Xcel smart thermostat users lose their cool after power company locks them out Thousands of Xcel Energy customers in Colorado this week discovered they’d been locked out of their smart thermostats, unable to adjust their air conditioning systems as local temperatures rose about well above 90°F this week. At a time when cyberattacks are a daily occurrence, customers would have been forgiven for thinking …
CBS affiliate KHOU Channel 11 reported at the time that many customers returned to find their homes approaching 80°F.
I'm from Texas, anything under 80 is too cold for my liking. Houston's problem is the humidity. And the fact that it is Houston...
icon selected to indicate preferred thermal environment.
I was just thinking something similar. Northern Oregon here, and to save money I've been leaving the AC switched off for long periods this summer and opening windows instead. Once the interior temp gets to around 84F my wife has been known to shut the windows and temporarily turn the AC back on, but not much below that mark.
AC can turn your house into a prison. I've frequented the Caribbean a lot and I found that those staying in AC hotels (mostly Americans) seemed to be trapped there unable to do much more than hit the beach and the ACed bars and restaurants. One couple I shared a dive boat stopped because they couldnt cope with the heat while heading for the dive sites. The dive master said it was quite common.
And we all know how honest the texas power brokers are. I expect the fine print on the sweepstake says something like "if you have won the sweepstake but you have expired (whether due to heat or otherwise), your award will be given to the CEO."
With respect, and regret, it ain’t quite that simple. The issue isn’t so much the amount of power being used, as is the rate of change of usage. Remember that generation has to exactly balance consumption. Generators of any type can not respond instantly - even the famous UK “TV pickups”, which Dinorwig is very good at managing, come on load over 5 minutes or so, rather than having NTP-synced clocks meaning the load comes on all within 10 ms.
On the other hand, if the power company has so many people on "smart" thermostats which they can control, surely the obvious thing is to inform all those thermostats to "jitter" their startup times across something like a 15 minute period. Or are they not actually all that "smart"?
CBS affiliate KHOU Channel 11 reported at the time that many customers returned to find their homes approaching 80°F.
I only assume the temperature was encoded in imperial hogsheads per nautical gallon because El Reg didn't want the rest of the world to know that this means 21ºC.
If this is how air conditioning is used in Texas then I'm not surprised there are so many brownouts. If the outside temperature is 32ºC as it said at the start of the story then setting the air conditioning to 25-26ºC is perfectly fine. It's supposed to be your house, not a fridge.
I did the conversion via Google when reading this because I don't do Fahrenheit; I'm not young - getting on for 50 - but I just can't grok Fahrenheit.
They said "approaching 80°F", whatever the figure might actually be, but 80°F is getting on for 27ºC rather than 21ºC, which is pretty much 70°F.
Oh FFS, for some reason I had it in my head that it was 11ºC for every 10ºF, but it's 5.5ºC, i.e. my half-remembered rule was 11ºC for every 20ºF. So I worked out 90ºF with a unit converter and wrongly converted 80ºF in my head.
So that's my fault, icon for me.
And this is from someone who is supposed to be passingly familiar with ºF (someone from the UK), so I'm not sure what the new house style of Fahrenheit must look like to 95% of the world. I guess debating units of measure increases reader engagement though, and that's what counts.
0°F is really cold—the temperature inside your freezer
100°F is really hot—ever so slightly over your body temperature, so when you start feeling completely miserable
Divide that interval into a nice percentage scale between miserably cold and miserably hot. 50°F is nice and even, 75°F is warm, etc.
Considered that way, it makes much more sense than Celsius. Human beings are never supposed to feel the temperature of boiling water (which varies considerably by atmospheric pressure too!), so 100°C is a useless reference point.
IIRC, 0F was the coldest temperature achievable using NaCl to lower the malting point of ice and 100F was body temperature (presumably measured rectally). The "boiling point of water" is really the temperature where the saturation vapor pressure of water is 1 standard atmosphere. The "Melting point" of water isn't well defined as compared to the triple point of water.
IMHO, both the Fahrenheit and Celsius scales are arbitrary and the only rational SI temperature scale would be electron volts...
"100F was body temperature"
Close. 96F was body temperature. The fixed points were chosen to be 0, 32 and 96, so that you could mark the scale by simply bisecting the gap a few times. Quite a common phenomenon with old units of measurement - the numbers look weird and difficult to work with when taken in isolation, but they were often developed because they made things simpler when working with the technology of the time. Celsius is much easier to work with mentally, but much more difficult to mark instruments accurately when all you have is a ruler and pencil.
Yeah, I think in degrees C ( and I'm age 65). But I know that 20c is 68f.
And the conversion C to F is pretty simple.
(C x2)- 10% then add 32. Backwards is harder if you want to be accurate.
Always the core of the second programme I used to write when learning about a new computer language - in my youth, when I used to play those games.
("Hello world" of course is the first).
If the outside temperature is 32ºC as it said at the start of the story then setting the air conditioning to 25-26ºC is perfectly fine.
In fact, 78ºF (26ºC) is a popular daytime air conditioning setpoint across the southern US, including Texas. (Lower setpoints are often used at night.) I'll let my badly-insulated apartment reach 82ºF (28ºC) but there are no people or pets present during the day.
If this is how air conditioning is used in Texas then I'm not surprised there are so many brownouts.
Texas has hit an Enron-like combination of poor regulation and unchecked profit-chasing that has led to inadequate power reserves and inadequate weather hardening in its grid. In the mid-1980s, Florida had winters comparable to Texas's 2021 winter storm (which hit -10ºF / -12ºC) but without Texas's grid problems. Florida also regularly handles extended heat waves at similar temperatures (104ºF / 40ºC) and humidity (70%+) without similar power shortages.
Point being: Texas needn't have this problem during summers, and the root cause isn't air conditioning set to ~26ºC.
and the fun fact is people setting their temperature very low on Summer are the same keeping it very high in Winter...
The common southern US domestic AC setpoint of 78°F / 26°C is higher than the monthly average outdoor temperatures of London from June to August. Non-rhetorical question: Does 26°C truly count as "very low" for air conditioning?
Can't you stand 28°C without AC?
"Without AC..." that depends on the conditions. Usually, 82°F is perfectly fine if you're in the shade and there's a breeze or a fan. However, that's a temperature I see outdoors primarily in winter. Even 86°F / 30°C is a mild day during the rest of the year. Last week it was 106°F (41°C) in the parking lot at work.
Further, those outdoor temperatures usually arrive with uncomfortably high humidity unless you're in Arizona or Colorado. Texas, Mississippi, Louisiana, Florida, Georgia, the Carolinas have plenty of humidity to go with spring, summer, and autumn. (Not to mention pollen.) The heat indices from those combinations of temperature and humidity are awful.
Air conditioning doesn't just control temperature. It also keeps humidity below 50%. (And lets you stay in a closed environment free of pollen.) Dehumidified, cool air works wonders for comfort.
Going back to the origins of my 82°F comment, my poorly insulated apartment would not level off at 82°F during the summer if, as you specified, I was "without AC." The last time my apartment AC broke during summer, it hit 95°F/35°C indoors. During summer months, I deliberately run the AC at 82°F to save ~$10 on my electrical bill compared to a daytime setpoint of 78°F.
I suppose you could say that settling the southern US was a "First World decision" that comes with "First World problems" like a dependency on air conditioning for comfort, but dealing with winter temperatures like 82°F "without AC" isn't the challenge. The challenge is dealing with summer heat and humidity, which aren't so mild as 82°F.
Also in Texas here. People in the UK have no clue about the climate here. I know. I'm originally from Scotland.
Getting off the plane, when I first came to Texas, the heat hit me like a wave, I stripped down to my tee shirt and asked my then fiancée if it was far to the car. It was nighttime, in November. The temperature was mid-seventies.
After about a month, I had acclimated, for the most part, to the point where locals putting on coats because it was sixty-something did not seem ridiculous to me anymore. The humidity is what make it unpleasant, for the most part - I'd spent time in Arizona years before and very much enjoyed the climate. I was there in summer, with daytime temperatures regularly approaching 110 degrees Fahrenheit.
A simpler predictor of comfort than relative humidity is dew point. A dew point of 10C (50F) is close to ideal and anything above 15C (59F) is getting uncomfortable. OTOH, a dew much below 10C (50F0 can lead to problems with dry skin - remember my two years in Carson City associated with the skin on my knuckles and elbows feeling like sandpaper.
One major difference. Florida can import power from other states. Texas being Texas (no commy Feds will regulate us!) has next to no import capability so when Texas has a generation problem like they did last winter, they are completely screwed. By choice, despite what the Governor would like you to believe (it was all ERCOT's fault).
Well, let's just crunch a few numbers here.
Apparently, a solar panel installation in Texas costs about $2.73/W. Let's assume the company could get a significantly better deal than that, call it $2/W. Then that $5.5m translates to about 2.75MW of generating capacity.
Divided between 22,190 customers, that's a shade under 124W each.
A central aircon unit in Texas uses, on average, over 3kW, so 124W isn't going to go very far toward keeping it on.
That seems a bit disingenuous though, a home solar installation is going to have a very different cost to a solar farm as it's a few panels over a small space so higher per household charge for the time spent vs number of panels.
But, that said, I'm completely talking out my arse as I have no idea how much power you could generate with $5.5m, but I feel it would be more than that.
You may be right. Also I haven't budgeted for the revenue gain from the increased capacity, which would presumably offset the cost.
But I think my basic point is sound: $5m is no more than a drop in the bucket you would need to meet that full demand. Reducing load is *much* more cost effective in the short term.
I have no idea how much power you could generate with $5.5m, but I feel it would be more than that....
Pity I didn't see this a half-hour ago since my daughter works for one of those "cover the desert in panels" solar companies, she'd give you an exact figure. Based on our experience both as a household and from our city my rough estimate would be "A lot".
She tells me that the wildlife likes them as well. You can also keep sheep in solar farms (not goats, though -- they eat the cables). Its really too early to tell but is possible that if we cover a large enough area of desert with panels it may start to influence the flora & fauna and even make its own microclimate.
A/C units are rated in 'tons' and a year or so ago I went through the exercise of converting this to rational units. I needn't have bothered, really, because the electrical circuit that runs the unit gives the game away. A typical domestic unit uses 6Kw. A larger "ranch style" house may need two or more of these.
Solar panels are quite good at offsetting most of the power drawn by A/C. Our roof knocks out up to 4.5Kw; its not a particularly large installation but it does have the advantage that its on all day, every day. (Panels also act as a roof shade.) If you've got a large installation, say over a parking lot or commercial building, you'd be amazed at the amount of power panels can produce -- one typical sized rank will produce over 40Kw (while keeping your car cool).
As for that 'ton' thing -- this measure of cooling is the amount of cooling you get from a ton of ice over a 24 hour period. Domestic A/C units are 3-5 tons.
My thought as well.
Power companies don't help themselves either though in pushing the benefits.
I regularly get e-mails from Octopus saying why not get s Smart Meter so you can sign up for our 'Smart' tariffs.
Umm, no. You know my monthly usage from the readings I diligently give you, why not give me even a ballpark estimate of how much my savings would be in the email itself and then I might have a think about it.
Only way I can really save on power usage is convincing the wife that bath towels can be used more than one time which would cut down on one washing and one drying load per day. :-(
I have a smart meter from octopus. It doesn't work. The previous supplier's did.
Why did you get a new smart meter when you switched suppliers? Even more odd that a newer meter didn't work where an old one did.
Or do you mean that your smart meter is now dumb after you switched suppliers? If so that's also a bit odd because nearly all meters have now been adopted by the new network. I don't think there's many left on the old networks. My SMETS1 meter was adopted a year ago. FYI I didn't get notified I only happened to notice when I was reviewing my billing.
Not that it's worth crowing about. The whole SMETS1 saga was unnecessary and took far too long to be resolved.
The only advantage of a smart meter is they can finally get rid of most meter readers, and if needed can cut you off without sending a man round to pull your fuse. Advantage for the consumer... Ermmmm..
Anyone know if they use 3G for coms? Cos I thought that was going the way of the Dodo fairly soon.
Not sure but suspect it is 3G. I was forced to get one in one property & they wanted access to my router so I told them to piss off. I assume that they are using 3G as 4G is very patchy around here.
They only want them installed so they can kill off meter readers, cut you off remotely and piss about with tariffs - to their advantage obviously, not the customer.
They only want them installed so they can kill off meter readers,
Good (speaking figuratively of course). Fewer people for them to pay means lower (or at least less high) bills for the rest of us.
... cut you off remotely,
Why would they want to do that? Firstly no company that is in the business of supplying you with something wants to cut you off. If they aren't supplying you they aren't making money off you. Secondly cutting you off requires an extensive and expensive legal process so, again, it's not something they want to do. Thirdly if they do eventually decide to cut you off they will have to send someone round at some point before then to assess your situation because that's the law. They can't just press a button whenever they feel like it.
The only people they will ever cut off are those people who refuse to pay despite being able to(*). The only thing remote disconnect avoids is the cost of sending an engineer. That's good because it will slightly reduce the cost to the rest of us who are paying for these free loaders.
and piss about with tariffs - to their advantage obviously, not the customer
Yet to be proven. So far it appears they are going to offer discounts to customers who move their heavier loads out of peak periods and that's an advantage to the customer.
(*)Refuse being the important word here. Unable to pay is something different.
Getting rid of meter readers is unequivocally a benefit to everyone except the meter reader. Remote reading is way cheaper, far more reliable, and can be done as frequently as necessary, typically daily.
Cutting people off remotely - well, if they do that without justification, I suggest you sue the crap out of them - exactly the same, in fact, as if you were cut off locally. At least it will only take them a few minutes to reconnect you, not the best part of a day.
Piss about with tariffs - if you don't think it's to your benefit, then don't choose those tariffs. Tadaa! Problem solved. You're welcome.
With a little Python script and spreadsheet magic it's possible to identify when power is being used regularly through out the day because there are 30 minute period readings - which over time helps build a picture of consumption rathher than just while looking at the mini-display gadget.
Yup. They have have adverts that are designed to make you think that having a smart meter will mean you use less leccy. It's bullshit of course. The meter will certainly show you when you are using most power. But if anyone wants to reduce how much they're using they already know damn well that this means turning off stuff.
> The only advantage of a smart meter ...
Another "advantage", that is frequently missed, is the opportunity for the smart meter to start taking into account the Power Factor of your house's loads.
Posts like this one https://www.lemonfool.co.uk/viewtopic.php?t=12645 explain it better than I would do (and you can watch a few Big Clive videos, his favourite Hopi meter shows the PF of whatever he is fiddling with today), but suffice it to say that the cheap LED lights using capacitive droppers have low PFs and that could (theoretically so far, but...) lead to higher costs if (when?) PF starts being used in the calculation of charges.
There are reasons for the grid to care about your PF (if it gets really low for everyone), so taking it into account may be justified, but then we'd be in a pickle trying to explain to Joe Bloggs why these LEDs may be cheaper to run than these LEDs over here.
In some states, they are mandatory. They tried that here, but there was so much public push-back*, that the public services commission allowed a short window to opt-out. The window was ridiculously short (a week or two) where you had to fill out some forms, and submit it to the utility company. I paid close attention, and got my paperwork in.
The catch is that you have to pay a monthly fee ($8 per month, I think) for having a "non-standard meter". Who says the PSC isn't bought and paid for by the utility companies.
* Some people were resisting in very creative and serious ways. When utility crews were coming around to install smart meters, some were met at gunpoint and chased off (open carry state, where brandishing is rarely prosecuted - my city councilman chased a utility crew off of his land at gunpoint a few years ago, the sheriff wouldn't do anything about it). Some people welded steel cages out of rebar over their panel. Someone else wrapped their panel in razor-wire.
$8 is approximately what it costs to send someone round to read the meter every month.
Obviously it wasn't always so expensive, because there's economies in reading a whole street at a time. But now the reader only visits maybe four houses on each street, it's much less efficient.
There isn't enough room in the meter box for both the meter -- smart or otherwise -- and potential disconnection switchgear (contactor). They might as well show up, open the box, and yank the cables, and you don't need a smart meter for that.
My beef with my utility is that I can't use their "insights" app to see the smart meter's live telemetry without paying them $2 a month. I would like to know the exact demand of (various devices around the house), but I'll be arsed if I'll pay them for it, since I'm already paying for the power itself.
Part two: Previous house "interruptible service" for the aircon: a separate meter (reduced rate) for with a radio receiver box for seasonal shutdown commands that could interrupt the thermostat signal to the compressor's relay. My electrician brother-in-law looked at it, how the receiver also needed 240V to run, looked at the thermostat wires, then reconnected the relay to the incoming thermostat wire but left the receiver powered so the power company wouldn't notice it offline.
I have a Belkin device which temporarily plugs in between the socket and device to give an idea of how much power it takes. For total consumption I have a remote unit powered by batteries that seem to last forever. It has a plastic clamp that goes round the main input live cable to make its power measurements. Presumably a current transformer. Then it sends a reading to a monitor about every minute. It seems to do some averaging over that period - so you rarely get any instant reaction to a device on/off. UK electricity suppliers were giving them to customers for free a while ago.
In both cases i do wonder at how effectively they - or the installed main electricity digital meter - handle the power factor of my various switched mode supplies. Not to mention the part-cycle patterns of dimmers and motor speed controls. A company in The Netherlands was found to be overcharging customers as their smart meters didn't allow for dimmers' reduced consumption during a cycle.
I presume the old revolving aluminium(?) disc meters were better able to handle such vagaries. A "hot wire" meter was good if the ambient temperature was reasonable.
Thanks for that emon link. Surprised they need a Pi as well as an Arduino. I have an Arduino Mega 2560 with small modules that handles 433mHz signalling, wifi, MSF RTC, SDcard, TFT screen - and monitors IR beam breaks.
I didn't realise that the red led on the supplier's meter might flash according to a standard (hopefully). Research needed - although the meter box is the hardest location in the house to get a mains supply for a device.
It's not the load as such, rather the sudden jump in load as thousands of thermostats all kick in at exactly the same time.
If the increase in load happened over, say, 15 minutes instead of simultaneously, there'd be less of an issue.
"...smart thermostats...unable to adjust their air-conditioning systems as local temperatures rose...well above 90°F..."
These homeowners got the crappy "Reward" combo that they signed up for. Hope that they enjoyed the company's surprise Dutch oven.
As for thermostat communism, just say no to "green" control freaks and profiteering power utilities that neglect to spend what is required for maintenance and capacity.
Don't usually agree with your comments, but in this case, yes the owners got exactly what they signed up for when they took the rate incentive. There's a good chance they didn't read the small print, or didn't think it would ever actually happen, so yeah they only have themselves to blame.
Not sure why you'd be downvoted for this.
Your comment about neglecting to spend on maintenance and capacity is also entirely true. There's been underinvestment for decades in the US and the UK. A combination of shitty energy policy and rampant greed. Generation and grid capacity simply hasn't kept up with increasing demand, and certainly in Cali PG&E's failure to perform even basic line maintenance in pursuit of profit has directly caused multiple wildfires, destruction of property and even some deaths.
This is indisputable and PG&E has been prioven responsible in several cases already. Again, not sure why you'd be downvoted on this, either.
Only leaves the hyperbolic "commie" rant. Trouble is, it's not communism. It's blatant, bare-faced capitalism and political corruption at work, doing what they do best i.e. shafting the populace.
My old house had power-company-run cutoffs for the hot water heater and HVAC. (Not "smart" thermostat, but a separate box.) The agreement very specifically said that it would only ever be off for 15 minutes or less per hour; presumably they split the houses into 4 groups and rotate between them. In 12 years, I never once noticed the house getting warm or water getting cool due to this.
The difference is that 15-minute restriction!
How long before you can purchase a blob ,that heats up the cooling thermostat when needed ?
A small heater, controlled by the real room temperature and the real temperature you want
if rooms to hot, heater turns on, and just heats heats thermostat !
or is that a business opportunity ?
I had a similar thought. A strategically placed resistor would do the job nicely. If it was 56k and rated for 2W, it would dissipate a fraction over a Watt at 240V, which I imagine would be sufficient to do the job if you mounted it suitably and with some simple ducting if necessary.
Smart for the generating company, and smart for its other customers. The stupidity was the subscribers', who clearly didn't read and understand the T&C — which were pretty obvious.
And that, kids, is why I'll never own a "smart" thermostat [*] or "smart" appliance that turns over control, without asking, to the leccy company, or anyone else for that matter. I may be stupid, but I'm not that stupid.
[*] In fact, a "smart" thermostat came with the installation of a new heat pump four or five years ago. I read the T&C for setting up a connection to the mother ship (in this case, the manufacturer of both heat pump and thermostat). It was when I got to the point where the "agreement" allowed the manufacturer not only to access all of my browser history, but to sell that information to anyone they chose, that I clicked "cancel" rather than "agree." Alas, I'll never be able to change the settings from afar (not all that terribly useful since the pandemic started) or put a pretty background on the colo[u]r display.
> it would be pretty easy to add a manual override switch that would engage your A/C, bypassing the thermostat
I'm pretty sure they'll have one built in already - the owners just have to remember how to get up off the couch and walk over to the unit and operate the switch.
If they have one of those IR remote controls with the undecipherable LCD hieroglyphics and 3 different temperatures displayed in a cycle, none of which appear to relate to the actual temperature in the room, then they are completely forgiven. :-)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
