Hello, I'm planning on building a house, climate zone 7 so quite cold. We want to insulate it well - at least R-30 for walls and R-60 for ceiling, with good windows. I'm exploring heating options and I've seen many people suggest that well-insulated houses can be heated with just a few mini-splits. I've been modeling house designs with BEOpt, and that can give the heat load for the entire house, but does not break anything down for individual rooms. Does anyone know of a tool (preferably free) to model heat transfer between rooms? I'd like to be able to say, the mini split is heating the common area to 69F but if I close the connected bedroom door, and the bedroom has 2 exterior walls, what temperature will the bedroom get down to on a cold night. Without an estimate of this number it makes me a lot more nervous to go with them. Thanks.

You can use software to calculate the heat loss to the outside and then manually calculate the heat gain from the interior and then find the balance point. But I agree that there should be something to do it all.

What you are desiring is a relatively simple heat transfer problem to solve if you know the steady state temps outside each wall (i.e., either the outside temp or the adjoining room temp) and if you know the area and R-values of each wall of the room that you are interested in determining what the unheated steady state temp will become. I think I could build you a calculator to accomplish this relatively easily and quickly. PM me and I will send it to you to try out and perhaps we can add it to the free DIY calculator section of our website.

If you calculate the heat load of a room you'll have the total BTU/hr that you'd be needing to supply via heat transfer through partition walls, etc. You can then roughly calculate the temperature difference across the partition wall area based on the approximate R-value of the partition walls (typically ~ R2) and the amount of wall area between the bedroom and partition wall on basic principles.

For example, say you have a 12' x' 12' x 9' corner bedroom that has 24' x 9 ' of running partition wall with the fully conditioned space, and say the design heat load of the room is 2000 BTU/hr (after factoring in the plug loads & sleeping human BTUs). That's 2000 BTU/hr moving through 216 square feet of partition, or about 9 BTU/hr per square foot. With an R2 partition wall & R2 door that imparts a delta-T of about 9/R2= 4.5F. To keep the bedroom a given temp, you'd have to keep the fully-conditioned space 4-5 F warmer than that on nights when the outdoor conditions will reach the 99% outside design temp.

Mind you in a US zone 7 climate with R30-ish walls it's unlikely the heat load will be that low, even if you had the minimum legal window area and triple-pane windows. With R50+ whole-wall (with all thermal bridging of the framing factored in) and min-legal-size U0.18 windows you might. Corner bedrooms would be the toughest, since they have the most exterior wall losses. There are existence proofs of houses in my area where it works with R45-R50 walls and somewhat more generous window area, but the 99% outside design temps here are 0F to low positive single digits, whereas zone 7 design temps are fully 20F colder than that, which will constrain it a bit.

http://www.greenbuildingadvisor.com/blogs/dept/musings/just-two-minisplits-heat-and-cool-whole-house

I get R4.6 using data from here. That's a huge percentage difference that makes it worth discussing.

Thanks for the responses.

sailawayrb: I sent you a PM, and yes, it would be great if there was an easy way of doing this as it seems like other people have this question. I am trying to understand the math of it all too.

Dana1: I have a question about your post. I'm not quite following you, and I'm not sure if that's because of a typo or I'm just missing something. So if my room has a 2000 BTU/hr design heat load, and there is 216 square feet of partition wall, then the room needs 9 BTU/hr to go into the room from the warmer space, per square foot. So far I'm following. But then you take 9/R2 = 4.5F to figure out the delta-T, and this is where I'm confused. Say my interior wall was insulated to R-4 instead of R-2, then I would be doing 9/R4 = 2.25F, right? But I would expect that if the interior wall were insulated more then the rest of the conditioned space would need a HIGHER temperature in order to maintain the temperature of the room, right? I am misunderstanding how this works?

jonr: Yes, it looks like if an interior wall is truly R4.6 that would make a huge difference. My walls would just be pretty normal with drywall. The "air film" on that website looks to be a big factor, if I read it the same way as you are.

BTW, my 99% design load is around -13F for this area, so quite cold. My target inside temperature is 69F. I think my family could tolerate around 60F in bedrooms, but I don't know that I'd want to go colder than that, so that's why I'm looking into these calculations before planning anything.

I'm also wondering how much of a factor HRV ventilation makes. If each room has a vent and the house has 0.35 air changes per hour, then wouldn't the air mixing from all the rooms be a pretty significant factor in reducing temperature differences?

Posted By jonr on 22 Aug 2014 04:12 PM
I get R4.6 using data from here. That's a huge percentage difference that makes it worth discussing.

You get about R1 out of the 2 layers of gypsum, another R1.5-3 out of the air-films, assuming it's dead-still air.  With convection it's really about R2 type performance.

If you assume R0.68 per air film and R0.45 you get a center-cavity R of

0.68 bedroom side air film
0.45 wallboard
0.68 cavity side air film

0.68 cavity side air film
0.45 wallboard
0.68 conditioned space air film

That totals up to  R3.62.

The framing fraction' R value  comes in at about R5, assuming R1.2/inch wood.  Partition walls have a framing fraction of about 12% (single plates, no window headers footers) so in a dead-calm low delta-T situation you're looking at about R3.75 whole-wall.

But in reality convection undercuts it by quite a bit- nearly in half. Those air-films are less stable than you think, and the air films inside the wall cavity convect freely at fairly low delta-Ts, and even the radiated heat transfer between the layers of gypsum starts to count once you're over 3F for a delta-T.

If you want to be super-conservative assume R2.5 or R3 for the partition walls, but an approximation of ~R2 seems to better match the (not third party verified but) measured delta-Ts.

I would assume that the interior air film R values already include convection effects (which is why they have different values for external films). And correct or not, I added in the "air space" listed at the bottom (R1). And ignored framing.

> But then you take 9/R2 = 4.5F to figure out the delta-T, and this is where I'm confused.

One needs to multiply, not divide by the R value. So yes, these fixes lead to a quite cold temp estimate, well below comfortable levels.

I PMed you back DarkNova and can send you the calculator for testing. Once we exercise and validate it, we will add it to our website too. It can currently handle 10 total room assemblies (i.e., ceilings, floors, walls, doors, windows, etc.) that could make up a room. You just enter the temps that are outside of each assembly, area of each assembly, and R-value of each assembly (if you have door or window U-factors, just inverse them to get the R-values before entering them), and the calculator will determine the steady state room temp.

And here it is:

Borst Steady State Room Temp Analysis Software

Edited to update the above calculator link and we also added the associated instructions for this calculator to our DIY software instructions manual:

Borst DIY Software Instructions Manual

Posted By Dana1 on 22 Aug 2014 01:57 PM
If you calculate the heat load of a room you'll have the total BTU/hr that you'd be needing to supply via heat transfer through partition walls, etc. You can then roughly calculate the temperature difference across the partition wall area based on the approximate R-value of the partition walls (typically ~ R2) and the amount of wall area between the bedroom and partition wall on basic principles.

For example, say you have a 12' x' 12' x 9' corner bedroom that has 24' x 9 ' of running partition wall with the fully conditioned space, and say the design heat load of the room is 2000 BTU/hr (after factoring in the plug loads & sleeping human BTUs). That's 2000 BTU/hr moving through 216 square feet of partition, or about 9 BTU/hr per square foot. With an R2 partition wall & R2 door that imparts a delta-T of about 9/R2= 4.5F. To keep the bedroom a given temp, you'd have to keep the fully-conditioned space 4-5 F warmer than that on nights when the outdoor conditions will reach the 99% outside design temp.

Mind you in a US zone 7 climate with R30-ish walls it's unlikely the heat load will be that low, even if you had the minimum legal window area and triple-pane windows. With R50+ whole-wall (with all thermal bridging of the framing factored in) and min-legal-size U0.18 windows you might. Corner bedrooms would be the toughest, since they have the most exterior wall losses. There are existence proofs of houses in my area where it works with R45-R50 walls and somewhat more generous window area, but the 99% outside design temps here are 0F to low positive single digits, whereas zone 7 design temps are fully 20F colder than that, which will constrain it a bit.

http://www.greenbuildingadvisor.com/blogs/dept/musings/just-two-minisplits-heat-and-cool-whole-house

That house is super tiny.  I assume it to be 1232 of total living space, and not per floor.  I don't think most people are looking at a 2-story that small.  The heating requirement of such a small house with such massive walls doesn't make sense to me.  I think that $330,000 for that house is pretty expensive.  You could easily build a much larger house for less using a normal hvac system, and the money you saved would probably pay for your utilities for 30 years, i.e. I think you could build an above code minimum house for $230,000, and then you would have $3300/yr to spend on utilities.

I just don't think the house in the that example would really suit many people.

First of all, you need to determine if mini-splits (or any heat pump) are capable of heating in your climate. The low limit is about 13-15 F. degrees below zero. Heat pumps get lower in efficiency as it gets colder, too. What is the lowest expected temperature in your area?

We are in a mild area, zone 3-4, and have found that one mini head keeps our 1400 square foot ICF home comfortable, at least when A/C. We haven't tested it in the winter yet. That being said, we keep the interior doors open all the time. There is usually less than 1 degree difference throughout the house. If you want to keep the doors between rooms closed, you could always put in vents, possibly even powered ones. A very small, quiet bathroom fan, coupled with a vent in the bottom of the door, or even just a healthy clearance at the bottom of the door would go a long way towards balancing out temperatures (and air quality) between rooms. In a really tight house, you are going to need an HRV/ERV anyhow. I wonder if there is a way to incorporate this system to circulate air between rooms?

I would have never believed that a single HVAC source could keep a whole (small) house comfortable, but I've never lived in a tight, well-insulated house before.

I think that $330,000 for that house is pretty expensive

The house you are referring to didn't cost $330,000. It sold for $195,200, which included the land and builder profit. What it says in the article is that Mr. Scott's company plans to build a number of other very efficient homes, some with 4 bedrooms that will sell for up to $330,000.

You could easily build a much larger house for less using a normal hvac system

That's the same greedy thinking that has caused the very problem people are trying to solve. A major tenet of Green Building is to build homes that don't waste energy just because it is unaccountably cheap. BTW, Carter Scott notes that a "normal" HVAC system for this house would have cost his company about $14,000 and they did this one for $7,500, or about half. Building more efficient envelopes pays off in terms of spending less on the heating plant to begin with.

I don't think most people are looking at a 2-story that small.

That's exactly the kind of home a number of new families are looking at, as well as a number of older couples where the kids have moved out and the parents are looking to downsize. The difference might be in the level of finish afforded. A nice feature of the energy-efficient homes for both couples starting out and for retiring couples is that the energy costs are more predictable and controllable, making it easier on budgets.

First of all, you need to determine if mini-splits (or any heat pump) are capable of heating in your climate.

I wouldn't hesitate to make limited use of electric heat - either because the HP can't always keep up or you want to occasionally close a door.

Re assist from a HRV, small amounts of air movement near room temps move very little heat.

CFM x 1.08 x temperature delta = BTU/hr

So say for a HRV:

10cfm * 1.08 * 10F = 108 BTU/hr

On other other hand, two duct/exhaust/register fans in a push/pull configuration (never pressurize or depressurize a room significantly):

200cfm * 1.08 * 10F = 2160 BTU/hr
That's actually useful in a well insulated bedroom. Might mask noise too.

Posted By ICFHybrid on 23 Aug 2014 10:45 AM

I think that $330,000 for that house is pretty expensive

The house you are referring to didn't cost $330,000. It sold for $195,200, which included the land and builder profit. What it says in the article is that Mr. Scott's company plans to build a number of other very efficient homes, some with 4 bedrooms that will sell for up to $330,000.

You could easily build a much larger house for less using a normal hvac system

That's the same greedy thinking that has caused the very problem people are trying to solve. A major tenet of Green Building is to build homes that don't waste energy just because it is unaccountably cheap. BTW, Carter Scott notes that a "normal" HVAC system for this house would have cost his company about $14,000 and they did this one for $7,500, or about half. Building more efficient envelopes pays off in terms of spending less on the heating plant to begin with.

I don't think most people are looking at a 2-story that small.

That's exactly the kind of home a number of new families are looking at, as well as a number of older couples where the kids have moved out and the parents are looking to downsize. The difference might be in the level of finish afforded. A nice feature of the energy-efficient homes for both couples starting out and for retiring couples is that the energy costs are more predictable and controllable, making it easier on budgets.

In one town, the houses start at $330,000.   Of course, you picked the town which says they go up to $330,000.  Whatever.  They don't specify the size.  It isn't greed.  It is the the concept of diminishing returns.  He is putting 17kw on some of these houses.  They probably only need 2kw.  It doesn't make sense.

I found a few more articles about his homes, and they say they are saving $500/month in utilities.  My sister who lives in Massachusetts in the crappiest house in the world with hardly any insulation, wet basement, minimum attic insulation, 2 story house with a basement, doesn't use $500/month in utilities, so it would be impossible to save that much/month.  Her house is 1450 sq ft and 45 years old.

So, I am doubtful of most of the numbers that people quote.  Maybe, he gave us his highest quote that he wouldn't use.  Who knows.

I don't think most older people want to buy a small 2 story house.  Maybe, in your part of the world, but most people say when they get older, they don't want to climb steps.

Thanks for the comments and calculations.

And thanks to Borst Engineering for that steady-state room calculator. It works well.

It showed that at a -13F temperature, in a corner bedroom with good windows and good insulation, with 69F in the rest of the house, that room could get down to a steady-state of 57.5F. Unfortunately that is pretty low and makes me second-guess pursuing mini splits for our climate.

I will still look at them more, as coupled with fans it could still possibly work, but as jonr showed it takes quite a bit of air movement to even out the temperatures.

The reason I am interested in pursuing an electrical heating option is that this is a rural property, so it is either electricity or propane. Propane isn't cheap, and our electrical rates are really cheap (~ $0.06/kWh) so it seems like the way to go.

GSHP/geothermal has a high up-front cost that would probably not be paid back in decades on a well insulated house.

I would be interested in an air source heat pump that could be fed into a central ducting system, but I have not seen one so far that can perform down to cold temperatures like the Hyper Heat Mini Splits.

If anyone has any other suggestions to look at, I'd be grateful.

At that price, ASHPs plus some electric baseboard assist for closed door rooms on cold days makes sense to me.

If you have a well anyway, open loop geo might be worth investigating. Hardness in the water is not much of an issue when using it only for heating.

They probably only need 2kw. It doesn't make sense.

I doubt very much that the contractor is putting arrays on that are larger than what the home needs A 2 kW array would generate roughly 8-12 kWh per day. Since the average home uses 1200 kWh per month, 240-360 kWh would represent a substantial shortfall. In addition, the homes are using electricity for the heat pumps, so it is likely they use at least the average.

If anyone has any other suggestions to look at, I'd be grateful.

The Daikin Altherma air to water air source heat pump works with fan coils and it comes in several sizes, 36kBTU - 54kBTU. It also supplies your domestic hot water and can tie in with solar. Daikin now has a low temperature unit that operates down to -13F and delivers full rated output at 5F, but you can also get them with backup resistance heat coils.

You are welcome DarkNova and glad you found it useful. We are going to add a couple more inputs that will allow including uncontrolled infiltration and/or controlled ventilation between the closed room and the exposed temps.

Much of our construction is in extremely rural areas where electricity, propane and wood are the only fuel options. If the landowner is lucky, they already have access to electricity via utility grid. However, if they don’t, it can often cost well in excess of $100K to get utility grid electricity to their home site. So, more times than not, electricity has to be generated onsite via engine-driven generators, hydro-driven generators or PV panels. However, one advantage of using electricity is that it can be converted to heat with nearly 100% thermal conversion efficiency and used to move heat via a heat pump at very high COP. So if you have created the onsite infrastructure to enable having low cost electricity without creating pollution or otherwise damaging the environment, electricity can be a great green heating option.