Geothermal Quote question
I am building a new home this Spring and I plan on having ground sourced geothermal. I received a quote from one contractor in my area. His plan is to do a 6 ton horizontal slinky loop field with a 4 ton heatpump. He claims that an oversized slinky loop field performs the best, however I’ve heard bad things about the slinky loops, and slinky is the only kind of loop they do.
It doesn’t seem efficient or cost effective to overbuild the field since that is a major cost. Does anyone have experience with this?
I’m also getting another quote for a straight line horizontal loop and vertical closed loop. I’m worried about the vertical since there is a lot of shale in my area and I’m not sure how deep the wells could go.
Thanks
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Jordan. You msy find that some will question your choice of a GSHP as opposed to a different system. It would be useful to know where you are and what the heat load is. Without an accurate heat load calculation, it is pointless to try to design a heating system.
I'm located on the edge of 5 and 6 IECC climate zone in New York. The Heat Load calculation came to 4 Tons. I would just like to hear from someone with GSHP experience what they thought of overbuilding the loop field.
The reason I'm going with GSHP is because I prefer an all electric build that will eventually be net zero.
Jordan, I selected GSHP for my new house, finished coming on five years ago. I like it, the system gives me summer AC as well as heat the rest of the year, and thus far maintenance consists of replacing the air filter now and then. I still feel that it was the right choice for my house, but that is because the house is superinsulated, and the unit is slightly oversized at two tons, so that the two-stage unit never has to upstage past first. Also, the house needed a well drilled for domestic water supply, and here the ground water is adequate for use in a Standing Column Well (SCW) design. The well didn't have to be any deeper than was needed for net water draw, so that there was no extra drilling cost for the heat pump. The well actually would support a three ton heat pump.
If your house needs a four-ton unit, and it can't use a well drilled for domestic water use, then GSHP would be a harder sell, and you might well be better off with the lower installed cost of minisplit ASHP. Whether you install GSHP or some other type of heating system, you really ought to put more planning effort into the design of the exterior shell. Four tons for a house suggests a house just built to code. If my house of close to 4000 sq.ft. is heated well with a two-ton unit running in just first stage, you perhaps can cut your heat loss in half, and wind up with a more comfortable house that needs a much smaller heating system. Put your first money into the shell design, rather than try to find a cheap source of heat to dump into it.
I am also on the edge of 5 and 6 climate zones in PA, and am a geo owner. We had the land and the machinery to do a slinky, so I read up on them awhile back. Since then, courtesy of a dud water well, we ended up with a 500' deep dual loop geo system.
Anecdotally I understand that some slinky systems under-perform as the winter goes on. The ground around your slinky (or my well) gets colder and colder and the geo unit works harder and harder come the end of February or so. I have even heard tales of systems that quit working altogether after a particularly cold winter. Surprise, the ground is not always the nice 50 degrees that the marketing literature would imply. Look for discussion on the anticipated "Seasonal Performance Factor" for your system as a key to understanding this. Upsizing from a "4 ton slinky" to a "6 ton slinky" may be a way your contractor hopes to avoid this problem.
Another thing to understand is whether your system will have aux heat. Sometimes, where geo systems are a bit short changed on well or slinky, aux electric heat is used to make up the difference.
You say you have shale, so we are very similar there. Most of the shale rock here is relatively soft so I wonder if that would really be a hindrance for a normal water well driller with a bigger rig. On the other hand, if it's close to the surface, it might make for more difficult digging and backfilling for your slinky.
We are investing into our envelope. We will have r29 walls (2x6 wall w/ fiber batts & and R10 rigid foam on the exterior taped for a tight envelope. As well as foam around and under the foundation, with 18 inches of blown cellulose in the attic. I'm thinking maybe the heat load calculation I was given might be high. The house will be 2500 sq ft of conditioned space (with a finished basement)
Check out geoexchangeforum. They might provide some additional help and are focused solely on geothermal. We chose a slinky loop system for our new construction home mainly because it was cheaper and the land was already "tore-up". They installed the slinky field and rough graded the land in a day. I'm not sure an oversized field will really add much to the cost. We chose the system a couple year's ago when propane and oil prices were so high.
A better-than-code 2500' house would not have anywhere near 4 tons of heating load. I'll bet it's closer to 2 tons. My 1920s 2x4 antique with 2400' of conditioned space above grade plus 1500' of insulated conditioned basement doesn't even have a 4 ton load until it's -15F outside. Your load is clearly going to come in quite a bit lower than mine even if you have 2500' of basement (is this a 1-story?). The calculated heat load is complete junk- even code-min houses that size won't have a 4 ton load at -5F or whatever your outside design temps.
The "...we will have..." indicates that the house hasn't been built or even fully specified yet?
If you were thinking R19 batts + R10 XPS = R29, the R19 batts only perform at R18 when compressed to 5.5" in a 2x6 cavity, and the thermal bridging of the 20-25% framing fraction of R1.2/inch wood robs considerable performance from the average, or "whole-wall-R" with all thermal bridging factored in. You're looking at only about R24.5 whole-wall with R19 batts and 2" XPS. A 2x6 wall with R23 rock wool batts and R10 continuous sheathing has a "whole-wall" R of about R26, still not R29. With a full on advanced-framing approach with a sub-20% framing fraction it might make a hair above R28, but that's about it.
To get to Net Zero with an array that fits on the house in a zone 5/6 location is going to take a real R30-R35 whole wall, which means 2.5-3" of polyiso on the exterior, or significant higher efficiency PV panels than are the current commodity product today. Download a copy of BA-1005, and read at least the first chapter, taking note of Table 2, which are "whole assembly" R values, not center-cavity framed construction R.
http://buildingscience.com/documents/bareports/ba-1005-building-america-high-r-value-high-performance-residential-buildings-all-climate-zones/view
XPS is one of the least friendly insulating products out there due to it's HFC134a based blowing agent. As it loses the HFCs over time R10 XPS eventually drops to about R8.4 too. Replacing the wall-sheathing with 2" polyiso gives you higher shoulder season performance, and comparable mid-winter performance with far less damage, since it's blown with much more benign pentane. You can't use it below grade on the exterior however.
A 2.5" + 2.5" EPS insulated concrete form (ICF) would be better than the code-min R15, and would also be more environmentally friendly since it too is blown with pentane. It also means you can run the sub-slab foam right up to the wall foam and have a good thermal break against the wall-concrete thermal bridge to the footings. If you take this route, with the foundation sill parked at the edge of the concrete, the half-inch wood sheathing + 2" wall foam are co-planar, for a seamless thermal break at the top of the foundation. If you're mounting the windows "innie" with the housewrap between the foam & wood sheathing, use EPDM tape as Z-flashing where the wall foam meets the foundation foam.
Use Type-II or Type-IX EPS under the slab. It'll be 20% thicker than an XPS solution, but it'll still be performing at it's day-1 R-value after 50 years of service. It's usually 15-20% cheaper per R too.
Jordan, I'll add to what Dana said about what appears to be an unrealistic four-ton heat load calculation for the size house and shell design you describe. I did my own heat loss calculation for my house, using a spreadsheet to add up the losses for all components of the exterior shell, including basement slab and expected worst air leakage, and correcting for mechanical ventilation through an HRV. Adding up the bits and pieces of heat loss is really all that a canned software app does, except that in my spreadsheet I knew just what all the "under the hood" calculations were and what assumptions I made, so I knew where all the uncertainties were.
Before I specified and ordered the heat pump, I gave all the size information, specs, and window U values to the regional distributor of the heat pump and to two of their "approved installers." All of them came back saying the heat load would be somewhat over 4.5 tons and recommended a five-ton unit. Only one would give me the details of the calculations done to support that size. On review, side by side with my spreadsheet, I found that they had assumed the existence of a fireplace, with air leakage costing 10,000 BTU/hr, or nearly one ton of the huge total difference. My house has no fireplace, only a small woodstove with outside air kit. Other air leakage in that installer's calculation reflected an air leakage rate selected off a drop-down list in the software used. The two air leakage numbers added up to 80% of the difference in totals. The rest was partially due to insulation levels chosen from a drop-down list and considerably below what I designed into the house and described in the information given to the installer.
I could only conclude that too many assumptions and incorrect numbers had been used in the installer's calculations. Perhaps that approach would work well with older houses or even one just built to barely meet code. I am convinced that a heat loss calculation for a high performance house must be highly detailed, taking a considerable amount of time, probably more than a contractor can afford to sink into a quote, considering he may not get the job. He may even not really know how to do one "by hand" (via spreadsheet), so has to rely on software and its assumptions, or at least doesn't know where he needs to override software when drop-down choices are too inaccurate. Then, too, he certainly doesn't want to get callbacks later if a system he installs underperforms, and also he makes more profit from a larger system. To specify a correctly sized GSHP takes confidence in the calculations done, and I suspect that to do this for a high performance house a typical installer is in the deep end without his floaties on.
In the end, I specified the two-ton unit, rather than the five-ton unit they and the two installers recommended, and the distributor agreed to provide it, but told me they would be of no help if it turned out to be undersized. My total heat loss at design minimum outside temperature was 22 KBTU/hr; my best calculation of actual indicates closer to 19 KBTU/hr. A builder of superinsulated houses in Maine told me that in his experience such a house tends to perform a bit better than the calculation says it would. That's understandable.
Dick,
I also have done my own spreadsheet. Simple math really. Do you have anything more accurate than the "Los Alamos Formula" for calculating heat loss through the slab?
For that matter do any GBA readers have a better estimate of slab losses than that?
Dick,
I also have done my own spreadsheet. Simple math really. Do you have anything more accurate than the "Los Alamos Formula" for calculating heat loss through the slab?
For that matter do any GBA readers have a better estimate of slab losses than that?
Ted, I don't know what that formula is. I made various assumptions about soil vertical temperature profile outside the basement wals and beneath the slab. In my mind, heat loss to soil has the most uncertainty in a model. With R20 on the walls and beneath the slab, and with perhaps a third of the total lower level perimeter walls being totally framed with windows, the heat loss through concrete wals or slab was a small fraction of of the total for the house. A fair amount of uncertainty in a small fraction of a small total isn't very large, so I didn't need anything better, no matter what my assumptions were.
An rough estimate for slab loss is fine for a total load calculation. It's less useful for optimizing the choice of sub-slab insulation thickness, but the penalty for going thicker than needed is pretty small.
Dana's envelope advice is excellent.
The economics of GSHPs are not very good right now, because mini-splits are offering the latest technology at bargain prices, whereas you pay a lot to get similar technology in GSHPs. And if you get an old-technology GSHP, it's not that much better than a minisplit, or maybe not any better if you compare a bad GSHP design and installation to a good minisplit installation. Despite the prevailing assessment on GBA, and the problems I've noted, I still see value in them. For example, they have a lower peak load on a very cold night when air-source heat pumps are at their worst.
I'm not a fan of slinky loops. They have higher pumping power requirements per unit volume of soil they draw from. That issue can mitigated some by using high efficiency ECM pumps; it can also be mitigated by plumbing multiple sections in parallel instead of series. But I'd ask to see details on pumping power calculations before accepting a slinky loop design.
Oversizing the loop is a better idea than oversizing the heat pump. Oversizing the heat pump will lower your efficiency, but oversizing the loop will improve the efficiency. If they are used to oversizing the heat pump, and matching the loop to the heat pump rather than the real load, you might actually do well to have a loop sized for say 3 tons even if your load is only 2 tons.
Despite all that, it's probably more economical to the money into the envelope and PV and use minisplits.