Heating and moisture strategy
I have some questions about how to proceed in terms of a home heating solution and dealing with humidity issues. The 2015 house includes the following:
· 656 ft2 basement, poured using ICFs, 2” foam below slab, 2×8 (R-30 roxul) framing on a portion of the exposed wall;
· 656 ft2 first floor, 2×6 framing (R-23 roxul);
· 576 ft2 second floor, 2×6 framing (R-23 roxul);
· Blown in fiberglass (~R-50) in attic.
A blower door test estimated an ACH50 of about 1.4. The house is located in Lubec, ME (04652).
The house is equipped with an NTI 151C propane boiler and is divided into three heating zones (one per floor). Generally speaking, only the first floor of the house is actively heated. The basement stays about 60 degrees without supplemental heating, doors to the two bedrooms on the second floor remain closed most of the time. Based on information gleaned from the internet, including Dana’s input to other forums, the boiler appears to be vastly oversized with its heating capacity of 74 MBH. When the heat is turned on, the unit appears to run continuously (i.e., it cycles infrequently if at all), not helped by the fact that there may be too little baseboard (~19’ on the first floor) to deliver the generated heat.
Short of a formal Manual J calculation, would you agree that the system is oversized? Will the efficiency of the system be improved if I increase the amount of baseboard on the first floor? The water delivered by the boiler is 170+ degrees which, based on info found on the internet, translates to about 500 MBH per linear foot of baseboard. Using a heating load of about 16,000 MBH for the first floor (based on a seat of the pants estimate using tools I found on the internet), something on the order of 30’ of baseboard appears to be more in line with what is needed. I suspect the boiler still won’t be operating as efficiently as it should even with the proper amount of baseboard, right?
I also have been experiencing moisture issues since I moved into the house. Indoor wintertime RHs generally range from the low to upper 50s; I use a dehumidifier to dry things out a bit at times. I tend to think the low natural ventilation rate is the primary culprit as the lifestyle of the inhabitants is not excessive in terms of generating moisture (two plants, religious use of exhaust fans during showers, no known leaks, relatively little cooking (albeit with a gas stove/oven), etc.). There is no active ventilation in the house, I am considering installation of one or more through-wall units such as the Lunos e2, Zehnder Comfortair 70, others).
Would the adoption of one or more ductless mini-splits help me heat my house more efficiently and help me deal with moisture issues at the same time? Presumably, the mini-split could be sized appropriately so it efficiently heats the first floor, and I have heard/read a bit about the ability of mini-splits to dehumidify. I found a 2013 discussion in which Dana mentioned the Daikin Quaternity series has a dehumidistat control that dehumidifies in heating mode. I don’t know if other manufacturers have come up with similar capabilities since that time.
Any confirmation/direction people can provide wrt the heating and humidity issues described above will be greatly appreciated.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Minisplits won't reduce humidity in heating mode. Field reports indicate that the Quaternity doesn't either, though it can dehumidify without significant sensible cooling. That series also isn't rated for cold climate use, and has neither the efficiency or capacity to make it worthwhile.
To deal with excess humidity in winter, increase the ventilation rate. From experience I can say that the Lunos e2 is a fairly simple retrofit, but others can be as well.
I'm not sure your analysis on the boiler is completely correct, but yes it's oversized for your house, but the Tx 151C combi boiler is able to throttle back to under 20,000 BTU/hr out. It's still going to cycle at 170F output on a 19' zone- just because it's "running" constantly doesn't mean it's BURNING constantly. To get a handle on the heat load you can run a fuel use load calculation to ball park it, but you're looking at less than 10,000 BTU/hr per floor @ -5F. With a reasonably open floor plan on the first floor, the bigger capacity 3/4 ton cold-climate mini-split Fujitsu -9RLS3H would likely fully cover the load of the first floor into negative double-digit temps, and with open doorways & stairwells much of the second floor as well. (The Mitsubishi FH09NA probably would too, but doesn't have quite as much capacity as the 9RLS3H.)
It's worth doing a room by room IBR type heat load calculation to ball park this. You'll have to calculate the U-factors on the non-standard wall framing, but it's possible to get pretty close without doing the full Manual-J.
A house that tight absolutely needs mechanical ventilation. And swapping out your gas oven/ cooktop for electric/ induction should help as well, since burning propane produces lots of water vapor. It'll improve indoor air quality as well, since burning gas also produces CO2 and frequently, some CO.
The amount of radiation (length of baseboard) does also affect the efficiency of the boiler. If you have too little BB, the boiler water must be hotter to achieve the necessary heating capacity. Hotter water means lower boiler efficiency. Using more BB will allow lower boiler water temperatures, longer burning cycles and greater overall boiler efficiency.
If you have no A/C now, adding a minisplit or two would give you that functionality along with dehumidification in cooling or dehumidification mode. It would also probably save money over the propane boiler, at least in moderate conditions. Once you know your crossover point, you can shut off the minisplits and turn on the boiler when the weather gets really cold. Since this system is supplemental, I would go for a minisplit with the best efficiency at moderate temperatures (above 0F or even a bit higher) and low cost, rather than trying for best performance at your design minimum temperature.
This is still pretty pricey, though. If you don't really need/want air conditioning, you'd probably do as well with less money by using a ventilation-only solution in winter as Dana suggests above.
It's much more (relative to other effects) efficient if radiators are designed such that outdoor reset will use temperatures well under 170F (into the condensing range) under most conditions. Your loads are evidently too small to heat without excessive cycling - so add a buffer tank.
In Maine's average three buck propane & 22 cent electricity markets there is never an operating cost advantage to switching from heating with mini-splits even if you do all the necessary changes to get it into the condensing zone.
At 95% efficiency you get about 87,000 BTU/gallon out of propane, running at 170F it'll be more like 85% efficiency and 78,000 BTU/gallon. At a winter average HSPF of 10 it take 8.7 kwh to deliver the 87,000 BTU "gallons-worth" of heat, at a cost of less than $2. When it's below zero a better class cold climate mini-split might only deliver 6000 BTU/kwh (a COP of about 1.8) so it would take about 14.5 kwh for that condensing gallons-worth, which at 22 cents is $3.19, about the same cost of just the fuel for the condensing boiler, not counting the electricity or distribution losses.
When it's -15F outside the COP might only be about 1.5, but it isn't that cold enough hours in a year to matter in Lubec, ME. The outdoor temperatures in Lubec are comparable to Eastport a few miles away, where the mean temperature even in the coldest week of January is over +20F:
https://weatherspark.com/m/27853/1/Average-Weather-in-January-in-Eastport-Maine-United-States#Sections-Temperature
At any modulation level at +17F the Fujitsu 9RLS3H has a COP greater than 1.8 (6000 BTU/kwh) , and at it's AHRI rated output has a COP over 3 (over 10,000 BTU/kwh) at that temp. The COPs improve from there as outdoor temperatures rise.
Bottom line- in most of Maine the money is better spent on a cold climate mini-split, not buffer tanks &/0r more radiation.
Thanks for the replies.
So, the ventilation issue should be addressed separately using devices such as the Lunos e2 or similar equipment.
Dana, you make a distinction between my boiler running all the time and burning all the time. I am used to oil-fired boilers that run until the desired temperature is achieved, turn off for a period of time, then start all over again. Is it ‘normal’ for my propane boiler to seem to be running most of the time? Is the fact that it runs a lot due, in part, to there not being enough baseboard to efficiently distribute the heat being generated? In other words, am I somehow ‘wasting’ the output of the boiler because of limited distribution?
I performed a heat loss calculation using the IBR approach laid out by Martin several years back. I compared the results of my effort to those given using the calculator at https://www.builditsolar.com/References/Calculators/HeatLoss/HeatLoss.htm. I did the calculations on a floor by floor basis (rather than room by room) because the floor of greatest interest (first floor) is wide open save for a small bathroom. I used indoor and outdoor temps of 70 and -1 degrees F, the latter is the 97.5% temp for Portland, ME.
My results indicate heat losses of 7700, 7000, and 6000 for the basement, first floor, and second floor, respectively. The loss for the basement was dominated by losses through the slab which were calculated assuming R-10 for the 2” of foam and the temperature gradient cited above. The gradient is no doubt the wrong one to use but I am not sure what should be used in this case. The losses given above do not include infiltration losses. I estimated those for the entire house by multiplying the temp gradient, a ventilation rate of 60 cfm, and a factor of 1.08. The ventilation rate was an assumption of what may result from active ventilation, the factor was taken from a blog by Martin. Losses from infiltration were estimated to be 4500 BTU/h.
My calculations compare favorably to those estimated using the calculator link provided above. So, if nothing else, there is that consistency.
In terms of operating costs, my last shipment of propane was $2.55/gallon while my last electric bill indicated 19.23 cents per kWh. By my calculations, these prices translate to 0.0028 cents/BTU for propane and .0056 cents/BTU for electricity. Assuming the boiler is 100 percent efficient, this suggests I come out ahead as long as I can achieve a COP of >2 (right?). Using the 85% efficiency cited by Dana (for 170 degree water), a COP cutoff of 1.7 is calculated. It sounds like these COPs are reasonable to expect overall.
So, operationally, it appears to make sense to get a mini-split. How much does a unit the size suggested by Dana cost to buy and install? I saw a (perhaps dated) estimated from Dana on another forum that suggested $3500-$4000 per ton (if I recall correctly). For simplicity, let’s assume an installed cost of $4000. In my time spent in my house (only 1.5 years) I estimate the use of about 450 gallons of propane per year or about $1100. At an average COP of, say, 2.5, I would save about 33% on the cost of a BTU, saving $350-400 per year. In other words, it would take about 10 years to pay off the installed heat pump. Given the assumed costs, the longevities of propane boilers and heat pumps (which I know absolutely nothing about), and the assumed COP, is this a reasonable result?
I don’t really care about the cooling abilities of the equipment. Very hot humid weather is relatively rare where I live and I prefer to just suffer a bit rather than pay to cool things down (a cheapskate at heart).
I have read on GBA about issues regarding heat pump operational efficiency and the height at which the indoor unit is mounted. Does it make sense to opt for a floor unit instead, assuming either placement option is acceptable aesthetically?
Thanks again for the input.
>"I am used to oil-fired boilers that run until the desired temperature is achieved, turn off for a period of time, then start all over again. Is it ‘normal’ for my propane boiler to seem to be running most of the time? Is the fact that it runs a lot due, in part, to there not being enough baseboard to efficiently distribute the heat being generated? In other words, am I somehow ‘wasting’ the output of the boiler because of limited distribution?"
Oil boilers will also turn the burners on/off during an extended call for heat when oversized for the active zone radiation. But an oil boiler has some thermal mass to work with, and the burn cycles will be long, potentially long enough to satisfy the thermostat with. With a low mass wall hung propane boiler that's ~2x oversized for the zone radiation the burn cycles will be much shorter.
If the boiler were set up with an outdoor temperature sensor the heating system would run extremely long cycles, even if the burn cycles were short. But you're indicating it's running at a fixed output temperature of 170F?
If the burn cycles during those long calls for heat are very short (<>5) it will impact efficiency due to the flue purge heat and fuel loss during ignition cycles. If the burn cycles are longer and fewer per hour the average efficiency will be pretty close to it's steady state combustion efficiency. At 170F=out that would be about 85%, at 120F=out it would be in the mid 90s.
The output isn't being wasted due to the limited radiation, but if it's short cycling most of the time the as-used efficiency can be as much as 10% lower than it's steady state combustion efficiency, and it's putting some wear & tear on the boiler.
The builditsolar load calculator is really a piece of junk, with wildly inaccurate U-factors. (Gary should really fix that.) But if you're only using 450 gallons/year your heat load is pretty low. If you have heating degree-day data for those years it's possible to estimate the heat load based on that fuel use.
Over the past 12 months the weather station at Lubec ME experienced about 6000 HDD, base 60F (per degreedays.net datasets.) That's 0.075 gallons per HDD. If it's running at 85% efficiency that's 0.075 x 78,000BTU/gallon= 5850 BTU per degree-day, or (/24 hours=) 244 BTU per degree hour for every degree below the presumptive 60F heating/cooling balance point (the degree-day base).
So when it's -5F outside the load will be about 60F - (-5F)= 65 heating degrees x 244= 15,860 BTU/hr, which is a hair more than the maximum output of the 9RLS3H at that temperature. Since some of that fuel use was for hot water, your real load is probably still less than that.
In my area a 3/4 ton cold climate mini-split runs about $3-3.5K, if bid competitively- definitely not $4K. With a mostly DIY installation buying the equipment online it's possible to install the 9RLS3H for about $2500 give or take, depending on how much the local refrigeration tech charges for the service call for the final refrigerant charging & commissioning tests.
If your heat load is ~14-15K at -5F, your actual average COP on that unit will be north of 3, not 2.5. For an ARI zone IV climate it's HSPF is 14.o, which is an average COP of 4.1. Lubec is in zone V, but at your high-R building envelope that's roughly the same heating hours as a zone IV location. But since its a bit cooler than most zone IV locations it will probably have an annualized COP average closer to 3.5. See:
https://s3.amazonaws.com/greenbuildingadvisor.s3.tauntoncloud.com/app/uploads/2018/08/08081647/HSPF%20map%20-%20FSEC-700x511.jpg
Do NOT conflate ARI zone numbers with DOE zone numbers, though they sometimes overlap. The ARI zones defined by the number of annual heating hours (at any temperature), whereas DOE zone numbers are based on base 65F heating & cooling degree-days.
If you have high Winter humidity, try to reduce source moisture. Sounds like that's mostly done, so install a HRV (not ERV).
Note: Efficiency Maine will subsidize cold climate heat pumps, but only $500/unit (independent of size), if professionally installed.
https://www.efficiencymaine.com/heat-pumps/
In MA (where you don't live) there's a MassSave subsidy to the tune of $1600/ton for displacing propane or oil heat, for any ductless units that appear on the NEEP cold climate heat pump list:
https://neep.org/node/3725/download/d4470144b29894e18ece9f2ff2eb3140
https://neep.org/system/files/ColdClimateAir-SourceHeatPumpSpecificationProductListing-Updated1.2.19.xlsx
It wouldn't surprise me that under the new-improved executive that Maine might eventually be good for more than $500.
So, in actuality, I do have an outdoor sensor. My comment regarding the 170 run temp was based on a subjective (wrong) assessment (i.e., glances now and then) of what it always seemed to be doing. Watching the boiler panel a bit more this morning indicates the unit cycles between the low 150s F and low to mid 170s F. Water temps increase over this range for 15-20 seconds and then fall back in about the same amount of time at which point the cycle is repeated. Sorry for the misleading information.
Re the builditsolar calculator, I used my U-values in the software. So, mostly, I guess I was just checking that my spreadsheet and the calculator were crunching the numbers in the same manner. So, probably nothing more than a relatively worthless consistency check.
I estimated the heat load from my fuel usage data and came up with numbers similar to yours for 85% efficiency at -5F (15.6K and 13.9K BTU/h at 60F and 65F). The info I have found about the Fujitsu 9RLS3H only provides min and max heat output (3100 and 22000 BTU/h). It sounds like you have something that provides more temp-specific results, is there a link for that? I suspect my fuel usage-based results are susceptible to uncertainty introduced by small sample sizes (of fuel usage and HDD). But they, along with my earlier calcs, help build a consensus.
The $3-3.5K pricing is good to know, I’ll have to see how that translates to ME. That price range and the $500 rebate cause the economics of a mini-split to improve significantly. The news gets even better as the average COP increases. As you note Dana, the new administration could bring even better news but its always hard to predict if and when that might come to pass. (I’m hoping Maine might also look into offering rebates on solar installations, currently we only have the federal rebates to look forward to).
Dana, I got the (mistaken?) impression that the 9RLS3H is an older model and has been replaced by the 9RLS3Y. The minimum operating temp of the former is -15F while that listed for the 9RLS3Y is -5F. Do you know anything about this and, if it is true, is there reason to consider other models?
Finally (for now😉), how are you ‘sizing’ the unit for my application? That is, are you simply looking at output vs temp and comparing it to ~15 BTU/h heat loss, working a 1.4 factor into the mix, or something else. I’m not questioning your answer per se, just wondering how you are arriving at your recommendation.
As usual, thanks!!
The NEEP spreadsheet has min/max capacity numbers for the 9RLS3H at +47F, +17F, +5F, and -15F.
It also has capacity numbers for the identical in all respects (except no pan heater for clearing defrost ice) 9RLS3 (no -H) at +47F, +17F, +5F, and -5F. If you compare the capacity numbers at all of the listed temperatures shared by both you'll see they are identical, so the inferred capacity for the -H version at -5F is almost certainly correct.
That spreadsheet is downloadable from the link on this page:
https://neep.org/node/3725/download/d4470144b29894e18ece9f2ff2eb3140
The ASU-9RLS3Y is the indoor unit, and is compatible with both the AOU-9RLS3, and AOU-9RLS3H outdoor units, as is the older design (but still current) ASU-9RLS3 head. The capacity numbers are identical, but the ASU-9RLS3Y delivers slightly higher efficiency.
At subzero temperatures the capacity difference between the 12RLS3H and 9RLS3H is pretty small. At -5F it's only a 1500 BTU/hr increase (about 10% more heat) and at -15F that shrinks to only a 500 BTU/hr increase (about 5% more heat.) Whether it's worth the upcharge for going with the 1-ton is up to you.
But moving up to the 15RLS3H is a much bigger jump and can deliver 16000 BTU/hr @ -15F compared to 11-11.5K @ -15F for her smaller sisters.