24

I am looking at replacing all the thermostats in my house. I have hot water baseboard heat and forced air A/C. Each room's baseboard heat is controlled by a thermostat in that room.

I tend to prefer simple mechanical interfaces to digital "smart" systems.

My family does not have a "normal" living schedule. I am often up until 3-5am. My partner can wake up anytime from 5am to 9am.

I have heard that the benefits of smart temperature systems (which adjust the temperature settings based on the time of day and can be pre-programmed) are over-stated, and that classic single-temperature controls (like this or this) are actually fine from an economical point of view.

Is there significant savings associated with those smart systems, or can I use the more simple thermostats without costing myself in monthly bills?

LShaver
  • 1,238
  • 11
  • 31
JoshuaD
  • 417
  • 3
  • 11
  • 2
    Do you have any plans to get a heat pump to replace your existing A/C? – ThreePhaseEel Apr 19 '21 at 11:32
  • Not this decade. The A/C units are relatively new. – JoshuaD Apr 19 '21 at 20:53
  • 8
    Depends hugely on the insulation. A well-insulated house does not lose significant heat over a mere couple of hours, so twiddling with the thermostats does nothing... – DevSolar Apr 19 '21 at 20:55
  • I bought a nest thermostat and it had an interface and aesthetics that I liked but I ended up using it like a 7-day programmable thermostat. It died and I replaced it with something more basic, no major regrets. One nice thing that I kind of miss was being able to use away mode when on vacation and then kick it back on during my trip home but at the same time I was paranoid that someone might be able to turn my heat off during the dead of winter - they have a feature which is supposed to prevent that, though. – JimmyJames Apr 20 '21 at 20:46
  • 1
    How is the billing done in your region? Is it per 15min/hour/day? And how much do the prices fluctuate "normally"? These things are wildly different across different regions. If it is normal that prices 08-09 is 50% higher than the prices 05-06, then you can save a bit by having a timer on your hot water tank. – epa095 Apr 21 '21 at 11:05
  • @epa095 Are you implying heating with electricity? Or how is consumption priced at different rates at different times? - Around here heating is usually done with gas from a pipe system (huge storage capacity in transit), gas from a tank (storage), oil from a tank (storage) or wood pellets from a tank (storage). Neither require nor justify load based pricing. - There are some regional heating networks, usually powered by large power plants which are running at high capacity, anyway, so also no load based pricing. Heating with electricity is mostly discouraged, though some older systems exist. – I'm with Monica Apr 21 '21 at 12:48
  • I have a smart thermostat and IMO the only benefit is being able to control it from my phone. It's terrible at figuring out its own schedule and constantly sets the temp colder than I ever would. It probably would be cheaper to use its suggested schedule, but I'd be miserable. – Kat Apr 21 '21 at 14:51
  • 2
    All the answers provided so far are great, however, they all seem to assume a "normal" get up in the morning, everyone evacuates the house for the office/school during the day, return home in the evening, go to bed about the same time type of schedule. You've indicated that you do _not_ live by that type of schedule, so be sure to apply the logic in these answers to _your_ situation. Are there long periods of time when nobody is in the house & you can allow it to change temp? Do you all want it a different temp for sleeping? Only _you_ can decide if these great studies apply to _you_. – FreeMan Apr 21 '21 at 18:24

13 Answers13

28

I only worked in the commercial/industrial field of steam and hot water boilers so take what I say lightly. Here is what the heating engineers told us; if you can set back the thermostat or reduce the buildings temperature for 8 hours or more then there is a cost savings and if you can't it is probably not worth anything. One of the main problems with temperature reduction was that when the room temperature was brought back up to the comfort level, the furniture, desks and everything in the room was still at a reduced temperature. If you were to use the furniture or items in the room, the room still felt cold.

I have a friend that is a lot younger than I am and he changes temperature many times a day. When I go into his home I always take a sweater, In my house I set the temperature to one setting for the winter and one for the summer. We have like houses and heating systems and our heating bills are about the same. My house is always more comfortable than his. My 2 cents.

d.george
  • 12,325
  • 2
  • 14
  • 23
  • 3
    "and our heating bills are about the same" means nothing to me unless you're trying to tell me that you both use almost the same amount of therms. – MonkeyZeus Apr 19 '21 at 13:56
  • 14
    since he lives next door to me and his gas bills are about the same as mine, that means that we use the same amount of fuel be it in $$, therms or cu ft. so yes it means something. And, if our houses are kept at the same temperature. I am just giving an example of temperature set back and what it does or doesn't do. – d.george Apr 19 '21 at 14:13
  • 5
    @d.george Thanks for clarifying. You did not mention that you are neighbors and paying the same price per unit of heat before; I hope you can understand why that's critical information. – MonkeyZeus Apr 19 '21 at 14:18
  • Great answer, yes, you can heat up air a lot faster than things in the room. I have a cabin with a WIFI controlled tstat. During the winter when we go there, we turn up the heat about 5-6 hours before we plan on arriving. Still everything in the cabin is still cold, but at least it beats turned on the furnace when we get there. – George Anderson Apr 19 '21 at 15:47
  • The eight-hour threshold seems about right. I've got a programmable thermostat for my living room, and the extra power usage for bringing the room up to temperature every morning amounts to several hours of steady-state heating. – Mark Apr 19 '21 at 22:21
  • 2
    One large variable missing from this answer is the difference between the two dwellings. Your friend's house could be lacking insulation, have a very inefficient heating unit, and have leaky windows... yet could still be surviving with a similar heating bill as yours. With the same level of heating that you're accustomed to in your home, your friend's bill could be double yours. We can't know any of that from this answer. Objectively, less fuel is less fuel.... and less fuel costs less money. – Aaron Cicali Apr 19 '21 at 23:17
  • @AaronCicali it'll be very odd unless one house had significant renovation done. The builder would've used the same stuff across the lot of houses since they're neighbours. – Nelson Apr 20 '21 at 07:33
  • @Nelson Assuming the houses are the same size, same floor plan, same shading and orientation, and by the same builder. – SomeoneSomewhereSupportsMonica Apr 20 '21 at 07:51
  • I get my heating bills to about 2/3 of what's typical for my type of house by acting similarly similarly to your friend, but with dumb hardware: I have my heat on for a period in the morning and another in the evening, and leave the thermostat untouched. I sleep hot so have to let the room temperature drop before bed. If I'm at home during the day I'll probably run it for another hour, maybe twice on a really cold day. The house is well insulated, but I still expect to adjust clothing to suit. – Chris H Apr 20 '21 at 08:08
  • ... over a cold spell lasting several days the background temperature when the heating has been off for a while definitely drops. A programmable thermostat would avoid that, by warming the house a little, so it would cost a little more than my setup but be more comfortable. I wouldn't go for it personally - the heating coming on in the night would wake me up, and the dumb hardware I have works reliably - but for many it would be good. They're fiddly to get right though – Chris H Apr 20 '21 at 08:11
  • @Nelson having neighbours with vastly different houses built decades apart is not uncommon. I don't think you can make any assumptions about construction style, years built, orientation ...anything really just because their neighbours. Just that they probably have the same gas/electric proivder. – Brad Apr 20 '21 at 14:14
  • 2
    A smart thermostat will do better than a *poorly programmed* programmable thermostat. "Better" may not mean saving money .. it may mean providing more comfort despite the owner's clumsy attempts at programming. It may even cost MORE to provide that comfort. – jay613 Apr 20 '21 at 19:34
12

Here is the science-based thermodynamic law that drives all this.

Suppose you have 2 spaces, A and B, at different temperatures. Between them, there is some level of thermal insulation.

No matter what is true about the insulation, this rule always follows:

Thermal transfer is proportional to the difference in temperature.

Take any random insulator (green) that happens to have nice round characteristics of passing 1000 BTU/hr per degree F difference:

enter image description here

You can see where keeping this thing cooler will result in less temperature differential.

How to apply this law to the question at hand

This is for a physics lesson, and includes exaggerated hypotheticals for educational purposes, designed to be illustrative rather than real-world numbers.

Two "example" houses. One is kept at temp, the other managed with a programmable thermostat which turned the heat off at 7:00 and kicked the heat back on at 3:30 so it'll be comfy at 6:00.

And let's presume cloudy day with constant temp, just for simplicity, because we're here to understand a law, not model a complex system.

enter image description here

Obviously the insulation factor, thermal mass and heating capacity of this house are unusually bad, and the "constant-temperature day" is unrealistic too. But we're here to learn. Some of us, anyway: for others, this is an 'inconvenient truth'.

What's positively true is that the house on the right lost less heat.

Here's the thing: Heat not lost is retained in the house. The house on the right had to make less heat simply because it lost less heat. Yeah, the house on the right has to run the furnace hard in the afternoon "to catch up"... but since furnaces run at only one speed, the simple fact is that the right house's furnace had to run less time in total than the left house's furnace. We know that because the house lost less heat, and there'd be nowhere else for the heat to go.

Except it's more extreme because of solar gain

In our house, the mornings are cold, and the furnace runs of course. And it runs all morning. But, solar gain has been heating the house's exterior, chasing off the cold of night... and this heat starts penetrating the insulation and warming the house in earnest. Since the house is already at 70F thanks to the furnace, the solar gain not only holds it there but also warms the house further - sometimes too much! So the furnace does not run again until dusk. We were there the whole time because of COVID, but imagine if a programmable 'stat had shut off our heat at 8 am.

The furnace would not have run. The house would cool off, and would be ~50F when solar gain really starts pounding on the house. The solar gain would lift the house to 70F just as we arrive home to enjoy it. The difference being, with a programmable 'stat, we would not have been 'out' the fuel costs in the morning. So a programmable 'stat lets you make better use of that solar gain, by letting house temp take a "morning dip".

Now, what does a 'smart stat' bring to the picture? In this hypothetical, the dumb programmable starts heating at 3:30 and reaches target temp at 6:00. On a 60F day it would also start at 3:30, be at target temp by 4:30 and waste the heat for an hour and a half. A smart 'stat would know exactly when to start. Again this is exaggerated for educational purposes.

Harper - Reinstate Monica
  • 276,940
  • 24
  • 257
  • 671
  • 6
    "Furnaces only run at one speed" - only primitive ones. Modern heating systems in the UK have furnaces with PID controllers that control both fuel flow and water circulation pump speed, linked to a continuous input of room temperature variation (measured to 0.1 degrees) not a crude on/off thermostat coupled to an on/off gas switch. – alephzero Apr 20 '21 at 10:48
  • 1
    Your example "cloudy day" doesn't have nearly the solar gain that a sunny day does, so there's less benefit there. – FreeMan Apr 20 '21 at 14:05
  • 1
    Are you trying to differentiate between an old one-temp mechanical thermostat, a programmable thermostat, and a smart thermostat? I feel like a programmable thermostat (multiple set points based on time of day/day of week) can give all the benefits you describe and then being "smart" like a Nest or EcoBee doesn't bring anything additional to the table. – Brad Apr 20 '21 at 14:10
  • 1
    This answer makes a good argument but I’m concerned that the science is not as connected to real world situations as I’d like. One real world wrinkle is that heat pumps have different efficiencies depending on the temperature difference. Another is that being comfortable often conflicts with thermodynamic ideals. Also the scenario neglects higher outdoor temps during the day that reduce the need to heat both houses. I’ve lived many places with almost zero solar gain. And any solar gain would benefit the left hand scenario just as much as the right. Etc. – Todd Wilcox Apr 20 '21 at 15:01
  • @J... *Clearly, you've never rented in the sun-belt LOL*. But you are right, I should not use "real world number" conclusions, so, removed. Other than that, please explain why you need "more believable" numbers. This is a lesson in basic physics, surely not your first one. Do you disbelieve physics merely because the textbook examples showed an 'arbitrary 1kg mass' that you could not relate to? If you'd like, I'll change the word "house" to "envelope"... – Harper - Reinstate Monica Apr 20 '21 at 18:26
  • 1
    @Brad Sorry, "programmable" is sufficient. I added a para to describe the advantage here of a smart 'stat. TLDR: it knows *when* to start reheating. – Harper - Reinstate Monica Apr 20 '21 at 18:35
  • 1
    @ToddWilcox My last section addressed solar gain. Actually, think of heat pumps as "pumping uphill". *The less uphill they pump, the less energy they need to do the pumping*. It's more efficient to pump heat from a 40F space to a 45F space, than from a 40F space to a 70F space. My first section applies to the freon-interior air relationship as well. – Harper - Reinstate Monica Apr 20 '21 at 18:44
  • 1
    I’m just saying none of this theory has worked out for me in real life in the past. Heat pumps change modes when there is a large enough temperature difference where instead of using refrigerant they turn on heating coils and operate like a central air space heater. That costs a LOT more when it happens. Also they have to de-ice the exterior heat exchanger so you pay for that even when it’s not heating. And a heat pump has a huge uphill climb to make when you turn it on/up before coming home from work. They suck in general and programming them doesn’t seem to help to a noticeable degree. – Todd Wilcox Apr 20 '21 at 18:47
  • @alephzero it's a hypothetical, exaggerated for educational purposes. Do you think discussing a variable-output furnace would make it easier to understand? I'm thinking the exact opposite... – Harper - Reinstate Monica Apr 20 '21 at 19:04
  • 2
    @J... The crux of OP's question revolves around a "folk wisdom" chestnut, that turning down a thermostat is pointless since it costs the same to warm the place up again. This chestnut is a frequent visitor here. I am debunking that chestnut, by showing "at a physics-textbook level" how that is rooted in a false presumption. If you want to build a more real-world example using hard numbers, it would be best-answer and I'd +1. – Harper - Reinstate Monica Apr 20 '21 at 19:23
  • @Harper-ReinstateMonica Perhaps this evening when I get a moment. I guess I just feel the numbers need some caveats that they are quite contrived and deliberately exaggerate the scale of heat loss compared to a typical home, that's all (which I see you've added). The physics is fine, it's just not representative of a normal home. You've got my + just the same. – J... Apr 20 '21 at 19:26
  • @J... That'd be tremendous! Yeah, I've disclaimed the hell out of it. The 1K BTU/degree-hour is to make the numbers nice and round, so a person can instantly see what is going on. I did it in part 1 and just kept it for part 2 for consistency. 1 BTU/hr would've worked too, but I went with the former, thinking about my 900sf cottage where the 50K BTU/hr furnace runs 2/3 duty cycle on a 35F night. Not just the insulation, it's the drafts! – Harper - Reinstate Monica Apr 20 '21 at 19:45
  • @Harper-ReinstateMonica I guess I figure sunbelt must insulate to keep cool too... 2/3 of 50KBTU/h seems high for 900sf. My house is twice that sq.footage with 80KBTU/h modulating at ~35% most of the winter - duty cycled on top of that. I maybe hit 50% when it's below -30C (-22F), still duty cycling on and off. And I'm having a major insulation job done this year because the house is old and drafty, so I don't consider my home to be a particularly good example of a well insulated home. – J... Apr 20 '21 at 20:00
  • @Harper-ReinstateMonica [How Much Can You Save By Adjusting Your Thermostat?](https://blog.directenergy.com/how-much-can-you-save-by-adjusting-your-thermostat/). Seems a typical home can save ~10% on heating by dropping to ~16C (61F) overnight (8h per day). I run 21C (70F) (home), 19C(66F) (away/night) and maybe 3-5% is the savings I see, which seems to match up with this article. Seems plenty of people have done these calcs already. – J... Apr 20 '21 at 20:53
  • @J... Yeah I have a preference for the sunbelt, but old cities and thus old stock. I have no use for postwar cities like Las Vegas or Phoenix or new developments like Mountain House that need artificial climate control to be habitable in the summer. In many cities, 100 year old rental stock is slums... but in the sunbelt it's often desirable. So here I am :) We share a 100A service with 7 other units and thus, landlord pays... so A/C is discouraged... – Harper - Reinstate Monica Apr 20 '21 at 22:22
  • @ToddWilcox: When you talk about a "large temperature difference", I think you're talking about current indoor vs. target indoor. But the efficiency / "pumping uphill" analogy being used in this answer and other comments is about the indoor vs. outdoor temperature: it's hard to move heat against the direction it would flow by conduction. Not an up slope in a temp vs. time plot! (But yes, it would make sense that a heat pump unit would give up efficiency if it thought it needed to heat really fast, or wasn't keeping up in a really drafty house. Maybe use an intermediate step, not 1 big jump?) – Peter Cordes Apr 21 '21 at 07:00
  • 1
    @alephzero a modern UK boiler with simpler controls will still modulate the burners to control the temperature of the water in the condensing heat exchanger. This is probably rather common as boilers are updated without a full system rebuild, and it works quite well combined with TRVs. – Chris H Apr 21 '21 at 09:14
7

It all depends how do you set up your not-smart thermostat VS how do you plan to set up your smart thermostat.

They don't just save energy for the same use.

If you set up non-smart thermostat for 'comfort always' you will pay more. If you will set up it as 'discomfort always' you will pay less.

For the same level of comfort you need different temperature at each part of the day. For example at night you don't need as much heat if you sleep in bed. But if you are active at night or live with unpredictable schedule, then smart thermostat won't be of much use because of how complex the pattern is.

So, overall, if your time schedule is predictable, and you are interested in cost for the same level of comfort, then yes, smart thermostat can save significant sums of money. About 20% of heating cost.

FreeMan
  • 37,897
  • 15
  • 71
  • 155
Surprised Seagull
  • 179
  • 1
  • It also depends on what you have for a heating system... – ThreePhaseEel Apr 19 '21 at 11:32
  • I wonder if it might be possible to dumb-down a smart thermostat, so it can provide a minimal level of baseline comfort and be used as a remote control to set the heating on as you leave your variable shift, in time to be warm when you get home – Chris H Apr 20 '21 at 08:14
  • @ChrisH You can indeed do some fancy things like that with smart thermostats, particularly if you use something like [IFTTT](https://ifttt.com/) to use multiple smart devices/services together. You can (for example) create an action that automatically triggers once you get a mile away from work and sets your thermostat to a particular temperature. Or, tell Alexa "I'm heading to work" and trigger a rule to set the thermostat, turn off the lights, close the garage door, and close the blinds. – bta Apr 20 '21 at 22:05
  • @bta interesting - obviously they should be capable but I wasn't sure about interoperability. It's not for me though. I'd rather do things manually than introduce new failure modes, not just technical ones,but expensive products reliant on ongoing support for basic functionality, and that's easily dropped when the manufacturer gets bought out – Chris H Apr 21 '21 at 05:29
6

One of the benefits of "smart" systems is you can make more than just time-based adjustments, though it depends on the system. I have an EcoBee (no affiliation) so I'll speak to that.

The selling point to me with this was the room sensors, which have temperature and motion. I have 3 in the house, plus the main thermostat (which itself has temperature, motion and humidity).

One difference though is I have forced-air heat and air conditioning, with a single thermostat mounted centrally.

Target temperature + location

During the day, the temperature is based on sensors on the main floor, and at night, based on the ones upstairs. This means the temperature is kept comfortable where we are most of the time. I think I notice it more for the A/C than the heat, but the effect is there for both.

In other houses with time-of-day programmable thermostats, I generally have had to mess with the night-time temperature (read from the main floor) to make it comfortable upstairs, and this changes throughout the year depending on outside temperature.

In houses I've been in with a non-programmable thermostat, it seems most people keep the main floor at a temperature I'd call "freezing" in order for it to be comfortable-ish upstairs.

Auto home/away

If there's motion detected, it activates the "home" schedule (even if we're normally "away"), and if there's no motion, it automatically says in "away" mode until we get back. This works so well, that I hadn't even adjusted any schedules until just a few months ago, despite working from home for over a year since COVID started.

Fan cycling

My house has some large south-facing windows, and that side of the house gets significantly hotter on a sunny day. I set up "run fan minimum 15 mins per hour" and though it's hard to be objective about it, I think this has led to better balance in the house. I can't say for sure it's run actual heating/cooling any less, but I can say I don't notice it coming on seemingly unnecessarily (eg: A/C coming on when it's already quite cool on the north side of the house).

Remote adjustment

I really want to say this is a stupid feature: if you set everything up right, there's no need to adjust the temperature of your house remotely.

However I have used it a few times to enable "vacation mode" when I've otherwise forgot. This just changes the temperature by a few more degrees than home/away does and holds it until you get back. Maybe saves a tiny bit of money, if you regularly go away for days a at a time and can't remember or be bothered to adjust your thermostat before leaving.

Is it worth it?

I would say purely based on energy savings, it would be worth it only if you can make good use of motion-based scheduling. If people are away/home/sleeping at varying and non-regular times, you can reduce the energy use a bit (by raising/lowering the temperate a few degrees).

For comfort? Almost certainly yes, though it would depend on some of the specifics of your house and lifestyle.

gregmac
  • 28,726
  • 15
  • 82
  • 159
4

I installed a pair of Nest smart thermostats in my house and my energy bills have not changed to a statistically significant level compared to my rather basic programmable thermostat. That is, the amount saved due to using less energy was less noticeable than the amount I saved by switching to a different electricity provider (YMMV, highly dependent on how much you currently pay for energy). The comfort level in my house is greatly improved, though. If you measure utility not simply in dollars but in return on investment (comfort received per dollar spent), they're definitely worth the price.

Here are the main features that made the biggest difference for me:

  • Smart thermostats tend to be more precise. Many can measure temperature to the tenth of a degree, so they tend to keep the room temperature closer to the set point. Cycles are shorter and room temperature is more consistent.
  • My smart thermostat can detect when sunlight is coming through the window and beating directly on it. It adjusts its measurements to compensate. My old mechanical thermostat would measure erroneously hot on clear summer afternoons and run the A/C when it wasn't really needed.
  • Smart thermostats often support a remote temperature sensor. I put one in the bedroom and have the thermostat set to use that sensor at night. My house has a noticeable temperature difference from one end to the other, so the smart thermostat lets me keep the room I'm in comfortable without over-cooling the parts of the house that are empty.
  • At the end of a cooling cycle, smart thermostats can turn off the compressor (a major user of energy) and run just the fan for a while (uses very little energy). Your ducts are full of cooled air, and this pushes that cooled air into the house. It's kind of a waste to leave it in the ducts, slowly warming back up to room temperature.
  • Smart thermostats often have a hybrid "heat+cool" mode that will keep the temperature between two set points. This keeps things much more comfortable in spring and fall when outdoor temperatures can vary dramatically from day to day (no more manually re-programming the thermostat every couple of days). It's also extremely useful if you're going to be away from the house for an extended period of time and don't know what the weather will be like. Set it to keep the house between 60 and 80 while you're gone and don't worry about it.
  • Smart thermostats can learn how long it takes to heat or cool from one temperature to another and use that to adjust their schedule. If you set it to be 67 degrees at 6PM, the thermostat can use past data to determine that it needs to start cooling at 5:42 PM in order to hit the target temperature at the specified time based on current conditions. Programmable thermostats generally start cooling after you hit the specified time, so the house may or may not be the desired temperature when you get home.
  • Nest periodically sends you email reports showing you how much heating/cooling you used on particular days, how your usage compares to the previous month, how weather impacted your usage, etc. These provide a lot of good data for understanding your energy usage, predicting future usage, and measuring the impact of any changes that you make.

All of these features will save you money by reducing energy usage, but the most noticeable impact they have is on comfort. You can make your house much more comfortable and save a little bit of money each month. A digital programmable thermostat might have a few of these features, but it really takes a smart thermostat to get the most significant ones.

Caveat: Your situation will be a bit different since your heating and cooling are two separate systems. Using a smart thermostat in every room for heating is likely cost-prohibitive (some smart thermostats aren't even compatible with certain types of baseboard heating). You'll likely get good use out of using one for your central cooling system, though.

bta
  • 1,505
  • 5
  • 13
  • "Cycles are shorter" is an *anti*-feature. Many HVAC systems will wear out quicker doing the same work hours on shorted work *cycles* vs. the same hours on longer cycles. Well... if you care about saving money over some minimal discomfort, anyway, which I would guess from the question that the OP does. Also, leaving cold air in ducts is not exactly wasteful; that lack of heat is just going to diffuse into the house anyway. Circulating air *in general* will feel cooler, but you could also install ceiling fans. – Matthew Apr 21 '21 at 17:27
  • @Matthew - Along those lines, most smart thermostats include logic that prevents them from cycling in a way that damages the equipment. Some models can even monitor the health of the A/C unit and alert you to potential issues before they get serious enough to notice. – bta Apr 28 '21 at 18:25
3

A smart thermostat is almost guaranteed to reduce your energy usage by 4 to 19%, depending on your climate and HVAC system type.

No need for hypotheticals -- there's hard data

Given the length of time that smart thermostats have been on the market (the first generation Nest was released in 2011), there have been a number of field studies to verify energy savings.

In 2017, Southern California Edison (the utility covering about 15 million customers in southern California) performed a review of existing literature on smart thermostat savings to develop estimates for California: "Work Paper SCE17HC054 Revision 0: Residential Smart Communicating Thermostat".

This study provides a thorough analysis of seven prior US studies, all published in 2015 or 2016, and includes a regression model with additional results for California. The paper is dense, but the prior studies are described starting in section 3.3.

Here's the table of results, with links to the prior studies (where I was able to find them). "Participants" indicates how many smart thermostats were installed. Some studies used a control group (with the quantity indicated), while others used a pre/post method, where energy use before smart thermostat installation was compared to energy use after.

Study Location Participants Author Heating savings Cooling savings
PG&E Smart Thermostat Study: First Year Findings California 2,207 (control: 1,520) Applied Energy Group for PG&E 5.3 therms 4%
Energy Savings from the Nest Learning Thermostat: Energy Bill Analysis Results National 1,369 (pre/post) Nest 9.6% 17.5%
Evaluation of the 2013–2014 Programmable and Smart Thermostat Program Indiana 197 (control: 2,611) Cadmus for Vectren 11-14% 9-19%
A Look Inside the Eye on the Wall: Sub-metering Data Analysis and Savings Assessment of the Nest Learning Thermostat Washington 176 (pre/post) Bonneville Power Administration 12.35% 12.35%
Energy Trust of Oregon Smart Thermostat Pilot Evaluation Oregon 280 (control: 800) Apex Analytics for Energy Trust of Oregon 6% N/A
Evaluation of the Space Heating and Cooling Energy Savings of Smart Thermostats in a Hot-Humid Climate using Long-term Data Florida 27 (pre/post) Florida Solar Energy Center 9.5% 9.5%
Nest Learning Thermostat Pilot Study California 505 (pre/post) Navigant for SoCalGas 5.4% N/A
Regression model for the work paper California N/A Nest for SCE 11% 12%

But what was the baseline?

The goal of these utility studies is to estimate savings for any baseline. They are using a state-mandated energy efficiency budget to generate kWh and therm savings, so they need an estimate of what the savings will be when installing a smart thermostat in any home that doesn't already have one -- regardless of the factors that other answers have mentioned:

  • HVAC system type
  • Number of occupants
  • Occupant schedule
  • Occupant thermostat use
  • Existing (non-smart) thermostat type
  • Home size
  • Home insulation quality or air-tightness

The only factor they do control for is climate, because that is something they can know without going into homes or talking to residents. All of those other factors will affect your individual savings, but these studies show that when you install a smart thermostat in enough houses, the differences average out to a range of 4 to 19% nationwide. The range gets smaller when you start looking at specific regions.

While your exact situation is unique, these seven studies represent nearly 5,000 homes across the country. Pre/post studies compare identical homes and control for weather, while studies with a control group have identical weather and control for home characteristics (each treatment home is matched with one or more similar control group homes for the analysis).

Try it yourself for cheap or free

I see in your profile that you are in New Jersey. Several NJ utilities offer rebates or discounts on smart thermostats (also true for many utilities around the US):

  • South Jersey Gas: $19
  • New Jersey Natural Gas: free
  • PSE&G: free

Try it for yourself, and if it doesn't work, you can switch back to your old one.

LShaver
  • 1,238
  • 11
  • 31
  • I briefly scanned three of these and did not see where they controlled for the "old" thermostat to be completely dumb, programmable but not programmed or ineptly programmed, or programmable and well programmed to the living habits of the occupants. IMO this is important ... Are we upgrading from a 1955 blob of mercury or from a $20 2005 7-day timer that has been thoughtfully set up? Also, those utility-provided thermostats come with some heavy big-brother Ts&Cs that may not suit everyone so I don't love comparing savings form "free" replacements. – jay613 Apr 20 '21 at 19:38
  • @jay613 on baseline: In general, these studies are intended to be broad enough that among participants, all potential baselines (anything NOT smart) would be covered. Thus the range of savings. At least two address the question: The 2015 Nest report mentions that a previous estimate of 20% savings was revised down to reflect increasing prevalence/use of programmable thermostats. The overall SCE report models the baseline using "comfort temperatures," which could either be a result of a programmable thermostat, or someone actively adjusting setpoints to stay comfortable. – LShaver Apr 20 '21 at 19:57
  • @jay613 regarding T&C -- that's definitely a valid concern for any electronic product, but generally the utilities are subsidizing commercially available products, so it's the same thermostat you'd get if you ordered it from the manufacturer. – LShaver Apr 20 '21 at 19:58
  • That's what I suspected re the studies, it's all averaged in. A Nest replacing a programmable device that was programmed by an overzealous spendthrift might *increase* utility costs, but provide greater comfort to the owner's victims (family). The NJ PSE&G Ts&Cs gives them the right to remote *and on prem* monitoring for two years. They can enter your house. – jay613 Apr 20 '21 at 20:11
  • @jay613 hmm I see the T&C are both better and worse than what you found -- it's four years, but access is just to verify that the thermostat is installed and operational, with notice. The data they collect is just billing data -- which they're collecting anyway. https://marketplace.pseg.com/psegsupport.html – LShaver Apr 20 '21 at 20:31
  • @jay613 One of the worst gotchas I've seen was giving the utility the ability to remotely adjust your thermostat if they decide that you should be using less energy (like an ISP throttling bandwidth). – bta Apr 20 '21 at 21:57
  • @bta In Maryland they have had for a **long** time programs to allow remote control of a/c compressor on peak days. Optional, and you get a rebate/discount (which varies depending on how many times a month they can shut you off). A lot of the issues have to do with understanding the programs and managing expectations. If you are at home almost every day during the day (retired or have very young children, **or lots of us for the past year**) then those programs are a bad choice because they affect your comfort - and if you forget you even have it then you get surprised and think your a/c is – manassehkatz-Moving 2 Codidact Apr 21 '21 at 00:35
  • busted. On the other hand, if you can tolerate a little discomfort occasionally, and are not at home during the day most of the time anyway, they are a great way to save a few $. If enough people do that, it can help avoid rolling brownouts (or blackouts!) that would affect everyone. – manassehkatz-Moving 2 Codidact Apr 21 '21 at 00:36
  • @bta I do not think the "free thermostat" deals that are in discussion here come with a requirement that the utility can adjust your thermostat or shut off your A/C. There are *rate plans* that come with the provision, but you enter those (hopefully) with open eyes. The free thermostats require that the owner actually use them, correctly install them, and allow the utility to monitor their use. The monitoring provision tends to be a little heavy handed for my taste, but it's nothing close to remote control that comes with off peak rate plans. – jay613 Apr 21 '21 at 15:48
  • They had a low-tech version of brownout avoidance in NYC before Covid. All the big building managers in the city had an alerting app where utilities would "alert" them, which just meant asking nicely, to raise the temperature 8 degrees, shut off half the elevators and all down escalators. The cost of non-compliance would be brownouts or another 2003 style blackout. – jay613 Apr 21 '21 at 15:51
1

Depends on where you would set things, the insulation, etc. You would really need to test it yourself (to the extent possible) to see what is possible.

Why might it help?

Your heater has one job: to replace the thermal energy inside the dwelling that has been lost. If replaced exactly, the temperature is maintained. For most structures with reasonable perimeters, this loss is a function of the temperature difference between the inside and the outside. 15 degree difference, lose X energy per hour. 30 degree difference, lose 2X energy per hour. So for any period that you can allow the interior to come closer to the exterior, the total loss is reduced and the heating load is less. (This is simplistic and ignores solar load, drafts, assumes the heater efficiency is constant, etc., but is good as a first-order assumption). Yes, you have to bring everything back up to temp at the end of the window, but there will have been less energy loss during that time, so when everything is back up to temp, less heating will have been required.

Note that the better the insulation, the milder the temperature, the less time you can have the system off, then the less savings you will be able to achieve by changing the setpoint. (At the same time, the less of personal impact changing the setpoint will have).

If you're a penny-pincher and are diligent about setting the thermostats to just what is necessary, then a programmable or smart system may not save much.

If you just leave the thermostats set to the same temp all season just because that's comfortable, a very simple programmable unit can make a difference. Even if you only have a few hours a day where it can turn down, a few hours a day over a season can add up to quite a bit. If you have a $100/month bill it doesn't take much savings to pay for a $20 unit. Admittedly, when you need multiple controls, the math isn't as good, but if you're planning to be in the house for several more years, that's a lot of heating days to spread out the ROI.

BowlOfRed
  • 979
  • 6
  • 17
1

A smart thermostat will only save you money if your house is already very poorly insulated.

The same considerations apply whether we are heating or cooling, so I'll just talk about heating, and you can do the hot/cold/heat/cool inversion yourself where necessary.

You pay for the energy your house loses. If you allow your house to cool at times you do do not need it to be warm, then on average it will be cooler, and you will lose less heat to ambient. This will save you a little on heating costs.

However for there to be a significant difference in cost between your house staying warm, and cycling between cool and warm, the house obviously has to cycle between cool and warm. This implies significant heat gain when going from cool to warm (big heating system) and significant heat loss when going from warm to cool (big heat loss due to poor insulation).

With a well insulated house, the house doesn't cool much when not being heated, so the reduction in heat loss with a smart stat is far less significant. With a well insulated house, you might as well run the heating at a constant temperature 24/7.

If you insulate your house properly, including heat-recovery ventilation, there's very little difference between the performance smart and dumb thermostats. You'll have significantly reduced heating bills, because you're wasting a lot less energy.

Neil_UK
  • 584
  • 2
  • 10
0

One benefit of a programmable (not necessarily smart) thermostat that I recently saw in a youtube video is that you can schedule your heating / cooling to occur while energy rates and demand are lowest - i.e., overheat / overcool your house. For example, cool your house to 65 degrees overnight, and let it gradually warm up throughout the day to 75. Depending on your insulation, which way your house faces, etc you may not need to run your A/C at all during the day in this scenario, or if you do need to run it, you'd only be running it for a much shorter period.

You could make these setpoint changes manually, but it's also very easy to forget to change the change the set point when you wake up and you lose all the benefits.

mmathis
  • 8,754
  • 12
  • 41
  • 60
  • This is an awful idea in terms of wear on everything in your home (furniture, floors, walls,, bath fixtures, etc.) because oscillations in temperature and humidity come with expansion and contraction - which will in turn translate to more economic and environmental cost having to fix/replace things prematurely. – R.. GitHub STOP HELPING ICE Apr 20 '21 at 18:11
  • 2
    @R.. I don't see +/- 5 degrees as a huge problem in a well-sealed, controlled humidity situation, and I own musical instruments. I just don't see the point. The densest "thermal mass" material known is water @ 1.0000 BTU/lb/deg F. The problem is when you have HVAC plants throwing around 20,000 BTU/hr, if you want to store 10 hours of production at a 10 degree delta, you need 20,000 pounds of water (that's like 9 totes). I don't believe a house's internal structure *has that much thermal mass*. – Harper - Reinstate Monica Apr 20 '21 at 23:19
  • The bigger use for this is battery storage. Doesn't help for natural gas, which is more typical for heating in my area. But if you have a power wall or similar, charging at low cost time and using at peak time can actually save some money, but likely only if the upfront cost is covered by other reasons (instead of a genealogy,etc) – manassehkatz-Moving 2 Codidact Apr 21 '21 at 00:16
  • @manassehkatz-Moving2Codidact: If you signed up to pay different rates at different times of day you're going to end up like all the folks in Texas who were hit with $60k bills. It's a stupid game, like shorting stocks - bounded (and tiny) wins, unbounded losses. – R.. GitHub STOP HELPING ICE Apr 21 '21 at 00:21
  • @R..GitHubSTOPHELPINGICE Demand pricing (cost per kW for the highest demand period of each month) and time-of-day pricing (typically highest during the day, middle early morning and evening, lowest at night) are based on very real issues of how electricity is produced and used. How they are implemented, how generation (and distribution) is regulated, etc. make a huge difference in reliability and resiliency. No simple answers. – manassehkatz-Moving 2 Codidact Apr 21 '21 at 00:32
  • @R..GitHubSTOPHELPINGICE variable rate electricity pricing does not always equal wholesale pricing – mmathis Apr 21 '21 at 01:15
  • @R..GitHubSTOPHELPINGICE expansion / contraction is largely an effect of changing humidity, not temperature. It is a seasonal issue, not a day-to-day issue, and you likely can't overcool your house enough to make it a daily issue. – mmathis Apr 21 '21 at 01:18
0

I don't believe there is a clear answer despite all the theories/guesses that are purported as fact (not here, but elsewhere). So, I default to logical scenarios to better understand.

Imagine a tire is filled with compressed air, but has a tiny hole in it. The tire is the house. The air is the heat. The size of the hole is the quality of the house's insulation. As mentioned by @Harper, the more air pressure, the faster the air leaks out. But, there is much energy required to raise the pressure back up from a lower pressure, once or twice a day (when the people show/wake up).

People are not going to allow their homes to cool to all the way down to 40 degrees Fahrenheit (F) when they're away for the day, only to wait a long time for the home to completely warm back up when they return. So, to me, it ends up being the difference between the "leakiness of a tire" filled to 75 psi (or F) versus one at 68.

With this, my guess is that that difference translates to trivial money, annually, for a given household.

kackle123
  • 209
  • 1
  • 3
0

Lots of great guidance in other answers, let's apply it to your situation

Say you have a completely dumb thermostat, no timer, and you have periods where the home is unoccupied or you don't mind having less heat or AC while sleeping, so there is good opportunity for savings. Say all the arguments in the other answers favor a smart thermostat. The temperature differentials, insulation quality, and so on, would all lead to your saving money by getting one. THEN ... since you can get a NEST for free through your utility, or other smart thermostat for cheap, of course you should.

Say you have a cheap programmable thermostat, just a timer, and say you know your living patterns and you have programmed it appropriately so you minimize the amount of heating and cooling, but you do it fairly bluntly, given the constraints of the timer. IMO there is probably very little opportunity for savings here, but no harm in upgrading to a smarter stat if and only if it is free.

Say you have a thermostat in every room and they are all programmed to the living habits in that room, and you're looking at a $250 Google Nest for every room. Obviously this will be a monumental waste of money. You are not going to recover $250 per room if there is already a $20 programmable thermostat doing roughly what it should.

Now, mix and match the above scenarios, look at how many free stats you can get from PSEG, think about the guidance in the other answers, and let us know what you think!

But wait

You say your baseboard heat is controlled by individual thermostats. What about your A/C? Is that controlled by still more thermostats? How many? Is the potential savings from each new thermostat limited to about 50% because it only works for half the year? .... You'll have to work this out and do the math.

jay613
  • 28,850
  • 2
  • 36
  • 112
  • The answer from @gregmac contains some really good additional considerations. Thinking about that answer and mine together it boils down to this .... How smart is the new set up and how dumb is the old one? What is the potential for a smart stat to make a difference? And what does it cost? – jay613 Apr 20 '21 at 19:43
0

Smart Thermostats - really aren't that smart.

Yes they can run your fan a little longer but you can program some non-smart thermostats to do the same. Smart thermostats will learn after you go over and bump it for weeks... what you like.

Smart Thermostats are really good - if you are too lazy to program a 7 day thermostat.

If you are just too lazy to program a thermostat, sure they have value. If you can't remember to turn down the heat while you go skiing for a week. Yep might save you some money. But there isn't anything intrinsically that they are doing that is saving you money that couldn't be done with a programmed thermostat.

Why?

Other answers already explain this so won't get much into it but the items in your room have to be cooled an heated. If you are heating to a minimum and shutting off the items may never reach the desired temperature. This is the same concept of a refrigerator. A huge fridge that is empty is not really relatively cold. If you put warmer things into it the temperature could shoot up to 50F. Now if the fridge is 3/4 full of items and you put those same warm things in maybe it goes up a few degrees for 20 mins but will soon return.

What matters

  • your smart device doesn't recognize the draft under the door or by windows. It could if it had temp gauges throughout the house. This might be where we are going but it isn't where we are. So not very smart. Instead of playing with your smart thermostat taking the time to audit your house for the obvious issues, adding insulation to attic, next time your siding is changed - add insulation and air gap protection... these are things that save thousands not a fancy gadget that plays with your furnace non-stop.

  • shutting down vents in rooms that are either self conditioned or not used. There is more money to save in segmenting your home than anything else that can be done for free. Again the smart technology isn't there. They don't see that room A is 73F and no one is in it. The temp gauge is in room B at 67F and wants the house to be 69F. So it blasts hot air to room A... Room A is not only getting too hot but its taking resources from room B. So smart technology would include more temp gauges and smart vents!

  • what else isn't smart? Windows are not smart. They sure are smarter over in Europe -they have blinds on the outside of their windows. They shut the blinds during the day during the summer, especially ones with sun facing. This saves so much energy. Just like opening the blinds to the sun in the winter. So a smart thermostat would open and close these blinds to save tons of money per year - but they don't and we don't even have those blinds in the US.

DMoore
  • 47,296
  • 15
  • 79
  • 187
-1

The only way to save money on heating bills is to run the heater less.

If you can figure out a way to do this manually and can stick to it, you can then find a thermostat ("smart" or otherwise) that will allow you to automate that pattern.

Hoping a "smart" thermostat will magically find a way to save you heater run-time will lead to disappointment. The biggest challenge isn't the technical one of turning off the heat. The biggest challenge is convincing the occupants to keep the heat off... and that's a political/social challenge, not a technical one.

JS.
  • 407
  • 2
  • 5
  • This is *not* true when you are dealing with "blended" heating systems using multiple heat sources (example: most traditional North American split-system heatpump setups) -- maximizing heatpump utilization is an important part of getting your money's worth out of your heatpump! – ThreePhaseEel Apr 23 '21 at 00:35
  • I don't doubt you a right. But OP is asking if a smart thermostat will be better for his baseboard heaters than the ones he posted pictures of. My point is that if the Partner doesn't think the room is warm enough, that person is going to find a way to turn the heat up regardless of what a smart thermostat thinks. I'm open to being wrong. – JS. Apr 23 '21 at 00:54