anonymous Coward

    I bet they're all getting time from the same time server and have clocks synchronised to a few ms too :D

  13. DS999 Silver badge

    "Smart" stuff with an internet connection

    Keeps accurate time. Dumb appliances that require the time will be off by some amount so even if everyone sets it for the same time it won't be a problem.

    They could probably fix this by having the default for that "6am" wakeup, "5pm" home from work etc. include a 15 minute random window either direction when it switches on. If you configure it off the defaults you could reduce that to a 2 minute window but not have a time exact to the second when it switches on.

    That way the grid would be not be hit by a million AC units turning on at the same second.

  14. Kevin McMurtrie Silver badge

    New York City winter

    NYC workers really do wake up at 6 AM in the middle of winter. That's why I'm in Silicon Valley, where the coffee maker can run on solar power.

  15. Anonymous Coward
    Anonymous Coward

    Oh! the joys of a forty-year-old timer and thermostat set to switch on the oil-fired central heating about half an hour before I even think about getting up. And a nice reverse cycle air conditioner in the bedroom, to boost the heat or chill me down as required. What's smarter?

  16. Muscleguy Silver badge

    No choice

    I start work at 08:15 meaning I rise at 06:00 so my heating comes on just before in the colder months. I do set my water heating for the early hours and early afternoon to ease the burden mind. I have an air source heat pump system. Installed last August so feeling a little smug though my elderly ailing back boiler badly needed to go.

    My gas hob is gone for induction (which I love and must be lower energy) and so I lose my gas meter next week so I won’t have to pay the standing charge. I have no use for gas, apart from the sodastream bottle with a regulator I use when bottling etc my brews.

    I currently have 3gallons of a pilsner fermenting in my beer fridge at 11C. Achieved with an Inkbird smart. I plug the fridge into it, a heating pad under the brew bucket and the temperature sensor goes in the beer. There’s an app to set it up and monitor. I can brew cooling beers in the heat. I might do a lager next but the fridge will be occupied for a while, after fermenting I will cold crash it slowly. Then back up to 11C for conditioning in Barrel and for bottles. This beer will NEVER be warm.

    1. GreyWolf
      Pirate

      Re: No choice

      When can we all come round for a tasting?

  17. Nifty Silver badge

    I have an idea. An alarm app that wakes me up at a random time each morning.

    1. Negative Charlie

      I have one that's synchronised to local sunrise, so I get the benefits of quiet sunny mornings in summer and toasty-warm lie-ins in winter.

      It's called leaving the curtains open.

      1. Neil Barnes Silver badge

        Uptick for you!

        I hate sleeping in a room with blackout curtains; my brain wants to get up when the sun does.

      2. LDS Silver badge

        Unluckily, that's very latitude-dependent.

    2. Dimmer

      Another reason to work from home. You get to sleep in till the sun rises or let’s just not start work till 10 and be home before 5

    3. TRT Silver badge

      I have a cat. Definitely a source of entropy.

  18. Anonymous Coward
    Anonymous Coward

    Call central control

    Just get the Chinese manufacturer’s to log in and set everyone’s thermostat to a random time.

    In between whatever nefarious task they are up too of course.

  19. ricardian

    Interesting articles about the frequency of the National Grid in Great Britain

    https://www.sciencedirect.com/science/article/pii/S2352484720301967

    https://www.drax.com/power-generation/maintaining-electricity-grid-stability-during-rapid-decarbonisation/

    1. Neil Barnes Silver badge

      I recall fifteen or twenty years ago that the grid suppliers were getting concerned about the increasing number of switch mode power supply units that shared a characteristic of all grabbing current at the same point in the AC cycle, but very briefly, and then turning off... leaving a hole in the AC waveform.

      Did they ever fix that?

      1. Solviva Bronze badge

        Yes, it's called power factor correction (PFC). Causes the SMPS to look like and behave as a nearly-pure resistive load which the grids like.

      2. Kevin McMurtrie Silver badge

        It's partly fixed.

        Big dimmer switches on tungsten lamps and junky CFLs beat up the waveform. Audio equipment and old electronics still use a 50/60Hz transformer, rectifiers, and a massive capacitor to make DC. It hammers the AC waveform down to a sharkfin shape and radiates enough EMI to make nearby metal hum. Audiophiles demand this for purity even when a switching power supply could be cleaner.

        Large DC power supplies have a boost inverter in front to draw power from the whole AC cycle with a power factor of nearly 1. I peeked at my home's heat pump repair manual and it converts 240V AC to unregulated DC around 320V using an IGBT and inductor as a power factor boost inverter. That DC can then be used to spin the compressor at any speed using more IGBTs.

        Small switching power supplies, like LED lights, use a buck/boost inverter with a moderately sized DC capacitor on the input. The power factor isn't perfect but it can use a good amount of the waveform.

      3. Anonymous Coward
        Anonymous Coward

        Mostly, the grid is reliant on the inertia behind it to resist this particular problem. Old school coal and nuke generators had lots of inertia. Gas and Windmills have some. Solar and interconnectors have none.

        You can routinely see the effects of the problem you describe between 11PM-1AM, while lots of street lights and/or domestic devices are being shut down at similar times. "See" in the sense that you can almost set your clock to when the lights start flickering. It isn't bad enough to be outside of the statutory frequency ranges the Grid is meant to operate. The increase in electrical noise is particularly noticeable to audio-geeks such as myself; and I ended up getting a UPS/rectifier to get a clean 50Hz signal. The UPS functionality is a nice-to-have extra.

        In Germany, some of their closed-down nuke plants were reverse engineered into Flywheels; spun up to provide additional inertia. Britain doesn't need these "today" but there is a engineering case for installing some IMO. FInancially, flywheels are unpopular because they won't make anyone any money to operate.

        Looking at the (totally predictable) response to the ESO's recent announcements of what the 2050 grid might look like I fear government-level intervention will be necessary to force through the relevant planning permission to build what we need.

        The East Anglia GREEN project and objections are particularly amusing, as they all insist on looking at Offshore as an option. I'm sorry, but the power has to come onshore somewhere to go to, you know, the demand. Bit bloody useless landing it on the shoreline and then not being able to move it inland because the capacity of the existing system is inadequate for the amount of generation expected to connect in those areas.

        A/C as I'm an employee in the sector.

        The plans of the networks to do what is necessary are about right; but the ability to execute in the face of NIMBYism is a massive risk to the rest of us.

        I am of the long term opinion the network is screwed; and you need to plan to be able to run off-grid because Britain can't do long-term infrastructure right any more.

  20. Binraider Silver badge

    Solar radiation impacting on an area is more or less a function of time of day and cloud cover. So, If the "ideal" air-con temp is 72deg. F, setting everything to 72 low and behold means lots of devices click on at the same time. Especially for similarly-constructed properties.

    Whereas if the split included proportions of users at 70, 71, 73 and 74, low and behold, I've just divided your number of simultaneous events by 5.

    From an Electricity POV the obvious solution would be to have flywheels and/or physical inertia in the power generator - something that with the scaleback of coal that is being reduced.

    Large windmills do exhibit useful inertia too (before the anti-green brigade jump on me), though of course solar has zero; DC-AC inverter has to track system frequency and aim at that.

    1. Anonymous Coward
      Anonymous Coward

      Wind turbines produce DC which is fed in to the grid via an inverter synchronised to the grid frequency, so they provide no inertia to the grid.

      1. Binraider Silver badge

        Consider yourself educated. Wind can and does provide an equivalent; albeit not as capable as a couple hundred runs of spinning AC gen.

        https://spectrum.ieee.org/can-synthetic-inertia-stabilize-power-grids

  21. An_Old_Dog Silver badge

    A little bit of randomization ...

    A co-worker made this sort of oops. Workstation anti-virus rollout via NAL to 10K+ PCs had all our PCs getting their AV updates from our one on-campus AV server at the same time, which slammed the server and made for loooong AV updates. I wrote a little script which periodically altered the container login scripts to in turn, alter a text file on the PCs which controlled when that PC hit the server for updated AV info. Ran it for a week and got the update times spread out, then everything was good.

  22. GreyWolf

    False claims in the frigging article

    Quote: "Electric cars create battery waste"

    Not true.

    Car batteries are 95% recyclable, and recovering the minerals from old batteries is much cheaper than mining them in the first place.

    Both Tesla and VW already have recycling factories, and more are on the way.

    Also: there are at least two other potential uses for car batteries. They will definitely have a second-hand value, so will not be dumped.

  23. JDX Gold badge

    Quite Interesting

    This is the same problem as "everyone puts the kettle on at half-time" in big sports fixtures, but worse because we're effectively programming our homes to launch a coordinated attack on the grid. Obvious in hindsight I suppose.

    Vendors could address with firmware updates, perhaps.

  24. Scene it all

    My thermostat is set at a constant temperature. As the Sun comes up and it starts to get warmer, the central AC starts GRADUALLY cooling the house. As the outside temperature gets warmer, the AC works harder. It is not a simple on/off unit, but variable speed - both the fans and the compressor are variable speed. So it does not make a sudden demand on the grid, it ramps up slowly, depending on the weather, not the clock.

    It did have an option to connect to the internet, but I told the installer to leave that off.

  25. Wesw94

    Global Warming in micro scale.

    There is always a delay between a set-point change and reaching the desired temperature. Ecobee and many other programmed thermostats "learn" an estimated value for the delay for a specific installation. The thermostat then uses thIs "learned" value to change the equipment set-point in advance so the desired temperature is reached at the desired time. Each installation will have a different learned delay. An oversized furnace will reach desired temperature far faster than a right-sized one, for example. Over a group, the individual equipment on-times will be a randomish distribution. So the postulated problem does not exist in reality.

    This is a little example of half-wit academic thinking.

    1. Anonymous Coward
      Anonymous Coward

      Re: Global Warming in micro scale.

      Must strongly disagree. ESO actively switch equipment and/or pay customers to turn on/off at particular times to manage issues with voltage control, frequency and active/reactive power.

      The studies behind these things are complex and most assuredly a half-wit academic would not survive more than 5 minutes in that environment, for system downtime is financially measured in the multiple-millions levels of pounds if it hits the wrong location.

      Not saying ESO get it right all the time, but the five-nines reliability we enjoy is more or less as good as it gets anywhere in the world.

  26. fishman

    Electric car rebalancing

    Electric cars will become the main type of vehicle in the future. To help the grid just control the charging of the cars - when the spikes occur reduce the charge rate on the cars.

    1. Binraider Silver badge

      Re: Electric car rebalancing

      All turning off at the same time is as bad as all turning on :-)

      Physics is a bitch, but y'cannae change the law!!

  27. nijam Silver badge

    Smart meters aren't actually smart, just smart compared with people who fell for the sales schrick.

  28. Ken Moorhouse Silver badge

    Stuff Smart Thermostats...

    Send the "Powers That Be" copies of TV Times (or the transatlantic equivalent) so they can work out when the adverts will be during Corrie, etc.

  29. DomDF

    It was admittedly a long time ago, but couldn't the pre-smart (dumb?) thermostats *also* be programmed to come on at a set time? How do smart thermostats make the obvious and unavoidable morning spike any worse?

    1. druck Silver badge

      Clock drift of up to a couple of minutes a week of the old thermostats takes care of that.

  30. AndrueC Silver badge
    Facepalm

    In hundreds of thousands of homes across the US that means a sudden jump in electricity use right before residents wake up – if people aren't changing default settings, which the paper suggests is the case.

    But that's the whole point of smart thermostats. If people had the intelligence and inclination to configure their devices they wouldn't need to waste money on smart thermostats in the first place. They'd just buy a 7-day thermostat - many have been available for over a decade - and set it up.

    1. Ken Moorhouse Silver badge

      Re: Smart vs 7-day thermostats

      Another factor is that "the time now" does not need to be setup on a lot of modern devices, meaning that there is precise synchronisation throughout a country.

      This is compounded by the fact that "most people" will arguably choose to set their thermostat to kick in or out on an exact time, such as 07:00:00, not 07:02:34. It would be interesting to see whether the load on the grid is punctuated by spikes at intervals, like on a ruler.

      07:00:00 Mega big spike

      07:01:00 minor blip

      07:02:00 minor blip

      etc.

      07:10:00 slightly bigger blip

      etc.

      07:15:00 Mini big spike

      etc

      08:00:00 Mega big spike

      The way round that is for Smart thermostat developers to incorporate a random offset into their products, which is big if you can't be bothered to set it up, smaller once setup.

      As a consequence there will be some that will complain that the switch didn't occur bang on the time they explicitly setup. If I were a coder of such a device I would say that someone setting up hours and minutes would be subjected to random deviations, but if they specify the number of seconds (other than zero), then no random offset would be applied e.g., 07:00:01 would trigger at that time, because the user has indicated a higher level of precision.

      In the old days, anybody buying a mechanical 7-day thermostat would be subject to two substantial random offsets in setting it up:-

      (1) Setting up the correct time of day is likely to be several minutes out.

      (2) Setting up an on or off event is also likely to have a leeway of several minutes either way.

      1. AndrueC Silver badge
        Facepalm

        Re: Smart vs 7-day thermostats

        Good point. My thermostat is 20 years old but its time will always be accurate because I bought the RCT add-on for it. However it also has optimal start so will vary its times throughout the year as will/should smart thermostats. But most of the time the variation won't be that big and presumably is roughly the same throughout a region.

        Unfortunately in the past while searching for information I've found that a lot of installers disable the feature because it supposedly confuses the user who can't understand why the heating doesn't come on at fixed times.

        You can lead a horse to water...

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like

Biting the hand that feeds IT © 1998–2022