My own home heat analysis

[Nanker Phelge]

I am just trying to get my ducks in a row, planning to make the big switch from electrical baseboard to combi boiler flat panel heat for next winter. I have put in a good deal of time reading, and completed, what I feel is an honest and extensive heat analysis for my home. I published the results in a Google spreadsheet online and wonder if you don't mind taking a look at my numbers to see if they do in fact appear to jive. I determined most of my details when I spent the few hours it took to go through the "Sizing your Hydronic System" offered up by FloProTeam, if anyone is familiar with that seeming useful resource.

The link for the spreadsheet containing my home's heat analysis is:

https://docs.google.com/spreadsheets/d/1zskNopx2M1XibIp4FCVn5r9xao3XmDiC4PqKKMuvQ3Y/pubhtml

Thanks in advance for any constructive feedback, as this project is a financially daunting one for us, and I would feel better having some idea that I am on the right track.

 

[Jadnashua]

Can you get a reasonable estimate of your electricity use for heating verses the rest of the house? 1W=3.412BTU. Using that energy use and the local degree-day load should give you a pretty good alternate analysis to compare with what you have.

 

[Sponsor]

 

[Dana]

Your U-factors are all wrong. It appears there is nothing factored in to account for the thermal bridging of the framing.

The U-factor of 2x6 framing 16" o.c. with R19s with half inch CDX or OSB with wood siding, half-inch wallboard comes in at about U0.69- U0.71, depending on the actual framing fraction.

The U-factor of an R19 floor over a pier foundation, well ventilated crawl space, or cantilever is about U0.065 (better than walls, since there is lower framing fraction.)

For a ceiling to hit a U-factor of 0.025 would require insulation north of R50. A typical R49 blown-fiberglass installation comes in at about U0.026, a bit lower in trusses depending on how much insulation there is above the truss chords. If you have and R40-ish 11.25" of perfectly installed cellulose between 2x12 joists, use U0.027. If it's R40-ish blown fiberglass use U0.030. If it's trusses, tell me the depth & type of insulation, and what the truss chord elements are.

The 99th percentile temperature bin and outside design temp in Int'l Falls MN is -23F, and the minimum legal interior design temp is +68F, or a delta-T of 91F. Use that as the starting point for sizing the equipment rather than -25F and +72F, a delta-T of 98F. Only in the rarest of cases would the equipment sized to cover a 91F difference not also have ample margin over a 98F difference.

Clearly your infiltration numbers are on the high side, and a significant driver of the load calculation. Blower door and infra-red imaging directed air sealing is usually the cheapest building performance and comfort enhancements you can buy, and the blower door numbers would give you a better handle on the range of natural infiltration you might experience.

In your climate U0.36 windows are almost a crime. Depending on the window type you may be able to cost effectively upgrade them by installing low-E exterior storm windows, which would bring the performance down to ~U0.28 or so, at a fraction of the cost of a U0.28 replacement window.

What foundation type, and what type of foundation insulation?

With mid summer (non air conditioning use) electric bills as a baseline, it's possible to derive the heat load from a few mid-winter electric bills if given the exact meter reading dates and power use between those dates, and a ZIP code for looking up the weather data for those dates at a nearby weather station on degreedays.net.

 

[Nanker Phelge]

Your U-factors are all wrong. It appears there is nothing factored in to account for the thermal bridging of the framing.

You are correct. The house is a bit of a hodge-podge, having gone gone through what appears to me like at least 3 major renovations-additions since it was first built in 1952. Original sections of the house have brown buffalo board still in pkace over 2x4 framing, while newer sections have wrapped OSB.

After observing frost build up on small sections of an exterior wall in the kids' bedroom this past (mildish) winter, I opened up a small section of the wall inside, discovering that what little insulation was there was matted down to the bottom of the framing. So my intention is to attach 2x2 strips to the 2x4 framing to extend it to 2x6 and fill with R19 and a vapor barrier. One room at a time.

I am also contemplating replacingng exterior sheeted buffalo board with OSB on the exterior and wrapping with tyvek-like wrap or foam board and possibly both.

So I didn't have any real concrete numbers to plug in for thermal bridging between framing.

Ifit's trusses, tell me the depth & type of insulation, and what the truss chord elements are.

Will do. Ceiling imsulation is a cellulose type material with saw dust. I will get up there and measure the depth, but it does cover the 2x6 rafters.

Clearly your infiltration numbers are on the high side, and a significant driver of the load calculation

I am definetely a worst-case scenario type person, so I tended toward the high side with most numbers in doubt.

In your climate U0.36 windows are almost a crime. Depending on the window type

I used generic numbers here for leaky double pane windows. There are no "U" values or other related info listed on the windows other than they were argon filled. But with the amount of condensation build up observed in most of windows, I surmised any seal between panes has long since been compromised. Maybe a naive assumption on my part?

Whatfoundation type, and what type of foundation insulation?

From what I have seen sofar, having pulled up some flooring in the split-level's lower level, is plywood over concrete.

it's possible to derive the heat load from a few mid-winter electric bills if given the exact meter reading dates and power use between those dates

Unfortunately I wasn't savvy enough capture any detailed usage beyond monthly bills. In addition, we did not keep the house evenly heated 24/7. The living space was kept as comfortable as we dared during the day while the bedroom temps were maintained @ about 60-62, with the reverse at night. We kept the lower level of the house at a constant 58 as well, all in an effort to avoid obscene electric bills.

So although I know my numbers are surely skewed toward the high side, I just thought that going with a boiler like Westinghouse's NG80 that modulates between 8k an 80k, it shouldn't matter so much, except perhaps to properly size radiators for the individual rooms.

I have come to appreciate learning what the detrimental effects over-sizing a boiler has on efficiency, but what about over-sizing radiators? I thought perhaps by using cast iron radiators with output BTUs tending toward the high side, but well within the chosen boiler's total maximum output that the boiler would be ensured of a solid load while the occupants would be assured of a comfortable even, yet affordable degree of heat throughout the house. Perhaps another foolhardy assumption?

I really appreciate your replies and look forward to additional feedback to aid me toward more knowlegable decisions that will have to be made well before next winter.

 

[Dana]

You only skewed the air leakage numbers to the high side. Your wall U-factors are MUCH lower than reality. The thermal bridging is the framing- the studs & plates, headers, jack studs, etc which all add up to about 25% of the surface area of the wall. A 2x6 stud on edge using most midwestern framing species runs only about R6.5, which is a heluva lot less than the fluffy stuff between the studs. And an R19 batt performs only at R18 when compressed to 5.5" (the depth of a 2x6 stud bay). With 25% of the area having about 3x the thermal conductivity of the other 75%, there's about as much heat flowing through the framing as there is flowing through the fiberglass in a typical 16"o.c. stud wall, which lowers the average performance by quite a bit. If you assume the exterior sheathing & siding and inteiror gypsum adds up to about R1, and you allow another R1 for the combined interior & exterior air films it comes out to about R14-R14.5 for a "whole-wall R", the average performance, and the U-factor is the inverse of that:

1/R14= U0.071

1/R14.5= U0.069

Reality will be somewhere in that range, unless you have insulating sheathing, a rainscreen siding stackup, an unusually insulating type of siding, or if the studs are 24" on center or something.

The Buffalo Board is fiberboard sheathing, which is about R1- R1.5 better than half-inch CDX or OSB and more water resilient, and more permeable to water vapor, which means it letsl the wall dry toward the exterior, which is a good thing. (Half inch fiberboard adds about R1, and 3/4" fiberboard adds about R1.5 above what it would be with half-inch plywood.) If it's not warped beyond recognition and isn't pulling off the nails, don't swap it out for OSB/CDX, since the OSB/CDX are vapor retardent, and will take on a significant moisture burden over a Fort Frances winter. You can seal the seams & edges with duct mastic when you have it open, which will cut down on infiltration.

If we add R1-1.5 to the whole-wall to account for fact that it's fiberboard rather than plywood your new-improved wall U-factor is roughly

1/R15.5= U0.065

1/R16= U0.063

That's better than a plywood sheathed wall, but still a lot more wall loss than the U0.05 you had used.

Most argon-field single-lowE windows run about U0.32-0.34, not that it makes a huge difference. Condensation on the window is an indication that it's actually cold outside, and that you either humidify the place (actively with a humidifier, or passively by cooking pasta without a lid on the pot, or taking long showers without using a bathroom fan.) Unless there is fog on the inside of the sealed panes, the panes are still doing their job. A low-E window with a broken seal and air instead of argon will still operate at U0.37 or so.

Oversizing the radiation is fine, and beneficial for condensing efficiency, since it can deliver the heat at a lower temperature. But the oversizing has to be proportional in each room, so getting the load numbers as accurately as possible from the get-go is the right approach. Be aggressive rather than conservative when it's an unknown (such as the actual air leakage) and you're more likely to get it right. It's only human to skew to the other direction, but that's always a mistake (ALWAYS!).

Run the load numbers realistically, but as aggressively as possible. (You at least have to get the U-factors in the right range, which you don't.) When it comes time to size the boiler and radiation, use ASHRAE's 1.4x oversizing recommendation as the absolute upper limit. eg: If the load numbers come out to 47,000 BTU/hr you don't need anything with an output more than 1.4x 47,000= 65,800 BTU/hr even during the coldest night of the decade.

With condensing boilers try to design the radiation so that you don't need an average water temp greater than ~140F at design temp, in which case it will be in condensing mode most of the time. It's nearly impossible to oversize the radiation to where it can't work, but undersizing can be a problem.

Say a room has calculated load of 2200 BTU/hr @ -23F. If you upsize the radiation 1.4x, you'd be at 3080 BTU/hr. If you used the 180F output of say, a Biasi panel radiator , the B-24.16 ECO would about cover the upsized number at 180F, but would come up a bit short with only 1,836 BTU//hr, but close enough that it's probably the right radiator for that room. If you sized the radiator for 3080BTU/hr @ 140F you'd take the B-20.32 ECO or B-12.48 ECO, which would pretty much guarantee that you'd be running in condensing mode 95% or more of the time once you have it all dialed in. To be conservative never size it for it's 180F output at the lower, unscaled number, only after the 1.4x scaling factor. THIS is the right place to oversize "just to be sure", not at the heat load calculation point. If it's sized for the 1.4x scaled load at 140F it will be guaranteed to heat the place, even if there have been errors of omission/commission and it actually needs 150F water.

You could also use fin-tube baseboard to save money in rooms you don't really use much and comfort isn't a premium, or reuse some antique (or now) cast iron, taking care to estimate it's output at 140F, not just 180F (or 220F).