-
I don't see how anyone could consider 15°C to be cold,
The Earth needs more people like you. As for me, any inside temperature lower than 22 C is cold (in autumn/winter).
(because of cold external wall, small but cold droughts).
-
Short RC indeed improves controllability, but for a high cost: small R means high loss of energy. Insulation slows down adjustments but quite obviously reduces the consumption.
Increasing R reduces the consumption, at a cost of raised investment, quite obviously. Had you been offered an R for free - go for it. However, keeping same R for two houses with vastly different RC constants and using them for N-th part of the time (lets say RC1>>RC2, N=1 / N=2 examples), the second house would require half of the energy of the first one. So instead of investing into raising R, you can invest into lowering RC, reaching same total costs. In this sense, I disagree with the quote.
I'd rephrase: Heating well insulated house 24/7 costs same as heating a not so well insulated low RC house occupied 1/N-th of the time. Whether this tie is for N=1.1 or N=13 for specific design is a different story. -
Depends, of course, as said above, on various factors.
One thing in favor of heating at all times is managing humidity. RH can be pretty high indoors during fall and winter in the absence of heating, depending on where you live, the kind of house, etc. This can be nasty.
Of course, if you go for constant heating, you can set it for lower temperatures when you're away. Just enough to manage humidity levels and make it reach a comfortable temperature in just a few minutes when you get home.
-
Homes built with below slab insulation and minimal above slab insulation will have too much thermal mass to get to a comfortable temperature from cold in a reasonable time frame.
well, it should also mean that it stays at a comfortable temperature longer
-
Short RC indeed improves controllability, but for a high cost: small R means high loss of energy. Insulation slows down adjustments but quite obviously reduces the consumption.
Increasing R reduces the consumption, at a cost of raised investment, quite obviously. Had you been offered an R for free - go for it. However, keeping same R for two houses with vastly different RC constants and using them for N-th part of the time (lets say RC1>>RC2, N=1 / N=2 examples), the second house would require half of the energy of the first one. So instead of investing into raising R, you can invest into lowering RC, reaching same total costs. In this sense, I disagree with the quote.
I'd rephrase: Heating well insulated house 24/7 costs same as heating a not so well insulated low RC house occupied 1/N-th of the time. Whether this tie is for N=1.1 or N=13 for specific design is a different story.
Yes you are right, and as a consequence, houses that are not constantly occupied do not require as much insulation, it would be money wasted. This holds even in very cold climates. It's an easy thing to do if duty cycle is a few % and the period is a month (for example, spending weekends somewhere once a month).
Now the question is what to do when duty cycle is say 50% but period is as short as 24hrs - people live in the house but go to work and so on.
Because given cold enough climate - Canada or Northern Europe for example - amount of insulation absolutely needed to make the houses liveable without exploding heating costs pretty much forces the RC time constant in the range of a few days or more, which then prevents optimizing on a 24hrs cycle period, but still allows doing that with longer cycle (holiday reduction, etc.).
But really, the equivalent circuit is parallel RC with a current source (heating supply) feeding power into it, voltage being room temperature. Small R (poor insulation) seemingly increases controllability giving the false impression that you can save energy by "dropping" the temperature, but in reality that drop is caused by wasting the energy so you are not saving anything compared to having higher R.
Except, as you mention, high R has an investment cost, you do save by not having to invest into insulation materials and thicker walls and more expensive windows and so on. This is a classical optimization problem which can be performed in Excel quite well.
In my opinion, designing buildings that last for a long time is the best idea. When building, insulating properly is not that big of an extra expense and it pays back within the first decade or so, and generates savings for the next 50-100 years. I don't like the idea of buildings lasting for only 30-40 years.
Retrofitting insulation is a significantly higher expense, usually. -
(..) Yes you are right, and as a consequence, houses that are not constantly occupied do not require as much insulation (..). Now the question is what to do when duty cycle is say 50% but period is as short as 24hrs - people live in the house but go to work and so on.
I am glad you asked. Forget about Finland and EU, lets move to the Dreamland for a moment. If we relax this problem of multitude of constraints, it can be quite easy to estimate the potential of OP's concept/question.
Imagine a house with super low RC, but with standard R. In terms of engineering design, imagine a model, a free standing single room house, insulated from INSIDE, with floors and walls layout and furniture made out of low Cp*mass materials and machinery for injecting/extracting heat. Not a concrete slab, in other words. Imagine RC=5s.
My conclusion is that in Dreamland a room only needs the amount of energy that goes through the walls when there is someone in it, and only this part of the energy is used really. All the remaining energy for heating demand is not used but wasted. Under this definition, when you are not inside, there is no need for the temperature inside to be different than outside temperature, when no energy escapes through the insulation, losses=0.
Now, I do not know how about the rest of you but since usually people can typically be present only in one place at a time, this gives an estimation of what this setback concept is worth, the bordering case. Had buildings been built with low RC, these would have required only the fraction of the energy that comes from the occupancy of rooms. So if P people live in Q room apartment, sleep 1/3 of the time, leave for shopping, work, etc, there is no way they could all be in more than P places at the same time.
Of course once you return back to Finland, add more constraints, insulated internal walls, RH, high powers required to heat up room to temperature in seconds, things get messy. But the lesson from it is that, depending on occupancy, there is a potential insavingnot wasting huge amounts of energy. Alternatively, keeping wasting same amounts of energy but at lower investment cost. -
Yes, your model is correctly build but it's impractical, it doesn't match with reality. But it's really useful demonstration to aid understanding.
Add thermal insulation to internal walls, build with low-C materials (so that significant part of heat capacity is in the air), and install a ducting system with shut-off valves and huge fan units so that you can, at any time, swap air between any two arbitrary rooms using laminar flow to prevent air mixing. Now you can keep one room heated, and when you are entering another room, the warm air is quickly swapped to that room, bringing it up to heat in seconds, while cooling down the room where you were earlier. You can also install water pipes in all inner walls, floor and ceiling, even inside heavy furniture, and use powerful (think about megawatts) heatpumps with ultracapacitor supply to swap the heat between rooms in seconds.
Doing this, average loss through the outer envelope of the house is only that of one room, yet you always have a heated room wherever you are!
It's left to the reader to decide if this could ever work in practice.
I have a better idea for anyone who has much larger number of rooms than occupants and are really concerned about their energy usage: move to a smaller house, with fewer rooms
.
Or, just insulate, and failing that, do the dynamic temperature control tricks as discussed, they do work and have quite some real saving potentials in poorly insulated houses, there is no question about it. If your temperature drops quickly after turning off the heat, that's a sign that you are losing quite a lot of energy due to limited insulation, but that's also a sign that you are saving every minute by turning off that heat. -
Yes, your model is correctly build but it's impractical, it doesn't match with reality.
By definition.
The purpose of model is not to convince someone to install water pipes in furniture. It is to show that total cost of heating the building (investment + energy) is a continuous function of RC step response of powers installed. For ridiculously short step response and high powers this total cost is astronomical. For ridiculously long RC this total cost is also higher than necessary because of the energy and insulation needed to justify heating 24/7 when in reality you are not everywhere at the same time. So this total cost has to have minimum, with minimum location that heavily depends on the occupancy. For occupancy=1 the 24/7 house is the solution. But for all other occupancies it is not. And assuming we are talking here about mammals who sleep 1/3 of the time, the 24/7 house should have never happened. IMHO. -
Low RC has advantages.
The problem is, with real building materials including furniture etc., C can't be arbitrarily low, so in reality, low RC is "achieved" by low R. This changes the "low RC advantages" into "low R workarounds". Sure, they still help, given that R is fixed and can't be changed (and this is quite fair, retrofitting insulation is many times more expensive than just do it right the first time).
However, you did miss my earlier points that high RC has different set of advantages.
I'm sure you have noticed energy storage is now a really hot topic, and for a good reason. We already have hourly energy pricing as an option, and I'm sure it will be forced down our throats want it or not. Uncertain and uneven production of renewables, and uneven consumption are the root causes, and it's not getting any better.
Another point is, heatpump efficiency is not a constant.
You can fight this by having Tesla install massive grid-connected li-ion packs and that's fine too, but that's not the only way.
Large RC households offer distributed energy storage capability with little effort and cost (often it already exists, and only lacks monitoring and control). If you have that concrete slab and decent insulation, you can just cut heating for the most costly hours no problem. This can reverse the whole thing; I'm running a heat pump during daytime, when my PV installation is generating, and outdoor temperature is at highest, so that heatpump efficiency is at highest as well. But I have fairly stable indoor temperature because the house itself has quite decent R, mediocre but not too low C, with another C (1200 liter water tank) in parallel. Mixing valve to that tank increases controllability but for now it just stays fully open. -
well, it should also mean that it stays at a comfortable temperature longer
When it's not needed.
At equilibrium to keep at a constant temperature takes the same power regardless of how much thermal mass is inside the insulation. Ignoring power failure all thermal mass does is force you to start heating earlier when you need it and carry heat to times when you don't need it. Thermal mass in your home is not terribly useful storage, you generally want to keep storage separate from your living room, so you can heat it to say 60 degrees Celsius and not die. -
... and when building from scratch, cost optimization for amount of insulation is quite simple. I did rebuild the upstairs here, completely insulating the attic, and did "napkin" calculation in Excel. I ended up with average 180mm of PIR (polyisocyanourate) sheet because calculated that way, payback time for increasing further 50% from there would be well over 20 years. But going from say 80mm to that 180mm pays for itself in just 10 years.
And the cost of energy source of course affects that payback time. Choices of poor insulation have been always made when energy has been cheap. -
Thermal mass is not useful storage, you generally want to keep storage separate from your living room, so you can heat it to say 60 degrees Celsius and not die.
This is demonstrably just a false claim. Thermal mass is very useful energy storage as proved by myself and many others. Of course the amount of energy that can be stored is limited by low dT range available, but it's still not meaningless. It's in the range of some 5-50 kWh depending on the house and living standards (how much temperature variation is acceptable), and can carry over high-cost hours with practically no further investments at all.
Similar li-ion storage system would cost like some $3000-5000.
Those who have a thick concrete slab with in-floor heating pipes and a heatpump, notice the advantages immediately by making the heatpump run by daytime.
Sure, on paper, given same R, it's best to use as small C as possible, then use external C (like that water tank) for more controllability. But this also costs more to build. -
well, it should also mean that it stays at a comfortable temperature longer
When it's not needed.
... and if you really want to simulate the leak-the-heat-out behavior of poorly insulated house, nothing prevents you from actively cooling a room with a heat pump, and putting that heat into a storage tank. Obviously such high-tech solution never pays back for itself, but this is just the demonstration of the fact that poorly insulated house leaking out the heat is not an advantage even if it seemingly gives you control "by only having to heat when heat is needed".
-
Simplified,
Heating(t) = k*(t_in - t_out) - sun_through_windows - human_generated_heat - non_heating_electric_power, when t_in > t_out
Cooling(t) = k*(t_out - t_in) + sun_through_windows + human_generated_heat + non_heating_electric_power, when t_out > t_in
Without heat capacity,
energy consumption = integral over time of (COP_h*heating + COP_c*cooling)
Every integration time step will have either heating or cooling.
With heat capacity, heating and cooling needs can be averaged together, and so the integral will be smaller. Ignoring complex math, you get into the ballpark by taking the average during the RC time constant. For example, with 24 hrs:
Heating(t) = k*(t_in_avg_24h - t_out_avg_24h) - sun_through_windows - human_generated_heat - non_heating_electric_power
Cooling(t) = k*(t_out_avg_24h - t_in_avg_24h) + sun_through_windows + human_generated_heat + non_heating_electric_power
energy consumption = integral over time of (COP_h*heating + COP_c*cooling)
With conditions changing between heating and cooling, the latter integral will be significantly smaller.
This is well evidenced in old massive brick houses of pre-WW1 where significant part of the year goes without any heating power yet they are quite comfortable in summer even without cooling, and reportedly measured energy consumption is significantly lower than calculated based on the U-values of the structures.
At the same time, many "passive homes" or "low-energy homes" of 2000's have failed to get even close to the calculated near-zero energy consumption because they have low C, they overheat already early in spring and require constant switching between heating/cooling to be comfortable. But their consumption has been calculated by taking the sun into account as a positive only, but when the calculated heating power goes into negative, it has been just clipped to zero, causing temporary overheating, which wont't be automatically stored due to low C. But people won't accept that so they let the aircon to switch into cooling mode, following what would have been the correct calculation (negative heating power in the integral -> cooling power).
The high-C old brick houses overheat much less due to sunshine, and that little amount of overheating carries long into the night and morning.
The trick here is that the old high-C brick houses also have surprisingly high R for their age. C is not any good if the insulation leaks the power out.
Value of R is most critical, different ranges of C can be worked around and have different advantages but claims that high C is definitely bad are wrong. In my opinion, high C have more advantages than low C, but only if R is properly chosen for the climate. This is an opinion and I might change it seeing different evidence.
-
Attached is a simulation (yes, I'm working on this problem) that fairly well matches my house, simulating May-June 2012*, model includes temperature, wind, solar radiation from different data sources (PV generation simulation is also included for cost analysis, but let's skip that for now), also waste power from electricity used indoors.
*) Why 2012? Because it happened to be a very "average" year here, also containing a balanced mix of extremes, so good for design of systems. Last few years have been warmer.
The top graph shows, in blue, the calculated heating power demand, in Watts, to keep +21degC indoor temperature. Negative values mean cooling is required. Orange line specifically shows the solar irradiance input power.
Mid graph shows outdoor temperature.
The last one is interesting as it is the simulation of the room temperature, just for this discussion I chose to set heating/cooling power to constant zero. Now the blue graph is resulting room temperature if thermal capacity of the house is 1kWh/degC, orange graph when it's 10 kWh/degC. The heating/cooling cost is the same by definition (zero in this simulation). Which one you would prefer to live in?
-
Now get it to 21 degrees 50% of the time, starting early enough to get it there 50% of the time, and let it coast for the rest of the time. Which power bill would you rather pay?
-
OK, I'll run the sim. But hold on a bit, I want to make sure it goes right. I'm happy to be shown wrong so if necessary I can fix my understanding, because long term I can't afford putting a biased opinion ahead of facts, because this simulation and understanding the subtleties of the control is the basis of the business and I need the model to correctly represent and control many different types of houses, including very low and very high thermal capacity.
-
OK, here are some results.
Now in the model, heating is arranged with simple electronic thermostat, which senses current indoor temperature and provides a variable heating power using a steep P controller (Google "heat anticipator" for explanation of the classic mechanical PWM device), capping max power to 10kW.
Again, at the top/blue is the calculated "required" heating/cooling power if steady 21degC would be desired. This info is not used in calculation, just for reference. Top/orange is the solar irradiance power. Again, second graph is outdoor temp. Third graph is simulated room temp. Fourth graph is cumulative output energy.
Blue is low thermal capacity house (3kWh/K), orange is high thermal capacity house (15kWh/K). Insulation (i.e., average heat flow through envelope) is the same.
First, bear with me, without hourly cutting of power
see thermo1.png
As you can see, the cumulative energy drifts apart whenever there is overheating due to solar irradiance. Higher thermal capacity house shows reduced consumption right after such sunny days as it stores the energy. This is even with a "dumb" thermostat which does not understand the thermal capacity. Two month consumption, 1015.8kWh vs. 976.53kWh in favor of higher thermal capacity house.
Now let's cut the power to both houses. Let's cut it for 12 hours per day, say between 6am to 6pm when going to work and, in very non-Finnish way, eat out after work.
see thermo2.png
Two month consumption is now to 948.45 and 948.56 kWh, basically the same result. As you can see, the average temperature of the low-C house dropped, which also dropped its consumption, but not very much. With high-C house, consumption also dropped, but not as much; but it's still as good.
You asked to "start early enough to get it there". Now this expectation comes from the mindset of low thermal capacity. I didn't implement such logic in the model, but as you can see the high thermal house keeps decent nice indoor temperature without such trickery. Here's a zoom-in on transient response:
see thermo3.png
So while you can see that high-C house heats up slower, it starts from higher temperature. You decide which one is better, especially if you miss-schedule and fail to predict your living patterns.
This example data was not cherry picked. I'm sure I could find examples of both high-C building showing significantly better result (now the difference was marginal), but also could find data to prove your point where low-C building shows better result.
-
... I made one more simulation where I swapped the on/off time, turning heat off overnight, from 6pm to 6am. With such case, lower C house wins, probably because the reduction now coincides with the time of highest demand so there is potential to drop the average temperature so much it obviously finally does what was discussed in this thread.
Consumption:
ans = 818.80
ans = 924.00
But who wants to wake up in such cold rooms. If heating is started already at 4 a.m., the difference reduces:
ans = 885.60
ans = 942.45
-
I thought high performance windows we're pretty much energy neutral in cold weather on a sunny day? (ie. no solar heating in winter.)
-
Depends on windows and temperatures but for example here the legislation now requires windows to have U < 1.0 W/(m^2 K), and low energy windows down to some U = 0.5 are available, this already equals the insulation level of the walls just a few decades ago!
Passive/low energy houses were built here in 2000's with such good materials but fairly low thermal capacity, and with a lot of those energy efficient windows (4-glass, with argon filling on the 3-glass element, with spectrally selective coating) facing South. The result was, houses already overheat in early spring. So need to bypass the ventilation (which obviously has thermal recovery) and vent manually. And then when the night comes, heat again. This causes such houses to show significantly, like 20-30% higher energy consumption that was calculated on paper.
In any case, my model uses a single 1.5m * 1.1m Window facing South-30deg because that what I have in the 1950's house. If I triple that window area (while keeping my approximately 0.55 U value, i.e., the new windows would be the fancy 4-glass version), the benefit of high-C widens (see attached thermo6.png).
OTOH, if we completely remove south-facing windows, then as expected, the low-C solution gets more benefit from the 50% power cutting (see thermo7.png).
So I think I'm right about the fact that low-C and high-C solutions have different advantages. And again, IMHO, the advantages of high-C are higher because even if the energy consumption averages to similar values, high-C is less sensitive about prediction errors in human patterns.
-
One aspect often overlooked is the layout of the house (or apartment) and the possibility to create zones with different temperatures. Bedrooms, hallways and storage spaces with lower, living rooms and offices with higher temperature. Not having to heat the total volume to comfortable temperature saves a lot of energy.
-
Technology Connections on youtube did a video a few months ago where he figured out that one room in the house could be made to act as a battery for cool air during the off peak power time and in the evening (when he's home) deliver that cool air to the rest of the house.
Very clever. -
It can be complicated. My house has an electric heat pump, and so it has better efficiency when it is 12C than -7C... according to the documentation that's a factor of 2 in efficiency.
Given that I work 8-5, and pay the same rate for electricity at all hours of the day, I have the thermostat scheduled to drop 3C at 7:30 and to turn back up at 4:30.
So if you imagine a day where the morning low is 0C and the afternoon high is 13C, I effectively eliminate running the heat when it is 1.2 or 1.3 efficient, and run it instead at the warmest part of the day when it is 1.9 or 2.0 efficient (all relative to that -7C figure).
If you change the conditions, the answers probably change:
With rooftop solar power, you'd want to to use as much of that as possible, meaning running the heat all day
If the power/fuel has variable pricing through the day, it may be beneficial to run the heat to take advantage of that.
If the house was less well insulated and the temperature fell by 6C during the work day, you probably would want to run the heat during the day.
I wish there was better tooling to help homeowners optimize their energy use/cost. It seems like some of the "smart thermostats" on the market could provide recommendations, but I'm not aware of any that do so. -
I wish there was better tooling to help homeowners optimize their energy use/cost. It seems like some of the "smart thermostats" on the market could provide recommendations, but I'm not aware of any that do so.
I have noticed exactly the same and think there is a lot to do here with little actual/serious competition. But we'll see.
Due to lack of real controls, I simply let my heatpump do the work between 10am to 4am and let it rest when it's coldest. Also having rooftop solar, the operation starting at 10am coincides pretty well with the production.
But lack of control also means lack of instrumentation. Even if instrumentation is available (for example, you can install output energy meter in a hydronic system (one that uses accurate flow meter and two accurate thermometers to calculate power), how do you verify if the actions have the effect you expect? Run for a year, then change your habits for another year, while hoping the weather patterns are identical year to year which they are not?
Large changes such as replacing COP1 sources with COP3 heatpumps are obviously visible as $$$ saved in bills, but how do you verify what the effect of dynamically changing temperature setpoints is? The expectation on the forums seem to be they hold a potential for significant savings, but there is little to prove that except too simplistic napkin calculations. Expect of some low hanging fruit cases, they are micro-optimizations which are easily lost in noise. As you can see from my simulations posted there is no significant difference in energy consumption in any of the cases.
The only solution I can see is simulation with capable enough simulation models. It helps that thermal energy flows are really analogous to electronic design principles and elements like resistors, capacitors, current and voltage sources can be used with Kirchoff laws.