Conflicting HVAC Quotes
The original design for my house was 1900 sq ft 1st floor (front faces due North). with a 1900 sq foot finished basement. 450 sq ft attached garage. It is a walkout that faces due West. Only the walkout wall is fully exposed (and will be covered with slate). Remaining basement wall exposure is 12″ at the top (also slate covered). First floor and basement ceilings were to be 9′. Using Marvin Integrity fiberglass windows with LoE2 coating for North & South facing windows. LoE3 for East and West facing windows (only 1 5×5 facing East, and one 5×5 and one 2.5×5 facing West). The planned insulation was 2″ sprayed closed cell foam in basement and around band. 2×4 framing cavities filled with sprayed open cell upstairs, and attic floor was to be sprayed with open cell to R-38. I’m in KY (climate zone 4).
Upfront, I got one HVAC quote from an installer who lives across the street from my build (installs heat pumps). He initially quoted me prices on a 3-ton system, given those design parameters. His quote is about 6 months old.
As construction has progressed, only a few things have changed. Basement ceilings only turned out to be 8’10” (so only about 8′ where the duct chase will be). To save $2000, my insulator suggested using 4″ of spray foam on attic floor, then blown in cellulose on top of that to R-38. Otherwise insulation is unchanged.
Last week I got two quotes from Geotherm installers (Water Furnace & Climate Master), both recommend 3 ton systems (one 36,000 BTU/hr manual J and the other 39,000 BTU/hr). Just to make sure I had the most current prices, I went back to the original installer to refresh the quote. I told him about the change to the attic insulation. He came back with a quote for a 4 ton system (raising the cost). He said his original calculation was 36,000 BTU/hr with an all foam attic, but now that I’m doing foam/cellulose, it goes up to 46,000 BTU/hr.
So I guess my question is, does this sound reasonable? Is there a reason why a heat pump tonnage might differ from a Geotherm tonnage? Ground to inside air differential vs outside air to inside air? Is going from all foam to a foam/cellulose hybrid in the attic really going to increase my heat load by that much? I though open cell at 4″ was an effective air barrier, so why would it matter what is providing the additional R value?
I don’t want to buy an over or under-sized system. I also don’t want some dealer giving me an undersized price to win the job. The heat pump dealer is sort of old school though, using software that is a decade old (even though he does go to regular training updates). I don’t know if that could effect the calc too. Do the calculations change with time? The Water Furnace dealer has 14,000 units installed in his service area, so I wouldn’t think he is in the habit of under-sizing people or he wouldn’t do that much business.
Not sure how to proceed with conflicting info. Any suggestions?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Even the 3-tons heating sounds slightly high for a very-tight 2x4 framed house that size in climate zone 4, but not ridiculous. Changing the o.c. foam portion of the attic insulation to foam + cellulose at the same R-value would have zero effect on either the heating or cooling load numbers. The air-permeabilty of cellulose isn't as low as o.c. foam, but it's dense and air-retardent enough that it doesn't lose performance to convection with colder temps even without a top-side air barrier the way some low-density fiber insulation can.
Even if you counted just the 4" of foam and rated the cellulose overtopping at zero, heating loads getting onto 4 tons heating would be high for a house that tight, and given the modest amount of west facing glass the cooling loads won't be there either.
Attic floor just use 18" of cellulose alone. Geo bigger better.
Provide an Air Barrier System along with the cellulose, You must
Drywall is the air barrier. You meant to say, "do proper air sealing of the entire building enclosure.
Well, I've done some more digging and called all these guys back and at least have some new info. The guy with the lowest heat load was ~33,000 BTU. After talking with him, that is sized based of the sensible load (95% of days) his calculator spits out. The guy with the highest heat load was 48,000 BTU, but after reviewing his numbers, we found a few inaccuracies (he had a sealed attic, mine is vented, and his U factor on the windows was way off). After the changes, he came back with 43,000 BTU, which is still 4 ton. However, he is basing that off maximum heat load out of his calculator. His sensible load was actually listed ~32,000 BTU.
So it sounds like the highest and lowest sizing are really the same, just one guy sizing using max load and the other sensible load. The Water Furnace quote is still up in the air, it is 38,500 BTU heat load, but I don't know if that is max or sensible. But then the sheet says 3 ton (36,000 BTU) with 2kw of aux heat. (6800 BTU), so I'm guessing it is probably sensible, with aux heat to cover max.
I think the biggest cooling load of the three was 20,000 BTU, so any system easily handles that.
If this is all the case, then everyone is really coming up with about the same max load (43-45k BTU), but two folks are sizing me based on sensible load, and one guy sizing me a 4 ton that should flat out handle everything and never need aux heat.
What is the best approach for sizing, sensible or max? All units are two stage so at least a portion can shut down when max isn't needed.
The additional cost of an extra ton of geo to cover the 99.99% condition will literally never be paid off in utility savings- size it at the sensible load and don't over-ventilate in winter. Since most load tools have some built-in margin up-sizing is rarely a good idea. (IIRC in California under Title 24 it's illegal to oversize from Manual-J by more than some small fraction- something like 10% or 15%.) It's not a disaster to cover the 99.5% heat load with the geo- you need to run a kilowatt of aux-heat for 0.5% of the heating season hours to cover the shortfall it's not a disaster, and over the anticipated life-cycle of the equipment the operating cost of the aux heat wouldn't come close to the cost of an extra ton of geo. In cheap-electricity markets sizing the geo to 90% or 95% of Manual-J might even make sense.
If you oversize a 2-stage unit by very much you end up re-inventing the 1-stage unit, since the second stage is never required.
Dana, you may be right but where I live a a decade of air to air heat pump installs with electric back up left all us customers with $1000-1500 monthly electric bills in the dead of winter. Late 80's York systems. Most added propane and wood stoves. Same problem back then with Geo systems, the systems could not take care of cold months well. Would love to have someone in the Adirondacks post here that they have systems running now around here putting out warm air, for warm homes and no back up electric at $1000 for January and February.
AJ- Different climate zone, different heat-pump source, different average utility rates, different problem.
But even at 15 cents/kwh a $1000 monthly bill would be 6,667kwh, which is more than 22.5 MMBTU of resistance-heat at a COP of 1, which would imply considerably more than a 15% undersizing for geo even for the Adirondack region, I implies a disasterously under-designed/under-sized heat pump system. If the actual monthly heat load adds up to 40MMBTU and the system COP was as low as 2.0 (plausible, for old-school ducted air-source heat pumps, or a worse than average geo design) the heat pump would have to be undersized by about 10% or more to hit those levels of power use (and without low-temp air source heat pumps that's a fairly likely scenario to have run into, since the COP of most air source heat pumps fall to 1.0 well before Adirondack region design temps are reached.)
If the average COP of the system was 3.0 (still on the low-side for geo, but an average now within the reach of best-in-class ductless air-source in that climate) it would have to be more than 30% undersized to chew through 6500 kwh in a month for 4MMBTU of heating load.
In Will's home-sweet-Kentucky-home a decently built house that size wouldn't require even 20 MBTU of heat in any one month TOTAL let alone 2 MMBTU of auxiliary heat over and above what the heat pump should be providing. Even if he's burning 1000 kwh/year (3.4MMBTU) in aux heat it would take decades to add up to the cost of another installed-ton of GSHP, and even photovoltaic solar might have a better payback on the cash-difference.
Mind you, there's a fair number of truly abysmal geo implementations out there. More so than the equipment manufacturer or even the heat pump size, it's the system designer is who ultimately determines the capacity & efficiency of GSHP systems, which is why I tend to prefer lower cost "systems in a can" like ductless air source heat pumps whenever that's a reasonable fit- 90% of the critical design work is already done, taking it off the shoulders of the installer. With GSHP there are 100x more opportunities to screw it up.
I've had some other HVAC knowledgeable people look over t he Manual J for the 4 ton system. A couple of other irregularities jump out, but we don't know if they are mistakes or design choices. First he designed in 141 cfm of air infiltration for what should be a relatively tight house. Results in a roughly 10,000 BTU/hr loss in the winter due to air infiltration. Going room by room, he even has a closet with one 5' exterior wall and no windows as having 2.2 cfm of air infiltration through the framing/wall. The master bath has one 10' exterior wall with a 4 ft^2 window and he has 4.3 cfm of loss (window is rated for <.1 cfm/ft^2 @ 25 mph). Other rooms with larger walls and window have major loss. My gut says more loss than there should be, after it was pointed out to me (assuming what I was told was true).
Someone told me that per ASHRAE, a leaky house would have around a 10,000 BTU/hr loss, with a tight house around 3,000 and a very tight house around 1,000. I wouldn't consider my choices very tight, but not leaky either.
The other thing that was pointed out was there is a 36' long 8" trunk line, and 24' total of 6" branch lines. He is using a duct loss of 5000 BTU/hr of loss (heating) and 8600 BTU/hr of gain (cooling). All this lines are in conditioned space. I was told the heat loss should be about 5% and cooling gain 10% of a good install in conditioned space. So that would be 2400 BTU and 4800 BTU for a 4 ton system. So his losses are about double, but maybe he designed in lousy ducting, or it could be a mistake.
By inputting some correct (or at least as close to correct as his software allowed) u factors for doors and windows and SHGC, the 48000 BTU/hr heat sizing came down to 45000 BTU/hr. IF these other two items are poor estimates or bad design that is corrected, you cold shave close to another 10,000 BTU of heating off, again putting his max loss about the same as the loss from the Water Furnace quote.
There are also lots of obvious errors. A 1900 sq ft floor plan with a basement floor footage listed on the Manual J as 2006 sq ft (adding an additional 5% to the heat loss). The R value on the basement wall is R-3 less than what it really will be. Windows were corrected but doors are still wrong (he used a U of .39 for all doors when they vary from .19-.29). Rooms on the other side of the attached garage share the same infiltration rates as true exterior walls (even though the will never be hit with moving air).
Just lots of little things that seem out of place, but may add up to something significant. He is the only quote that uses all metal ducting, but then he will only bubble wrap it, which folks on here seem to think is a waste (he swears by it of course). All the other quotes used duct board, which everyone says stay away from.
What is better, R-4 duct board (Owens Corning) or metal ducting, but bubble wrapped?
All three quotes also want to use interior stud cavities as returns, which I've read you shouldn't do (so I'm assuming all quotes will go up across the board if I require ducted returns too).
I also don't understand why the 4 ton quote states 1227 sq ft/ton in the check figures (so 4 tons is 4908 sq ft, when the house is actually 3800 sq ft. That would be a 3.09 ton system for that sqft/ton figure. Seems like you would use a 3 ton system with .09 ton of electric heat on the peak days.
I really need to decide this week, but with so much confusion, and no one quoting exactly what I want, prospects seem bleek that I will figure it out in the next few days. :(
Four tons heating still seems on the very high side for a house you're describing.
What's your zip code, and what were they using for the outside design temp?
If you have heating bills from last year we can work backward from that to come up with an upper bound for load at a reasonable 99% design temp.
My own house is circa 1923 2x4 construction with known gaps in the wall insulation, ~R19 rock wool in in full-dimension 2x6 rafters, with antique double-hungs + clear storms for windows, with a comparable amount of conditioned space (2400' fully conditioned upper floors, 1500' of 65F+ basement) , but it DOES have an insulated foundation. It comes in a bit under 3 tons @ an outside design temp of +5F, as measured by the fuel use. Manual-J methods come in at ~40K, but simple math says if it were that high my mid-efficiency gas fired heating system would have to be over 100% efficient if it were actually that high, since it's including the domestic hot water too.
Wall cavities as ducts isn't even allowed by code anymore in some locations, and are guaranteed to be leaky. Any ducts should be mastic sealed on every joint & seam (or at least FSK-taped, if it's an appropriate surface.)
The heat loss from infiltration is about 0.018*V*ΔT so with 141cfm =8460 cubic feet per hour 10,000BTU/hr would imply a temperature difference of [10,000/(0.018 * 8460)= )65F, which implies a 99% outside design temp of +5F, if the interior temp is 70F. That's lower than any of the ASHRAE design temps listed for KY:
http://www.energystar.gov/ia/partners/bldrs_lenders_raters/downloads/Outdoor_Design_Conditions_508.pdf
And if your house is pretty tight, you won't have that much air moving without active ventilation (and unless you have a lot of indoor pollution sources you can turn the ventilation down when it's going to be super-cold out.) Allowing even 3000BTU/hr for infiltration losses might be high, but it's a very difficult thing to measure accurately since blower-door tests don't get at stack-effect and wind-penetration issues which vary a lot with where the air leaks are located.