I just had a J load run on our new house [1744 heated space] that is now in the frame stage in East Texas. 

With the 2x6 stick frame with foam insulation and R-38 ceiling [average ceiling height is 10'], energy efficient windows and doors, the  GSHP installer says the size needed is a 2.4 ton [vertical loop].

He said that upper end HPs are not in .5 units and recommends a 3 ton.  I have been reading that it might present a problem with humidity.

Will this cause a problem with summer cooling and humidity? 

Should we use a 2 ton unit?  or is there a 2.5 unit out there?

We also want the hp to heat our water.  What do you recommend for heating the water?

I will be looking forward what you all say.

Thanks.

2.5 ton single stage geothermal heat pumps are available. The higher end extra high efficiency dual stage heat pumps generally only come in whole increments such as 3.0 ton, but one this size would generally operate at around 2.0 ton capacity when in the highly efficient first stage.

But a better solution than the 3.0 ton dual stage unit might be to spend a little extra improving the building envelope so that you can get that load down to 2.0 tons. Maybe spend a few hundred extra and bump that R38 ceiling insulation to R49. Or add an extra 1" layer of rigid foam on the outside of the framing. Or go with a slightly better or slightly smaller windows.

To ensure you meet efficiency requirements for tax credits 2 stage equipment is your best bet.
Little down side to a 3 as it can operate most of the time as a two ton.
By and large most loop installers do not reduce for 1/2 ton increments, so you generally get 3 tons of loop for a 2.5 ton unit anyway (meaning your not buying extra loop to go to 3 ton).
If envelope improvements can get you to 2 tons that is not a bad idea, but depending on where you are in East TX you may have a significant heat loss as well.
Ask installer to try different insulation configurations on his loss software to see what it would take to get the load down.
j

I agree with others that you ought to do what you can to the house's shell so that a two-ton two-stage unit will work comfortably. In particular, make that shell as tight as you can get, with fresh air brought in under control through an HRV or ERV. Run the numbers on air leakage, and you'll see that air leakage makes a large difference, relatively, as the total heat/cool load goes down.

You'll want to review carefully the details on the Manual J or equivalent load calculation, especially as to any assumptions made in setting up the calculation. I have found that it's very easy to assume something that isn't so or to overlook something important. You can build your own spreadsheet to add up the U*A contributions to heat gain/loss through the building envelope, including sensible and latent heat effects of air leakage. That can be used to check various parts of the Manual J result. You can afford to spend a lot more time on a detailed calculation than an HVAC contractor can.

In my case, for house we just completed and moved into (almost 4,000 sq.ft. conditioned space, in NH, zone 6, heating dominated), there were several outside calculations done to size the heat pump. All three were close, and all called for a 5-ton unit. My own spreadsheet called for a 2.5-ton unit, and later a 2-ton, which ultimately was installed. I obtained details for only one of the outside calculations done, so I can only speculate that similar assumptions were made in the other two. The house is superinsulated and very tight, something I suspect is outside the experience of most HVAC equipment contractors. On comparing details for the outside calculation vs my own, I noticed first a line item for "fireplace;" there is no fireplace in the house. Next I noted that heat loss due to air leakage was considerably larger than by my calculation, using my target for tightness. Together, the two differences accounted for the great bulk of difference in total heat loss between the outside calculation and mine.

My calculation at that time showed that I was just above what the two-ton unit could provide, with no allowance for COAX fouling over time or "as built" differences in the structure. I made some changes to the house details and ultimately got the load down comfortably below the two-ton unit's capacity. Finally, the blower door test showed that the house was somewhat tighter than hoped for, making the two-ton selection even better, so that's what I went with. The well itself will support three tons of load, the depth having come from what was needed for domestic needs. The duct work also was sized for three tons. Temporary heat through the winter months was supplied by a small woodstove operated part time, with a firing range of 11 to 28K BTU/hr. Very crude numbers from operating that confirmed that the house's thermal performance was ballpark correct vs the calculation.

Well done.

We get similar results down here in Florida with new construction. I'm putting a single 3T in a 4000 SF house now under construction. Most "rule of thumb" outfits would want to go with 2 units totaling 5 or 6 tons

So Dick's experience is that 3 out of 3 geo companies did a manual J incorrectly by 100%. Not a good reflection on the industry.

Dick
I assume you are living in the house now? Is everything working as planned for you? Also do you have any hydronic heating with your unit? That is what I am trying to decide at this point. We are putting up about 3400 sqft in zone 6 as well (northern Iowa), single story with walkout basement. 2x6 OVE w/ 3" exterior XPS, full sheathing, ICF basement, triple pane (.17) north windows, high SHGC (.45+) south...no east or west windows at all. 9' ceilings, no fireplace, r60 ceilings. My loads are showing around 26k btu heat loads. I used .2 air changes. What number did you use? I would like to get that down closer to 24k or 22k even. A 2 ton waterfurnace puts out around 20-22k btu, so I would still need a small electric strip backup. However if I want to run the desuperheater on it and do infloor int he basement, they are telling me I need to jump up to a 3 ton/3 or 4 wells.

If the basement slab is well insulated and you were planning to heat the space anyway (that is, it is already reflected in your Man J), doing it via the floor shouldn't add load.

The DSH shouldn't add load during design conditions unless you have folks up using hot water after midnight during winter. My reasoning there is that design cold weather happens late at night and by then the DSH pre heat tank will have been completely heated. The DSH pump will either have gone off on high limit, or refirgerant will return from the DSH at about the same temperature it entered, i.e. there will be little or no heat transfer to the preheat tank.

Don't fear a strip heater - if your design calc is solid the strip shouldn't run but a few hours on the coldest days of the year. If a 5kw strip ran 100 hours at 10 cents per kwh, that works out to $50 bucks per year cost to run. Getting that heat from a larger heat pump would cost $15 or so, leaving a savings of $35. The larger heat pump and loop would add several thousand bucks to the project and might operate less efficiently in part load conditions, to boot.

Izerarc, we moved into the house mid-June, so we haven't had any chance to heat the house with heat pump. It works well in cooling mode, coming on now and then and doing a great job with ease, even during the recent hot weather. The unit (CM Tranquility 27) has the Climadry feature, but when I tried imposing a humidity setting while the thermostat was otherwise satisfied the unit tripped out with Fault (code 2), high refrigerant pressure. I suspect the Climadry reheat coil wasn't purged of air during installation. As soon as I get a chance, I'll fish out the bleed coil and follow the instructions for doing that. I paid extra for that feature, and I want it to work. There are times when it is quite humid, yet the outside air temperature isn't all that hot, and I want the dehumidification without overcooling the house. The unit is sized for heating and is considerably oversized for cooling, given the nature of the house.

My heat load model assumed 1.0 ACH at 50 Pascals, or about 0.05+ ACH natural in cold weather. It also figured 0.3 ACH total ventilation, the balance coming via an HRV (Lifebreath 195ECM), assuming 70% overall efficiency. That unit has a double core, with a claimed efficiency of 88%. The final blower door test showed 460 cfm, or about 0.8 ACH/50, better than hoped for.

My last heat model showed just over 21K BTU/hr. CM's tables for the 026 at full load, 6 gpm water, 50 F EWT, 820 cfm air claims 24.5K BTU/hr. so I ought to be ok, with some room to spare.

As to electric strips, CM's smallest is 5KW, or 17K BTU/hr. That's a huge fraction of stage 2 capacity, too much in my mind to use for 3rd stage heating. I did get the unit with that strip installed, but I also picked up separately a 1 KW strip. That heat is equivalent to another 10 degrees of outside temperature, which will cover me in the worst situation I've ever seen here. The smaller one will go into the duct just above the 5 KW strip. The 1 KW strip will come on as 3rd stage, with the 5 KW strip added in when emergency heat is called for. Between the thermostat (CM's ATP32U04) and the HP is an EWC UCZ4 zone control board. In addition to handling two stages of heating or cooling, it also handles the two stages of electric heating. On a call for emergency heat, the first strip will come on, followed by the second three minutes later. Together they will heat the house in the event of compressor failure.

As Curt notes, the number of hours supplemental heat is needed typically is low. In our location, design outside temperature is -3 F, and statistically it's lower than that for 53 hours total. If that is covered by that 1KW strip, and juice goes for 15 cents/KWH, I'm out $8/year. The 5 KW strip would cost $40, so the smaller strip pays for itself in the first year, statistically. I don't think a strip more properly aimed at emergency heat ought to come on when one more appropriate for "supplemental" heat is needed.

Rhetorical question: as highly insulated houses of more modest size get built and are heated by GSHP, will CM and WF start making units smaller than 2-ton? I suspect not right away, but who knows?

In a typical house load panel there is really no difference between a 5 kw strip operating 20% of the time vs a 1 kw strip operating continuously, so I'm not sure I'd have gone to the trouble of two separate strips.

I wouldn't look for units smaller than 2 tons - too much of a niche market for the foreseeable future. Houses with very low loads can be fairly cheaply heated with strips or propane. Payback on a geo system lengthens as load falls.

Look instead for inverter-based true variable speed systems able to ramp down to a ton or less.

Posted By jonr on 21 Jul 2011 08:39 AM
So Dick's experience is that 3 out of 3 geo companies did a manual J incorrectly by 100%. Not a good reflection on the industry.

Most contractors are not familiar with super insulated homes and are suspicious of their claims. One who builds a home like this and does not seek out contractors with the appropriate experience are likely to see the same result. It is a very predictable outcome indeed.

With Leed and other homes of the ilk, attention must be paid from the foundation to the shingles. It is not a job for anyone with a builders license, and it is not a job for someone with run of the mill heating or geo experience.


j

Posted By engineer on 25 Jul 2011 12:05 AM
In a typical house load panel there is really no difference between a 5 kw strip operating 20% of the time vs a 1 kw strip operating continuously, so I'm not sure I'd have gone to the trouble of two separate strips.

I don't follow your reasoning. Mine is this. If for each hour I need another 3412 BTU of heat, above what the 25,400 the table for the HP says it gives in 2nd stage, the 1KW strip provides that for the whole hour that the HP operates. If my 3rd stage strip is the 5KW size (17,060 BTU/hr), the 28,812 BTU needed in that hour will be provided by HP and strip operating together for 40.7 minutes (0.678 hour). My electric strip use for the hour would be 1 KWH for the small strip, but 3.39 KWH if the larger strip is used for 3rd stage heat. Use of too large a strip for 3rd stage (supplemental) heat reduces the overall COP of the HP/strip combination. That's why I'd rather reserve the larger strip for true emergency use (compressor failure, etc), and have a "right-sized" 3rd stage strip for supplemental. True, the savings in absolute $$/season is small, but the small strip is cheap, too. I figured that if I'm doing GSHP I may as well tweak the electric strip issue while I'm at it.

You are thinking that the thermostat is dumb and will over use 3rd stage and then turn off the HP completely. It shouldn't.

No, that's not what I meant. On those statistically few hours per season when even 2nd stage can't keep up with heat loss and third stage (electric resistance) comes in to supplement the HP, I don't want a flood of electric heat, just a modest increment, so as to keep HP output dominant and minimize resistance heat (COP=1) contribution.

Perhaps I misunderstand what you mean. If the thermostat engages 3rd stage, will it keep it on until the temperature setting is reached, then shut off everything? Or will it monitor rate of heat recovery and downstage to just HP as setpoint is approached? This is the CM ATP32U04 tstat I have.

I don't know but if it does the former, then you need a better thermostat.

Posted By joe.ami on 25 Jul 2011 10:03 AM

Posted By jonr on 21 Jul 2011 08:39 AM
So Dick's experience is that 3 out of 3 geo companies did a manual J incorrectly by 100%. Not a good reflection on the industry.

Most contractors are not familiar with super insulated homes and are suspicious of their claims. One who builds a home like this and does not seek out contractors with the appropriate experience are likely to see the same result. It is a very predictable outcome indeed.

With Leed and other homes of the ilk, attention must be paid from the foundation to the shingles. It is not a job for anyone with a builders license, and it is not a job for someone with run of the mill heating or geo experience.


j

IMO....if you are in the HVAC business you should have been trained for all situations!! Anybody that doesn't, needs to take some more classes. Everything I've read over the years is that 'most' HVAC 'always' oversize the units!! If the folks that install geo would get more educated, the units would be smaller.....costing alot less. That would make geo more competitive giving the installers more business! It just seems that HVAC installers always go much more(in tons) than needed to CTA.....costing folks alot of wasted money!

Dick R:

Fair point.

I see now that "rightsizing" the strip allows the HP most opportunity to meet the load with minimum use of the strip. I don't know where you go to buy a 1 kW strip, but some minor re-jiggering of a 240 VAC 5 kW strip so as to run on just one 120V leg would yield 1250 Watts / 4266 Btuh. I'm not saying this can easily be done; just throwing out the possibility.

robinnc, As a hvac contractor I will agree that proper training is a must. The problem with load calcs is they often come out in an area that does not match will with any equipment size. A building structure may need more heating than cooling or vice-versa. Then there is the need for proper ventilation. Buildings get altered in the construction process and equipment gets ordered early on to meet construction timelines.There is nothing worse than a job which can not cool or heat a structure during extreme weather conditions. Engineers are notorious for mistakes and the attempt to blame the contractor when things do not work. I make a living correcting jobs which never worked right from day one.

I already bought the 1 KW strip. I got it through the Grainger catalogue. It's not all that big, is a finned surface unit, SS.