# Geothermal – Sizing Issues – 4 Ton or 5 Ton?

| Posted in Mechanicals on

I live in Cincinnati, OH and I am zeroing in on purchasing a geothermal heat pump for my 5800 SF (conditioned) home. My home is all electric is currently heated by an 8 year old 4 ton air source heat pump with 15Kw electric backup. My issue is the two geothermal companies I have contacted to quote the job disagree on heating load of my home. Both companies did Manual J calculations, but are coming up with different numbers. One company has the heat load at 55k BTU — says a 4 ton is the way to go. The other company has heat load at 68k BTU – says go with a 5 ton unit. All calculations were done based on 6 deg. outside air temperature and inside set point of 68 deg.

I am trying to figure out which size is best for my home. The current air source unit performs well. I have closely monitored it’s performance over the last couple of years and it appears the balance point for the unit, the point where it starts needing help from the electric back up, is 28 – 30 deg. At this outside temperature, at night, when there is no solar gain; the unit will run consistently and never turn off, but it does not engage the electric backup. Based on the TDS of the current unit, at 30 deg. outside air the unit is outputting 32,500 BTUs. I feel very confident in the analysis of my current system’s performance. Is it possible to use the above information about the performance of the current unit to help decide which company’s sizing is correct?

Based on my own calculations from above, at 30 deg. the heat load is 855 BTU/deg. 32,500/38 (the difference between the set point of 68 deg. and the outside air). So based on this, at the 6 deg. outside air temperature that the geothermal companies used, the heat load for my home is 53000 BTU. This puts the sizing squarely at 4 tons.

By biggest question is about my sizing methodology — does it work using by current system performance to establish BTU load per deg.

Thanks for your help GBA friends,
Joel Perrine

## Join the leading community of building science experts

### Replies

1. GBA Editor
| | #1

Joel,
The heat loss formula for determining transmission losses through floors, roofs, and walls is Q = A • U • ΔT. In other words, the rate of heat flow through a building assembly (in Btu/h) is equal to the area of the assembly (in ft²) times the U-factor (in Btu/ft² • hr • F°) of the assembly times the ΔT (in F°).

When the outdoor temperature is 30°F, your ΔT is 38F°.

When the outdoor temperature is 6°F, your ΔT is 62F°.

When your ΔT increases 63%, as it does in this example (62/38=1.63), then your heat loss (ignoring infiltration) will increase 63% -- from 32,500 to 53,000 BTU/h. So your math is correct.

There is a wrinkle, however: the stack effect is stronger when the ΔT increases, so your infiltration rate will increase at cold outdoor temperatures. Whether this is a small detail or worth calculating depends on how leaky your house is.

2. Expert Member
| | #2

The efficiency of an air source heat pump varies non-linearly with outdoor temperature and it also varies with system/duct design, which makes it difficult to assess the heat load directly as a measurement of power use per heating degree-day, which would would result in a linear approximation model.

At the cost of geothermal it's worth hiring an energy nerd to run an aggressive ACCA Manual-J load calculation on the place, since HVAC contractors have little incentive to down-size it. The 68,000 BTU/hr @ +6F already has a thumb on the scale, since Cincinatti's 99th percentile temperature bin is +12F ( http://www.energystar.gov/ia/partners/bldrs_lenders_raters/downloads/Outdoor_Design_Conditions_508.pdf ) for a seems on the high side for a 5800' house unless it's older 2x4 construction or leaks quite a bit of air, or perhaps it's a sprawling 1-story ranch(?). No telling what other "garbage" went into the "garbage-in, garbage-out" load calculation.

The load/conditioned space area for the 68K calc is about 12BTU/hr per square foot. That would be roughly the right ratio for for a 1 or 1.5 story 2x6 construction house half that size. Heat loss does not grow linearly with the size of the conditioned space- it varies with the size of the exterior wall & roof (& window & door) assemblies and their construction. Bigger houses are usually more efficient shapes (such as 2 or more floors), and have an inherently lower exterior-surface/conditioned-floor area ratio, resulting in a blower BTU/hr-ft^2 ratio.

A ratio of 53K/5.8K' @ +6F is about 9 BTU/hr-ft^2. That is a credible number for a 5800' 2x6 2-2.5 story with code-min windows and reasonable air tightness.

Using +6F instead of +12F adds about 10% to the total load. Given the expense of GSHP many designers design to only 95th percentile temperature bin rather than the 99th, since the marginal cost of auxilliary heating to cover the rest is lower over the lifecycle of the system than an up-sized geo system. If they're already starting out designing to the 99.7th percentile temperature bin (the +6F outside design temp) they aren't being very aggressive in their system sizing at all- if anything they're padding it. An engineer or energy auditing company that is only selling the accuracy of their math is worth hiring as the sanity check, and to have something to push back with. It won't cost anywhere near what an "extra" ton of geothermal does, even in low-cost geothermal markets.

What IS the marginal up-front cost difference between the 4-ton quote and the 5 ton quote? How much rooftop PV solar would that buy? (If you don't know local rates for PV, use \$3.50/watt, x 0.7,, or about \$2.50/watt, assuming you'd take the 30% tax credit.)

It's quite possible (or even likely) that applying the difference in cost to rooftop PV would more than make up any difference in power used by the auxiliary heating strips that covered the difference during the deep coolth, should the 4 ton system not quite keep up.

A competent energy auditor might also be able to point to cost-effective things to upgrade, and tell you what the "after" upgrades heating/cooling loads would be.

3. | | #3

Martin and Dana,

Thank you very much for taking the time to answer my question. It would seem that the answer to my question is a little less straight forward than I might have imagined. As such, I will provide more detail about my building envelop to hopefully zero in on some consensus.
*** addendum to previous post - the 5 ton company calculated 68,000 BTU/hr at a 70 deg. thermostat setpoint - just found this out, so as a matter of comparison, I am going to call their number 66,000 BTU/hr when adjusted for the lower thermostat setting.

Style of Home = 1 story ranch, 2900' upstairs, 2900' in the walk out basement.

Age = 34 years
Above grade walls = Cedar Siding over 1/2" polyiso, over 1/2 plywood, over 2x4 walls with batt insulation
Below grade walls = concrete -- will be adding 2" XPS to walls & rim-joist, plus requisite air sealing
Windows = Anderson casement windows, no air leaks noted, I believe they are tight
Attic = blown in fiberglass at R30 *** no ice dams ever seen ***, no major air leaks

Overall, I believe the home is not very leaky. There are some can lights that need addressing and the rim joist/sill need sealed. There are also a few doors that come use some attention as well. But we do not have any comfort issues in the home.

The company that has the heat load pegged at 66,000 Btuh listed this in their Manual J report concerning the infiltration rates of the home. This just their opinion, no blower door was conducted.

Winter Infiltration Specified: 0.430 AC/hr (318 CFM), Construction: Semi-Loose
Summer Infiltration Specified: 0.230 AC/hr (170 CFM), Construction: Semi-Loose
Heat loss through structure = 45,000 Btuh, 21,000 Btuh added for infiltration

Are these number indicative of the home I described above? Would it be possible to have a blower door performed on the home and compare it to the above numbers they entered as a way of validating their final number?

The cost difference between the two sizes is \$5k - when factoring in the tax credit it is obviously less. Costs are coming in around \$5k/ton in this area.

Here are my concerns:
1. A 5 ton unit could be too large on the AC side. Current 4 ton unit has no problem keeping up. Based on run time and comfort - sizing seems perfect.
2. Adding heat pump water heater as well during the mechanical make over -- how much extra heat load will this impart during the winter?
3. Agreed that getting an outside consultant might be my next course of action. Dana said the key word "competent" in this case - I am worried I will not be able to find one. Can anyone recommend one in the Cincinnati area?

Much Thanks

4. GBA Editor
| | #4

Joel,
It's almost always cheaper (especially in the long run, considering potential fuel savings) to perform blower-door-directed air sealing -- thereby lowering your design heating and cooling loads -- that allows you to downsize your HVAC equipment, rather than just buying a bigger heat pump.

You should certainly pay for a blower-door test. Ideally, you would hire a contractor who is able to perform air sealing work while the blower door is operating.

If the person who performed your Manual J calculation is correct -- and your house has a "semi-loose" envelope -- it's time to tighten it up.

5. | | #5

You corrected for the error in inside design temp, but you're still hanging with the 99.7th percentile temperature bin of +6F rather than the 99th (+12F). Call it 60K max, and there is probably another thumb on the scale in that calc.

Use foil-faced polyiso rather than XPS in the basement. It's far greener due to blowing agents issues, is easier to air seal, and is modestly higher performance in this application. Depending on just how much above-grade foundation you have, air sealing & insulating the foundation & band joist will cut about 1-1.5 tons off the heat load.

Air sealing the house will reduce the heat load signficantly, and slightly lower the cooling load, mostly on the latent load end. Insulating will have a much smaller effect. The R30 fiberglass in the attic is sub-code, and fiberglass is somewhat translucent to infra-red radiation. Overtopping it with 3" of cellulose will bring it up to IRC 2006 code (but not IRC 2012), and would block the IR coming off the roof deck for a modest reduction in cooling load.

There's no way you'll need more than 4 tons of GSHP after upgrades, and 3-3.5 tons isn't out of the question. Five tons would only be called for if you had unusually high late-day solar gains, like a whole lot of west facing window, which you clearly DON'T have if a 4 ton ASHP was keeping up.

6. | | #6

Dana,

Thanks again for your help. Do the air infiltration values the company listed in their manual j program seem high? I have no way of contextualizing this relative to other homes. Here is the % of heat loss they are saying for the envelope:
30% -walls
32% - infiltration
8% - fore
23% - glass

Do those number seem right compared to the relative norm?

I will be doing extensive air sealing and insulating in the basement. It is good to know that should reduce the load significantly.

Overall I am leaning towards the 4 ton system . Current 8 year ASHP performs well to around 20 deg. - with minimal involvement from the back up elements. A 4 ton GSHP has the capacity to function more like a 5+ ton unit because of the medium it is exchanging heat with - correct?? If this is correct I would agree a 4 ton is plenty big enough.

I was hoping to gain some "thumb of scale" as you called it that might weed out one of the sizes. It appears as you have pointed out that the 4 ton unit would be more than adequate for more than 99% of the heating days in Cincinnati and extra air sealing would really help on the coldest days to minimize use of the backup elements.

Thanks

7. | | #7

I agree with Dana that given the information we have 4 ton or smaller makes sense, for all the reasons he has stated. I think the infiltration numbers being used are reasonable, but conservative, especially given that you will be sealing further in the near future. Doing a blower door directed air sealing exercise would reduce the leakage and give you the data to do a better calculation, if you don't mind waiting for that.

Note that in sizing a GSHP, there are more choices than just one number representing the size: there's some educated guesswork involved in sizing the loop for a given size heat pump, because the soil type, etc. affects the performance. Adding a safety margin and oversizing the loop at bit is a better thing to do than oversizing the whole system. But your two quotes may already be doing an appropriate amount of that.

The other issue is that some heat pumps now have various degrees of variable power operation possible, instead of just cycling on and off. That capability can signiificantly improve their efficiency, and can reduce or even reverse the efficiency penalty for oversizing.

Considering all of that, if you get a high-end 4-ton system, you'll be oversized, but with no efficiency or comfort penalty for the oversizing.

8. | | #8

The infiltration loss as a percentage of the total indicated an extremely leaky house, something that can probably be cut by 2/3 or more with blower-door & IR directed air sealing.

The wall loss as a percentage of the total is in the typical range. The glazing loss is on the low side of typical, but as a fraction of the total that will vary by quite a bit depending on the house shape, and the window/floor-area ratios, window type etc.

Note: When you fix the air infiltration issues it doesn't affect the wall & window losses, so those will increase as a percentage of the (now lower) total. If the infiltration losses get cut to ~10% of the total, the wall losses will probably be bit to the high side pf average as a fraction of the total, and the windows closer to the middle range, indicating that your window/floor area ratio is probably a bit lower than the ~15% typical for new housing. But the ranges are quite wide- I can't say I've done a statisitcal study on the break-down. This is just by seat of the pants WAG. The infiltration loss is the glaring outlier. It might in fact be real, but is probably fixable if that's the case.

Infiltration / ventilation being 32% of the load wouldn't be completely out of line for a super-insulated house, but that's not your house.

9. | | #9

Dana,

Thanks for helping me contextualize the infiltration numbers. Our old home was an 80 year old Cape Cod - no wall insulation, block foundation and very little air sealing - I know what an extremely leaky house is - my current house is nothing like that. This confirms my assumptions that the one contractor was hedging too much with the 5 ton system and is sizing it to deliver total heat from the heat pump down in to the negative digits (i.e. 5 times a year). I was concerned he was not adequately taking into account the reduction in load that would be experienced once air sealing was done to the level we GBA readers understand it should/will be. I do not hold it against the contractor, I do not think the majority know how to factor things like this in. Most people would rather overpower a buildings deficiencies with mechanicals than fix the source.

10. | | #10

Charlie,

Thank you for posting some of the sizing intricacies in the various components of a geothermal system. Yes, the loop is absolutely critical and I will inquire a little more from the contractors as to the sizing methods used there.

As an aside, does anyone believe the over all design of a geothermal system is too complicated to get right - which leads to poor results should the system be designed incorrectly. I am using Water Furnace GEOPRO rated contractors only - they are the two biggest names in the business in the Cincinnati area - so I don't believe this will be an issue. I was just curious if anyone thought a GSHP was that much more difficult to get right than a ASHP system.

11. GBA Editor
| | #11

Joel,
Q. "Does anyone believe the overall design of a geothermal system is too complicated to get right - which leads to poor results should the system be designed incorrectly?"

A. Yes.

Q. "I was just curious if anyone thought a GSHP was that much more difficult to get right than a ASHP system."

A. They do.

An excerpt from my article called Residential Commissioning:

"The residential systems with the greatest opportunities for commissioning problems are ground-source heat pumps (GSHPs). These systems often require five or six different contractors to have a finger in the pie: the engineer who designed the system, the well driller or backhoe operator, the GSHP contractor, the plumber, the duct installer, and the manufacturer’s rep. Considering the opportunities for screw-ups, it’s remarkable that any of these systems work at all.

"Vernon McKown is a builder who owns Ideal Homes in Norman, Oklahoma. In 2005, McKown built a zero-energy home that included a ground-source heat pump. One of the consultants involved in the project was Mark Sevier, who at the time was working for the Building Science Corporation. When I interviewed Sevier in 2005 about the project, he told me, “A ground-source heat pump installation is not a straightforward deal unless the installer has a lot of experience with different types of installations. We had to commission the ground loops, to make sure there was no debris in the loops. We had debris in our case. I would recommend to anyone installing such a system that they get someone who knows how to commission the system. The overall energy transfer of the system is heavily dependent on what the ground loop does. You have to select a water temperature from the manufacturer’s chart. But how am I supposed to know what the water temperature will be in this hole in the wintertime? As it turned out, the loops hadn’t been properly flushed and pressurized. This was a split system. The supplier didn’t have an air handler to match their outdoor unit. A significant problem that we experienced was matching the indoor unit with the outdoor unit. The coil size is substantially different for ground-source systems than air-source systems.”

"I also talked to Vernon McKown about the same project. He told me, “When Mark Sevier was here, he tested the temperature of the fluid in the lines coming in and going out, and we were way off of what the manufacturer’s specs targeted. So we asked why. They started digging into that. Checking these temperatures needs to be part of the commissioning process. I heard the contractor say, ‘We are changing the coil and maybe the fan.’ And I’m like, ‘What the hell is going on?’ Hell, they were replacing the whole son of a bitch. Here is the sad part about it: We build these test houses, and we’re looking at emerging technology. When we were building the models on paper, it looked like the big winner was going to be the ground-source heat pump. The models looked great. But now the performance piece is not working, and that is disappointing.”

12. GBA Editor
| | #12

"One thing I really like about the minisplits is how they are packaged systems from a single supplier, and are highly engineered as a system and therefore very reliable. GSHP [ground-source heat pump] systems are, at least where I have practiced, essentially custom engineered and installed, usually by several entities who have a shared responsibility to make sure the systems perform.

"Knowing that this will cause howls of au contraire to arise, I will say that GSHPs have been the most problematic HVAC technology I’ve worked with, and so I choose the Japanese air-source equipment without a second thought at this point (and I have worked with GSHPs on projects ranging from zero-net-energy homes to buildings up to 70,000 square feet)."

13. | | #13

Everything Martin says about it being difficult to get GSHPs right is true. On the other hand, the same could be said about a lot of other aspects of buildings. I had more trouble getting a dense-pack cellulose job done right than I had getting my GSHP done right. And it sounds like you have experienced contractors, which helps a lot. It would be well worth your time to call some of their references and check them out on Angie's List and anywhere else you can find stories about them.

14. Expert Member
| | #14

You really ARE dependent upon the design skills of the GSHP installers for getting the most out of it, and "typical" performance is nowhere near "best practices" on system efficiency. A best-practices GSHP in your climate could hit a seasonal average heating COP of 4.5-5, but an over-pumped/inefficiently pumped system with insufficient ground heat exchange can be barely half that. The hope is always for the former rather than the latter, but typical is somewhere in-between.

The fact that they seem unable to nail down a heat load calculation with any precision or seem to be padding the size "just in case" is not in their favor. Many competent GSHP contractors looking at house leaky enough to have a 66K heat load would be advising that peeling off a ton or more of load with air-sealing and spot insulation upgrades would be in order before going with an up-sized system. Most of those folks probably only mention the most obvious low-hanging fruit on the building envelope upgrades- they're mechanical guys, not building efficiency nerds, but they would at least point out the cheap & obvious. I'm aware of one GSHP contractor in Florida who goes as far as to insist any house with the ducts & air handler in the attic spray-foam the roof deck and air seal the attic, verifying the air tightness with a theatrical smoke machine before letting the foam guys leave, since that's cheaper than the extra tonnage of geo. It ends up being a smaller system for him- he makes no margin on the foam, but call-backs and complaints evaporate, since comfort levels and latent load control are higher than with the oversized system in the vented attic scenario. Unfortunately he seems to be a rare species in the HVAC biz.

Competent independent auditors would be able to give you both before & after upgrades load figures using Manual-J methods. Even when the extra tonnage costs are lower than the building upgrades it's important to keep in mind that the building upgrades have a lifecycle 2-4x as long as the mechanical systems, and the "extra" ton of GSHP has an associated annual power use, whereas a ton of reduced load does not.

It's unlikely that you'd be able to get this antique down to mini-split capacity territory without a MAJOR deep energy retrofit, and at \$5K per marginal ton for GSHP, after subsidy it's probably not a lot more expensive than a best-in-class modulating 4-5 ton ducted ASHP with a seasonal COP of around 3.0. But it's harder to guess where the as-used efficiency is going to be when you go GSHP unless the designer's systems have been carefully performance monitored by a disinterested third party.

15. | | #15

Martin,

I absolutely agree that if I was building new or had a smaller home I was remodeling a mini-split is the way to go. We currently have a ducted system for a large home, so I have ruled a ductless system out. I appreciate the articles you provided. They highlight some of my concerns in that the savings I hope to achieve will not be realized due to poor installation and design. I am truly wrestling with this decision. Very much concerned that the up front cost, even with the tax credit, and the potential down side if the system is not designed and installed correctly, could be more headache than it is worth.

• |
• |
• |
• |