Sell me on geothermal vs. building envelope upgrades and an air-source heat pump (new build)
I know these questions are hard to answer online, but I will do my best. Most of the information I receive locally is from HVAC suppliers pushing a particular system.
So, a little information that will help.
We are building (in June) in Northcentral PA, in zone 5, on 25 acres. We’re at ~1000ft of elevation with a mixture of shale/loam soils. We have no access to NG. It’s Propane, oil, or electric. Current rate is 0.070390/kWh. We have ~6300 HDDs.
We’re building a single story with a fully exposed walkout basement. 2986 sq.ft on the first floor as well as the basement. Basement will be ICF with R10 subslap insulation. I’m putting Creatherm panels and running per for a future hydronic system. On the main floor, it’s traditional 2×6 with NuWool, and R49 spray in ceilings. We’re using Low E Paradigm windows, and have ~781 sq.ft of glass.
We were quoted by two different contractors for a Waterfurnace 5 series, 4ton system, single zone with electric backup at ~28,000 before tax credits. That was for vertical well systems. Only one contractor was willing to do a horizontal loop, and that came in at $24,o00. Adding a desuperheater, and extra zones would drive the cost well north into the mid to upper 30,000 range before tax credits.
Right now, we’re considering adding foam sheathing to the outside of the building envelope to create a thermal bridge an increase our R value above code (waiting on the quote from the contractor). I think in doing that, along with the insulated slab/hyrdonic heat, that we might end up better in the long run with a higher efficiency ASHP + the hydronic and possible wood stove for rare sub zero nights/days.
Granted, I’m not a pro, so I may be overlooking some obvious things, or maybe I need better estimates.
I would love to hear everyone’s thoughts on this.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Your plan to improve your home's thermal envelope, and to specify one or more air-source heat pumps instead of an expensive ground-source heat pump system, makes sense to me.
Note that adding a continuous layer of exterior rigid foam to your walls won't "create a thermal bridge" -- it will actually greatly reduce thermal bridging. (Which is what you want, of course. A thermal bridge is not desirable.)
Yes, I had typed this up twice only to have the site lock up when I went to post this morning. I suppose I was rushing that third post. I meant to say, "reduce thermal bridging."
You might rethink the amount of glazing. 781 square feet is a lot. It'll require a lot of heating energy input to overcome all that R-4 ish area.
Are you sure that .07 per kwh includes both power and delivery? That's super cheap.
As for the rate, I was mistyping as that was just the transmission charge. That and the distribution charge totals out at .1104/kWh.
You should check if your utility offers a “time of day” rate. These rates charge more per kWh during on-peak time (usually late morning through early evening on weekdays), and then significantly less for off-peak (nights and weekends). The rate I’m on is about 30% higher for on peak (relative to the fixed price per kWh rate), and about 60% cheaper during the off-peak time.
Unless you work from home, you’ll probably be using most of your electricity during the off-peak time. Most of your heating load will generally be at night too, so that auxillary electric heat will mostly run with off-peak power too. You can shift other loads (laundry, some cooking) to off peak times as well. I save on average about $30/month with this type of rate, and I have gas heat so you’d save more with an all-electric system.
Some utilities also offer a reduced rate for electric space heating and your auxillary heat would probably qualify. Interruptible rates might not be a good idea for space heating use so I’d stay away from those.
If you combine the right rate with better insulation you can save a lot on utility costs.
Bill
I'm still confused. Here in Maine, our bill has two parts. One is transmission and distribution. Right now, it's about $.08. Then we have the charge for the power we use. It's about the same, approximately .08. Total is about $.16 per kwh. Is your total cost $.1104? If so, that's pretty good.
Yes, ours is broken into generation and transmission. We can shop the open market for our generation, but PPL charges for transmission. I don't have the bills in front of me, but i pay under .07 foe my generation, and roughly .04 for the transmission from PPL.
With geothermal sometimes the upgrades are upfront "free"/cost neutral because it's cheap to upgrade to reduce you heatloss by 12000btu than it is to add a 12000btu of additional capacity.
Overtime though it's much better to have the upgrades, as they continue to pay dividends. Off course there is the point of diminishing returns.
For the $30,000 geo budget cost, splitting it into $15k of air source heat pump, and $15k of upgrades would likely be cheaper in the long run, and the upgraded house will be more comfortable since the temperature gradients inside will be lower. And you will probably inherently have some zoning which is nice.
My thoughts on infloor in the basement slab. Don't do it, spend half of what you would've spent on the infloor system on additional wall/perimeter/underslab insulation. Your surface temps will be high enough that you won't miss not having infloor heat. Save the infloor for bathrooms where it is a nice luxury, and even then just do it for surface warming electrically. I would put a floor mount/low static ducted style air source heat pump in the basement to pull the cool air off the bottom with a nice low return.
I would put a Sanden Heatpump water heater in before I did a desuperheater unless your getting the desuperheater for less than $1000.
Interesting thought on the hydronic heat. I figured since I was insulating the slab anyways, the Creatherm would be perfect. Plus the initial investment in Pex is small and all DIY with the right layout program.
It usually doesn't take anything like 4 tons of geothermal to heat/cool even a code-min 3000 square foot house above 3000' of walk out basement, but it might take three. Did you or anybody else run an aggressive (per the manual's instructions) Manual-J?
Assuming it's about 6000 square feet of fully conditioned space 800 square feet would be a 13% window/floor area ratio, which isn't excessive. The ~800 square feet of "low E" windows could be anywhere from 9500-15,000BTU/hr of load though, depending on the actual U-factosr. Paradigm makes some relatively inexpensive U0.18 triple panes, but also code-min Uo.32-ish windows. Often times with windows "less is more", unless judiciously chosen to minimize heating & cooling energy use, optimizing the size and type by the cardinal orientation and shading factors.
But just for yuks let's say the actual design load is 48,000 BTU/hr @ +5F. That much heat can be delivered by a cold-climate 5 zone multi-split like the Mitsubishi MXZ-5C42NAHZ at a typical installed price of ~$15-17K, and if the heads/cassette are all reasonably sized for the zone loads it'll hit it's HSPF 11 (= 11,000 BTU/kwh) efficiency numbers, which is a seasonal average COP of 3.2.
With individual mini-splits (rather than multi-zone) the efficiency can be even higher, and the installed cost lower. Three Fujitsu 15RLS3H could deliver 61,500 BTU/hr @ +5F, but could modulate down to 9300 BTU/hr @ +47F, and if sized correctly for their zone loads would deliver over 13,000 BTU/kwh efficiency at an installed cost of about $11-14K in competitive bidding.
Without a Manual-J that you keep updating with envelope changes it's impossible to spec the heating & cooling correctly. Using (DOE freebie download) BeOpt to model the energy use of the house as the changes will make the trade-offs easier to get a handle on than doing it all with a Manual-J type tool though.
At the average US installed cost (before subsidies) of rooftop solar the difference in upfront cost and the $24-28K geothermal would buy about 3500 watts of PV, with annual output that exceeds the difference in efficiency between typical geo (with all pumping power factored in) & mini-splits. Applying the 30% tax credit subsidy to the PV the difference in cost would buy about 5000 watts of PV.
Thank you. No one (including myself) ran any calculations that I’m aware of.
As for mini splits, we have a two zone Mitsubishi mini split right now at our current place, and it does well on our 900 sq.ft first floor during shoulder season heating and is great with AC in the summer. It doesn’t do great once temps get into the 20s and below, but we also live in a poorly insulated house. I’m not keen on the large bodies hanging on the wall (although I know smaller profile systems are available as are ducted units).
Designing an HVAC system without establishing a good handle on the actual loads is always a bad idea- even worse when it's something as expensive as ground source heat pumps. Contractors will always oversize when guessing the loads, to avoid the 5AM call from the irate shivvering customer. By right sizing it you save more than the fees charged to have an engineer run the numbers on it.
Not all Mitsubishis are designed for cold climates. The 2C20NA has only about half the capacity at +5F as the 2C20NAHZ (~11KBTU/hr vs, 22KBTU/hr), yet both are 1.5 ton 2-zone Mitsubishis.
https://nonul.mylinkdrive.com/files/MXZ-2C20NA2_Product_Data_Sheet-en.pdf
https://nonul.mylinkdrive.com/files/MXZ-2C20NAHZ2_ProductDataSheet.pdf
A 100% increase in cold-temperature capacity is a pretty big difference!
The vapor injection compressors used in the cold climate mini-splits also give them substantially higher efficiency at low temp too, not merely more capacity.
Mitsubishi now has mon0directional ceiling units (flush with the ceiling plane) designed fit between 16" o.c. joists, as well as floor mount models that can be set into walls, and mini-duct cassettes, etc. The objectional aesthetics of high wall blob coil units aren't a requirement for going ductless.
Happy with my 4 minisplits in a pretty good house (R70 attic, R40 walls, R30 foundation, R16 subslab, with triple pane windows). 2x 6Kbtu mitsu ductless, 1x 9Kbtu mitsu ductless and 1x 9Kbtu fujitsu ducted. Many cold days this winter in the -25 to -30C (-15 to -25?F) and the house has maintained set points.
I have 2 ditra infloor heating zones (kitchen is 370w and ensuite is 450w) they run 2x per day when highest use would be common. The comfort was worth the $500 premium each to install (pure luxury, not a "green" building type of thing).
I feel uniquely qualified to chime in here!
I took possession of my new home in October of last year. A friend of mine was December.
We built similar homes, with a focus on energy efficiency. Both houses are a little over 2k Sqft, in southern Ontario. Identical climates.
I improved the envelop. 2" of foam insulation on the exterior of the house, insulation under the basement slab, and R34 basement walls all contribute to a low design thermal of 18k BTU.
I then installed a 2.5 Ton Mitsubishi hyperheat heat pump, seasonal average COP of around 2.75.
My friend went the other direction, and bought a 5 ton Geothermal heat pump, horizontal loop, and an otherwise code standard house. Design heat load of 48k BTU.
We recently had a very cold day, -27C. My air source heat pump was just a humming outside, all day, but kept the house quite toasty.
His geothermal system was also working fairly hard all day.
In the end, his Geothermal system used 80 kWh to keep his house warm. My less efficient, less costly air-source heat pump he smeared at? It only took 65 kWh.
My total cost was also less.
Geothermal is dead.
Hi Steve!
Paul in London (just down the road from you)......I have a 1900 square foot home (2003 build), running a 46k BTU output NG furnace currently. I don't need a new furnace yet, but I'm looking at the possibility of replacing the furnace/air conditioner with a air-source heat pump.
I'm glad to hear your system was able to handle the cold blast we received last week. My house isn't as efficient as yours (ach50=2.9) with triple-pane, but I was curious what kind of numbers your Mitsubishi cost to have installed. Could it run through my existing furnace, or does Mitsubishi make an air handler to replace it? Sorry for the green questions, I'm just doing number crunching to see whether going your route is even worth contemplating......
Many Thanks, Paul
Mitsubishi PVA air handlers are fully modulating when married to a PUZ-HA xxx cold climate compressor, and can be a drop-in replacement for a small gas burner in full sized ducts. The biggest cold climate version is the 3.5 on HA42, with 48,000 BTU/hr of capacity at +5F/-15C, but the HA units come as small as 2.5 tons
http://meus1.mylinkdrive.com/files/PVA-A42AA7___PUZ-HA42NKA_Product_Data_Sheet-en.pdf
Use the existing furnace to measure the design heat load before picking a heat pump, since the large air handler versions have a limited turn-down ratio:
https://www.greenbuildingadvisor.com/blogs/dept/guest-blogs/out-old-new
Interesting analysis. So his per kBTU cost is .28 kwh and yours is .6 kwh. It's obvious that you get the most bang for your buck by spending on insulation, but it seems that geothermal is three times more efficient. It would be good to know the lifetime expectancy and costs per heating system.
This is one data point for each system. I think we need to be careful about making generalized declarations based on one case. Especially when we know that the GSHP is so reliant on system design, and we have no way of knowing how well this system was designed for the specific application.
Funny math- how does 0.28/kwh vs. 0.6/kwh energy cost come out to be "...three times more efficient..." in the first place ?
0.6/0.28= 2.14 x "more efficient", a much smaller factor than "three times".
And how are those output kwh numbers being derived, anyway?
Without modeling the two houses we have no idea what the net output energy used really is, only the kwh input numbers.
You're right. Sorry about the math. As to kwh numbers, I am curious to know where to find electricity meters so I can find out the efficiency of different appliances.
>"We built similar homes, with a focus on energy efficiency. Both houses are a little over 2k BTU, in southern Ontario. Identical climates."
I'm confused about what is meant by the statement, "Both houses are a little over 2k BTU..."???
Did you intend to say 2000 square feet?
>"Geothermal is dead."
Just like that miserable Chucky doll, it refuses to go away forever. :-)
An oversized overpumped geothermal system can easily underperform a right-sized air source ductless system on raw efficiency, even though a right sized well designed system will beat it.
From a raw comfort perspective, lowering the heat load has a far more satisfactory impact than a bigger capacity heat pump (of any efficiency/cost.)
Whoops! You're right Dana. 2000 Sq Ft. I've corrected the original post.
I built a home with very similar choices to yours.
Whenever you talk about geothermal operating costs ask how much of that number is for the flow center that is the pump that moves the water around the loop. They will hem and haw and if there are honest they will say zero because it varies so much between installations. The other thing the geo people do not talk about is the February loop temps that is they are almost always 32° while the glossy brochure is talking about the 55° constant temp of the earth. Steve Grinwis story is very interesting.
If you have added windows thinking the solar gain will lower your power bills, I think you will be very disappointed. I say limit the windows to the ones you will be looking thru.
I spent a lot of time with BEopt modeling of my home, I think it was worth the effort. This program will compare the operating costs of different heating systems and insulation packages. You can enter the cost of each system and its fuel.
My guess is the numbers will suggest you add the foam to the walls and remove the sub slab foam.
Generally wood stoves do not work very well in tightly sealed high performance homes. Be sure to ask your insurance agent about a wood stove. Many will not cover any home that uses solid fuels.
I have been very happy with the performance of my Rheem 17 SEER heat pump. At present I have the strip heat locked out above 2° without any problems, I think I can change that to 0°. We had one night at -6° with only 1 hour of low strip heat usage.
https://beopt.nrel.gov/
Walta
My contractor got back to me and quoted $3600 to install R5 foam on the entire exterior. I think that's fair.
Matthew,
Exterior R-5 rigid foam will work in Zone 5 -- but only if you are framing with 2x4s.
If you are framing with 2x6s, the minimum R-value for exterior rigid foam is R-7.5.
More information here: "Calculating the Minimum Thickness of Rigid Foam Sheathing."
Good points. Never thought of that implication.
Also, R-5 foam sounds likely to be XPS foam. XPS has an outsized climate imact from the gases used to blow the foam, 1400X worse than CO2. So look at other types of foam: EPS, neopor, and polyiso.
How much validity is there in the argument that geo systems require less to replace/upgrade down the line since your loop system (regardless whether it’s an open or closed loop) is already in place, and you’re just replacing the indoor unit vs an ASHP? I always find that statement hard to swallow since in a traditional ASHP, you don’t have a loop system regardless, so you’re essentially still replacing the same amount of equipment (compressor and air handler), but it’s an argument my contractor and most of the HVAC guys keep throwing at me.
Additionally, the overall complexity of the geo system worries me as well, since your ROI calculations don’t take into effect the repair or replacement costs during the life of the system. I would imagine both would be cheaper on a traditional ASHP than a Geo system, but then again, I could be wrong.
They would require less than the original install cost, not necessarily less than an ASHP. That's assuming the loop never leaks. My in - laws had to abandon their geothermal system because of that (didn't help that it was no more efficient than pure electric, as far as I could tell based on a couple of electric bills I saw).
I think the argument is really just that the extra expense of a GSHP is a one-time thing. Future replacements are cheaper than the initial install, but not cheaper than replacing some other equipment.
I'm the rare happy geothermal owner. But I'm an engineer and got heavily involved in the design and installation, and I've hear plenty of stories of systems that haven't been designed so well. My electric consumption is lower, in my retrofitted 1970 house, than the electric consumption of mini-split heated houses with much better envelopes here in NH. So it's possible, but I wouldn't recommend it without having a lot of confidence in your system designer and installer. If they didn't even do a proper load calculation, I'm not sure I'd trust them.
I will say thought that there was a period when minisplits had all the best modern variable-speed compressor technology and GSHPs all ran on single-phase induction motors. Mine is from that period. Now you can get GSHPs with the same level of technology as in minisplits, plus the advantage of a warmer source to draw from in the winter. You can also get variable speed ECM pumps for the loop circulation which cut the energy use for that piece that people worry about. My installation performs great and it would perform even better if I upgraded to all the latest technology. But I'm not saying you would have the same success or saying you should do that.
I would agree about currently being dead financially, but earth coupled heating is not a bad tech. There are a lot of poor installs out there that cost a lot of money.
My father built a house and put a GSHP in back in 1995-96 (back then the loop/heatpump/ductwork was about $14k I think). His alternatives at that point were basically propane or electricity. Since then the only failure was one of the Grundfos pumps had to be replaced for about $400. His loop/HP have been flawless for almost 25years.
The neighbors built a similar house but went with a propane furnace. They have continued to compare notes/utility bills over the years. My dad figures he could've replaced his heatpump twice and still be ahead (he is somewhere between $20-30k ahead financially of his neighbor in accumulated costs).
Back then there weren't good ASHPs available, but a GSHP was the right answer at that time.
Personally, I have a Mits FH heatpump/NG boiler, but I feel the geo backlash is a little harsh, especially in the rural world dominated oil, propane, and wood boilers. There are worse things to spend $20-30K on than a GSHP (look at all the fancy pickups on the road), It kind of reminds me of the discussions/echo chamber arguements on car forums between EV owners and Prius owners (or tongue in cheek Diesel owners :-) ).
Horizontal ground loops done right on a rural property with lots of room, which good equipment sizing (which usually means matching significantly less than your peak load, and relying on cheap to install electric for the few hours per year of actual design+ loads) can be a reasonable solution.
That being said, I've come to the realization that it's always better to spend on the building, than the heating system. Good buildings let you use heating systems that otherwise wouldn't be viable, and let you potentially eliminate some costly heating solutions (like in slab radiant) without giving up the comfort. I'd rather have a high performance building with a simple properly sized conventional furnace, than a code min building with a fancy GSHP or ASHP, or hydronic system. A fancy heating system doesn't make a high performance building on it's own.
Josh,
I think you make a good point on horizontal loop systems and I completely agree on putting the building envelope first in terms of investment. I am math challenged but do understand therms, kWh, etc.
On your fathers house, could you give a few more details. What is the kWh usage for heating with the GSHP. What heating zone is it and heating degree days? Do you have any building specifications? As I like to say, every project a case study.
What would the comparable kWh winter usage be if electric baseboard was used instead of the GSHP?
We must once again consider electricity to heat homes with all of the renewable power coming online. In the Midwest we have plenty of wind and I believe the utilities in greening their portfolio will promote electric heat coupled with off peak incentives.
Homes off of the natural gas grid have very high heating costs so building envelope improvements plus the compounding of electricity for heating need to be considered. I think the ROI would be favorable.
I'll have to get back to you about the energy consumption details.
Climate Zone 5 (Mid Southern between Ontario, 20minutes North of kitchener Waterloo)
But I believe his heat pump is spec'd to deliver around 40kbtu from 30-32degf entering water temperature. He has backup electric duct heat, but the breaker has always been turned off unless he goes away in the winter.
The house is 2x6 construction with a brick exterior. 2400sqft 2 story with a walk out basement. and is well insulated by 1990's standards. But it is pretty exposed in a field to winds. Electric hot water heater, with a desuperheater.
He got a wood stove installed 3 years ago, and hasn't used the heat pump much since then. All the kids are grown up and out of the house so the wood stove keeps the main floor and the upstairs cozy for free (his wood is free). He burns 3 cords (3 x 128cuft) per year to keep the house at least as warm as the heat pump, and seems to enjoy the physical work of cutting/splitting/stacking by hand.
Where he lives there are "farm geo" style contractors that focus on straightforward horizontal loops (usually a parallel 500-600ft circuit of 3/4 hdpe piping per ton of capacity, about 6ft down). In the 2000's (last time I looked) it wasn't uncommon to get a GSHP installed for around $18-22k, before incentives. I just looked up one them who is now saying $20-30k assuming land is available for straightforward horizontal loop.
Deleted
So would you all still recommend insulating the slab regardless of the system chosen? Further, should I just use rigid foam and no setup for possibly future hydronic?
IN a zone 5 climate 1.5-2"of EPS (R6-R8)is "worth it" under the slab, 3-4" if being used as a heating radiator. See Table 2, p10:
https://buildingscience.com/sites/default/files/migrate/pdf/BA-1005_High%20R-Value_Walls_Case_Study.pdf
The value is a combination of mitigating mold & "musty basement" smell risk, and lower heating energy use.