GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

How high a percentage of glass (sq. ft.) before radiant floors become a reasonable option?

user-5967000 | Posted in Energy Efficiency and Durability on

In 1998, we purchased a house in NY right on the border between climate zone 5 and 6. It was dark…. I mean, really really dark inside and out. The house was 1.5 stories and about 1500sqft with a large room in the front with a high cathedral ceiling. Limited windows, large 2-car garage taking up the southwest corner.

In 2000, we (over)react to the lack of sunlight and I put on a 500sqft addition with cathedral ceilings and remove as many interior walls as I can. Much of the west wall became french door panels (some fixed). The three operating windows of the addition were large (40+ sqft) and there was a window seat with a similar sized fixed window. There was so little available wall space for hydronic baseboard that I explored radiant heat. I go all in and remove all the baseboard in the rest of the house. I purchase a 55gal oil fired water heater for $1k and $5k for a mile of piping and a ~1000 aluminum sheets. After endless hours of knuckle-breaking labor (and hand-molding every one of those aluminum sheets) I had a system that heat our house comfortably…… but it wasn’t cheap: I burned about 700 gallons of oil a year.

In Jan of 2016 our home burned to the ground. Nobody was (physically) hurt, but the fire was intense enough to crack the foundation. I have that rare opportunity to start completely from scratch. There are some things that will be different (get that garage out of my house footprint! let my roof just be a roof and get those mechanicals out of my attic!) but some things will be the same: Our property is beautiful and we like to “bring it inside” as much as possible, so the floor plan probably has even more glass than before. The current design has an 18ft folding (or sliding) wall into the screened porch taking up nearly half the west wall. Another key difference is that the first floor ceilings have been raised to 9ft, but no more cathedral ceilings.

The thought of re-installing the radiant heat system in DIY-fashion again fills me with dread. (The heat source would be different this time at least.) There is a good chance that I will use 3/4in maple flooring again. While I won’t have to deal with two sub-floors in the new house, I seriously doubt that the under-floor screw-up aluminum tracks will be sufficient. I made a promise that I would never ever never mold another sheet of aluminum. However, when I look at the prices of warmboard I get the uncomfortable feeling that I will be a man forsworn…..

I’m satisfied with my envelope, a 12in double-wall, and I have a good plan for air-infiltration. I have not had my mechanical engineer buddy do a true manual-j calculation yet, but the freely available software I have used puts my heating load in at ~25k. Unfortunately, that 25k is a composite number and won’t really tell me how it will feel to sit in the great room next to all that glass. I’m prepared to invest a fair amount of $$ for good glass, but glass is still glass…..

Good grief, that was a long preamble…… The actual question:

Can a locally-under-performing region of an otherwise high-performing house justify the use of radiant floors?

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. GBA Editor
    Martin Holladay | | #1

    Q. "Can a locally-under-performing region of an otherwise high-performing house justify the use of radiant floors?"

    A. Yes, but it's not the only solution. The upgrade to triple-glazed windows may be cheaper than the installation of a hydronic heating system.

    First, perform your Manual J calculations. Then come back to the Q&A forum.

  2. STEPHEN SHEEHY | | #2

    I lived with radiant in my old house. My biggest gripe was that since, like you, I like the sun streaming in, the heat often didn't run at all during the day because the sun heated the space. Then it would get chilly when the sun went down. But since radiant is so slow to react to the need for heat, by the time the floor got warm, it was time to go to sleep.

    My new house is tight and well insulated. Heat is two mini-splits, although most of the time we only turn one on. The last few days have been cool (30-50F) but sunny. The sun heats my space, but since it is so well insulated, the heat doesn't come on at all, even at night. But earlier in the winter, when the space cooled off in the evening, the mini-splits did a fine job of keeping the temp where we wanted. We're in Maine, zone 6.

    You mention a screen porch. Will you install windows or just screens? Our "porch" is on the east end of the south side and has pretty good windows and is insulated to some extent. In winter it tends to insulate the adjacent conditioned space. I think if your porch has even cheap windows in winter, you'll be comfortable in the great room even with all that glass, assuming the glass wall has a good U value.

    The rest of our great room has a south facing wall with a section of windows about 11' x 11.' We've never experienced any drafts or felt cold sitting next to the windows. The windows are triple pane, about U.12.

  3. user-5967000 | | #3

    I'll get two Manual J reports for the house. One with mid-range and the other with high-range windows.

    I lean towards the minisplits (we have ~15Mwh/year coming in on two pole-mounted dual-axis trackers) but it is a tough sell for my wife. I think I need to package the minisplits with something else: "Look at what could fit in the budget now!"

    I'll be back when I have more hard data.

  4. charlie_sullivan | | #4

    What about the mini-splits is a hard sell? There are various options that might mitigate or eliminate those concerns.

  5. user-5967000 | | #5

    It is a case of engineer-marries-artist.
    The first minisplit she saw was rather "prominently displayed".
    The nature of the loss injects a certain amount of emotion into decision making. (rather strong desire to replace what we had)

    I'm seeing some movement on #3, but I'd need to find ways to really hide the downstairs ones without killing its efficiency.

  6. charlie_sullivan | | #6

    Hmm, there are two issues then. One is desire to regain what was lost; the other just the aesthetics. The aesthetics can be mitigated by opting for "ducted mini-splits" which can be hidden between a ceiling and a floor, or in a soffit or the top of a closet, on the top floor, which ducts running short distances.

    But regaining what was lost is harder. One thing that might help a little is to consider radiant ceilings rather than floors. They sound counter-intuitive, but they can work well, and in fact if you are sitting on the couch, your skin receives more radiation from the ceiling than the floor. They can be a little cheaper to install. But a hydronic system is still going to be expensive overall, and it can be hard to find people to design and install it right.

    If part of the appeal of mini-splits is to move to a heat pump as the source instead of combustion, good air-to-water heat pumps are now available. Martin had a good summary of those options in a recent blog post.

    You can also supplement minisplits with radiant electric heat, such as radiant cove heaters. One of those over big windows, turned on only when you are sitting there, could eliminate local comfort issues.

  7. Foamer | | #7

    I will second Charlie Sullivan's suggestion if you decide on radiant after all. Put it in the ceiling. I did so in my house and it works extremely well. Because of the low mass of the drywall, reaction time is fast and the system provides very even heat. Comfort is very subjective but in my book, nothing beats radiant.

  8. Dana1 | | #8

    A high-R wall and high performance windows raise the average radiant temperature of the room too independent of what's supplying the heat. The higher the building envelope performance, the lower the surface temp of the radiant floor or ceiling. With a 12 inch double-wall in a zone 5A/6A boundary climate you'd notice the radiant heat on the coldest hours of the coldest days, but during the average wintertime mid-day, not so much, unless you concentrated the radiation into just the areas where you might be spending your time.

    Low temp panel radiators installed under windows does a LOT for average radiation temp for enhanced comfort, at lower cost & complexity than radiant floors/ceilings, and it's more responsive too.

    A wood stove delivers a more concentrated radiant heating effect near the stove, and due to the high air temps convecting off the stove, ends up raising the ceiling temperatures considerably, which further enhances radiant comfort. When it's -10F outside a room heated with wood stove is more comfortable than a room heated with a radiant floor. (Subjective, but most would agree, not counting the 5% of people who are never ever comfortable.)

    A 25K heat load would be on the high side for a 1500'-2000' house with R40+ walls, which implies a lot of glass even with code-min windows. Run a BeOpt model of the house to figure out where the load is coming from, and how it might most cost effectively be reduced.

  9. user-5967000 | | #9

    I actually already have a BeOpt sim in flight...... I could not figure out how to specify exactly where my windows are..... just generic % of wall..... and even that was drop-down menu style.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |