Howdy;
I’m trying to figure out how to best heat a volume 30×48 with around a 30ft peak at the top of the room. This is the sanctuary of an old church being turned into the “great room” for our home. I don’t want forced air for all the normal reasons. The floor is 100 year old 1″ thick pine t&g. The interior walls are currently stripped back to the brick, which I believe it 3 layers thick to the outside. The brick gets cold but not freezing on winter nights, though there’s little direct heat in this room yet (leaked heat from a duct that won’t close too tightly). It’s a hot roof with icynene from peak to eave. We’re located in Central Illinois.
I really like the idea of radiant hydronic, but have been told by one professional that under-floor heating (assuming we keep the old flooring, which we fully intend to) would have to run pretty hot to radiate effectively through that wood, and even then would not be sufficient if we ever put an area rug down. Also, my guess is that without any real “good” mass, you would be running the heating a lot to maintain a temperature, and not get the stored benefits of something like concrete.
So this is my next thought. We’re thinking of putting a knee wall around the outer edge brick walls. It would be about 4-5ft tall, and provide a nice visual feature to offset the very tall brick walls. I think knee-wall is the right term, please correct me if I’m wrong.
Here’s my thought – what if I poured a series of concrete panels, say 3″ thick and 3’x3′ in size (the sizes are debatable and not critical at this point). In each panel I would have the normal metal lattice for strength, and also run a loop of PEX the would enter and exit on opposite ends. These panels become the thermal masses. Then I put some foam insulation panels between the brick walls and the cement panels, to radiate the heat in the right direction, and connect each of the panels in series around the room. The center of the room will have a loft overhanging it that could have a standard radiant ceiling panel, but most of the areas people will gravitate to are near the edges and corners of the room.
My understanding is that radiant works like line-of-sight/infrared behavior, and that firing into the room from the walls would produce a nicely heated set of surfaces low in the room and not waste a lot of hot air up in the rafters.
Any thoughts on this idea? Go ahead and be brutal. I’m making this up as I go along. =)
My kingdom for more tools
Replies
I like your idea - but pex lines are often spaced every 8" (2 per typical joist bay) I don't know the calcs, but I am guessing you would need a lot of pex to heat your space, maybe more than would fit in your proposed wall system.
A few questions:
Can you get to the underside of the pine floor? Staple-up with aluminum plates can send up a lot of heat. I have 3/4" yellow pine over 3/4" advantech and 145 degree water - its warm through the socks...
Could you pull up and re-lay the floor? Whats under it?
Would you consider a "belt and suspenders" approach - since you have high heat loss walls, and the floor may not be adequate - tubes in the floors and in the 4" wainscot/concrete/shelf 4' high around the room - nice potential design element, btw, the shelf created could be topped with a nice piece of walnut etc.
Is your ceiling up yet? Radiant in the ceiling adds heat too...
Let us know what you do, and definitely run heat loss calcs - this sounds like a great project...
Brian - you make a good point about the spacing. I think we could maintain the 8 inch span between lines, but it's still a question of being able to produce enough heat from the knee-wall, which is going to be less surface area (but a better mass) than the floor.The concrete wouldn't have to be run as hot as a wood floor, to give the equivalent feeling of warmth, right?We do have access to the underside of the floor for staple-up, if that turns out to be the best solution. I'd like radiant to be the primary heating mechanism for this room though.So many questions. I ordered a book on radiant hydronic heating. Hopefully it can enlighten me a bit.--------------------------
My kingdom for more tools
I think we could maintain the 8 inch span between lines, (not necessary) but it's still a question of being able to produce enough heat from the knee-wall, which is going to be less surface area (but a better mass) than the floor.
Think of your decorative concrete walls as oversized, mass-oriented radiant baseboard.
The concrete wouldn't have to be run as hot as a wood floor, to give the equivalent feeling of warmth, right?
You will need all the tubing you can get to get that concrete hot, since it will be less surface area than the floor. Higher design temps might help too, but you have 624 s.f. with the walls (assuming 4' high) versus 1440 for the floor. (your room is 30x48, right?) I would put tubes under your floor either way, since it feels so good on the feet! (plus tubing is the cheap part of radiant) I would think the same amount of tubing within the envelope will generate about the same heat gain, regardless of where it is. The trick is getting the heat out of the tubes, which concrete excels at.
We do have access to the underside of the floor for staple-up, if that turns out to be the best solution. I'd like radiant to be the primary heating mechanism for this room though.
Absolutely - stick with radiant if at all possible. I'll never go back...
Treat every person you meet like you will know them the rest of your life - you just might!
So you're saying floor and walls would be the best possible option...I won't disagree, but I'll admit I was hoping that the wall based heat would be sufficient and comfortable without the added time and cost of the staple up. To do the floor I'd need the Pex, the staple up reflective panels, and also... insulation? or do the metal panels reflect well enough?Also, will I need a bigger boiler to handle the water for floor an walls? I'm planning on a closed system, if that makes a difference.--------------------------
My kingdom for more tools
boiler size - you've gotta run heatloss calcs...
Treat every person you meet like you will know them the rest of your life - you just might!
My heatloss calcs consist of:hl = 1/(cw + cd)Key:
hl: heatloss
cw: # of closed windows
cd: # of closed doorsYou mean there's more to it than that? =) Any suggestions on a source (book, website, etc) that can provide some heatloss calculation education?Thanks for all your replies, too, BTW. It's nice hearing this isn't a completely useless idea.--------------------------
My kingdom for more tools
http://hvaccomputer.com/talkref.asp has a program for the heat loss calculations. Check out http://www.hvac-talk.com/vbb/, but don't ask them DIY questions.
http://www.ashrae.org
Just some thoughts--I'm no expert on heating, that's for sure--if you are thinking about radiant heat in knee walls, could you just get some old cast iron radiators and put them in and then build shelves over them and maybe grills in front of them and insulate well behind them (like with foil-faced foam). Seems like it would be easier than making the panels you describe.
Mass is irrelevant for heat transfer. It can buffer lower load demands, but then you need to manage the mass to prevent over/undershooting.
This type of space is high risk for not keeping up with radiant floor alone for sure. You definitely need a heat load calculation to determine whether floor will do the job and if so, at what temps.
An over-subfloor method can work very well, if you can accomodate any height buildups.
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Hey Danno:
I thought about using "normal" cast iron radiators of some sort, but I like the thermal smoothness inherent in a concrete system - it doesn't need constant heating through it, and become a passive radiator in a few areas due to very large windows in our east/west expoosure.I'll entertain just about any idea at this point though, because it's free to *think* about different options. Once I start doing it however, I want everyone to tell me it's awesome, and never hear a negative comment. =) heh.NRTRob:
Thanks for those points. I definitely need to get my calculations in line before making any decisions. Too many guesses and unknowns. Like I've mentioned, my "ideal" option would be to avoid having to do an underfloor radiant of any sort. Less work and materials. But if the wall-based concrete masses wouldn't be enough, the floor would provide some supplemental help.As for the over-subfloor methods, I have to say I'm pretty leery of pulling up all the old floor and re-laying it - we like that floor a lot as-is...unless there are other methods to achieve what you're talking about...(?)--------------------------
My kingdom for more tools
Realize that "radiant heating" is an idealization. Any hot surface in contact with air will exchange heat with the air itself by natural convection, and with the surrounding space by radiation. Your big flat panel high mass "radiators" will also be convectors. Hot air will rise from them to the ceiling, where it will cool and sink. Without also heating the floor, you're likely to get a cold spot in the middle of your "great room", where all that cold air sinks and crawls back along the floor to get to the bottom of your "convectors". And without insulating the walls, you WILL lose considerable heat out of them also, regardless how you heat the space.
Run the calcs and make sure you know how much heat it will take to keep this lossy space warm- then you may have some tough decisions to make about those stone walls and that ceiling...
Concrete can "smooth out" a "bang bang" system operation... but then you risk over/under shooting. Slamming a bunch of heat into a mass and waiting for it to radiate out, and then slamming it with more is not particularly precise, and not as effiicient as it can be. If you use higher water temps with tight tubing on centers (fast response from cool start) and have a high heat load (small chance of overshoot, since the heat dissapates quickly) then ok, sure.But typically whether you use mass or not, you'd be much better served by striving for as close to constant operation as you can. The ideal in all cases is to always replace exactly what you are losing as you lose it. Anything else is playing "catch up" and must run harder to do it... that said, almost no systems are true constant circ (though there are a few) but you can minimize the "catching up" the system needs to go through for sure... use some sort of reset and a good thermostat with pi/pid logic.If you don't want to pull up the floor, then if you want a heated floor a joist system is the only choice. I would wager you'd be looking at heavy extruded plates though in a cathedral space like this, if floor will in fact do the job by itself.If you want to do wall, we typically just strap the wall with 1x3s and use lightweight plates w/PEX-AL-PEX and reset water temps... no need for concrete panels. But unless you were very conservative with your windows, it's unlikely you'll have the wall space you need, and you'll probably need a more robust radiator.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
so you say mass is irrelivent?heres a question,,i have 2x12s then 1x6 t&g would it be better to run the hose under the house with reflective insulation? or build the floor up to fit the hose .. i do have under access to the house as i am lifting it..coastal oregonthanks
tj
well first you want to do a heat load calculation ;)In general, over floor is always more efficient than underfloor. You can run at lower temps, so you have a lower joist temp to lose heat downward with. However depending on your install method, it may or may not be a big deal. If you're over a cold crawlspace, for example, it is probably a bigger deal.I never rely on reflectivity for insulation. Use real insulation with particular attention to the ends of the joist bays. Reflective insulation is fine, but it's not a replacement for real R value.Mass is irrelevant for heat transfer. If you are storing passive solar gains, it's relevant. It's relevant to how you need to control the system, in many cases. But for heat transfer, it's irrelevant. Heat transfer is based on the overall conductivity of the system, not how dense or massive it is.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
"Mass is irrelevant for heat transfer."
So you disagree with all of those brilliant individuals that have been extolling the magical thermal properties of high mass houses? Constantin will have to "quivel" with you!
Mass is irrelevant to heat transfer in steady state conditions, I should add to my previous statement, you're right.Mass is not irrelevant to heat STORAGE. If your walls store solar gain and buffer temperature swings with their stored energy, then they provide benefit.But they do not reduce steady state heat loss. In the real world, with outdoor temps, you don't see steady state a lot.. but inside, you do want a steady state. So having a mass emitter doesn't help (you want it to be at steady state). Having a conductive emitter does.As for what makes sense for a building SHELL, since you don't have steady state conditions outdoors, the mass can be of benefit in some climates. In others, regular superinsulation is a better choice.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
The combo approach sounds the best, a couple of ideas-ques, In the kitchen a radient conc floor would get you warm as I would think the kit cabs would interrupt your baseboard idea. Or a tile over mud base could work. Would the existing floor handle the added mass? Are lg east-west windows gonna increase your summer ac load? How about a ground water system? How thick would rad. slabs be? I imagine a 2" thich would be the minimum. good luck Barmo