GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X 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

Radiant Floor and Heat Pump Capacity

rtrask | Posted in Plans Review on

How many radiant floor runs on a single manifold?

I have been installing radiant tubing in the floor joists of a modular home that we built over a full basement.   In the concrete floor of the basement we installed the tubing on 12″ centers ~300 ft runs.  It works great.  I am installing 1/2″ pex in the floor joists with aluminum fins 8″ centers and approximately 300 ft runs.  My concern is that it will require 12 runs to heat the main floor.  While I can buy a 12 run manifold, I wonder if I will either need larger circulation pumps or to split it into two zones.   Are there any guidelines available to match pump capacity to the total run length?

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.

Replies

  1. Expert Member
    DCcontrarian | | #1

    > Are there any guidelines available to match pump capacity to the total run length?

    Well, BTU/hour= gallons per minute times temperature change times 500.

    With 12" PEX spacing, each foot of tubing covers one square foot of floor, so a 300' loop is 300 square feet. With the surface of the floor 10F above room temperature you can get 20 BTU/hr per square foot, or 6000 BTU/hr per loop. You typically shoot for about a 10F drop -- if it's too much you end up with inconsistent floor temperature -- which works out to 1.2 GPM.

    With 8" spacing you cover 200 square feet per run, with the same 20 BTU/sf and 10F drop you need 0.8 GPM and get 4,000 BTU/hr.

    I'll add that to get 20 BTU/sf with 12" spacing you have to have a highly conductive surface under the floor.

    At 1.2 GPM PEX drops about 2 PSI per hundred feet, or 6 PSI over 300 feet.

    If you're using a manifold, each loop would have the same 6 PSI, so all the loops would run at the same GPM if they're the same length. That shouldn't be a big deal, the 8" loops will overperform a bit.

    You'd sum the GPM for the loops. So for 12 runs you'd need 14.4 GPM @ 6 PSI. So probably a 1-1/2" pipe to the manifold.

    The question I would have is, are you really going to run all 12 loops at once on the main floor? At 4,000 BTU/hr per loop that's 48,000 BTU/hr, that's a lot for one floor.

  2. gusfhb | | #2

    Since they have enough tubing to heat the space they can play with water temp to make up for floor coverings
    FOr a concrete floor running a 'counterflow spiral' as I recall helps prevent temp banding.

    The real problem in a modern well insulated house is having the water temp low enough to avoid overdriving .
    So then you want to limit the amount of tubing so you can raise the water temp soas to feel the warm floors.
    I love radiant[as opposed to some here] but it is not that useful in well insulated houses

  3. rtrask | | #3

    Thanks for your thoughtful replies.
    I paid for the upgraded insulation. We have R25 in the walls and R50 in the ceiling. The basement forms were ICF. and without any additional insulation are just under R25. I have read the articles that claim modern well insulated homes are poor candidates for radiant floor systems. I have resigned myself that the floors may not be warm to the touch. The concern about thermal gain yada, yada yada... I still think I made the right decision. We are off grid, we live in a high cold mountain valley we are in the time of year where night time lows are in the teens and highs are around 30, interspersed with some real cold spells. The idea that I would be reliant on an electric power source for heat with short daylight hours and snow covered solar panels is not attractive. I could go on, but that is not what this post is about.

    Right now we are heating the main floor off of the basement zone. It is about 80 degrees in the basement and about 66 on the main floor. I need to chew on the information that DCcontrarian provided. I only have 1" pipes coming off of my boiler and the manifold I am looking at has a 1" intake. As I feared, the pump I have on the distribution panel is insufficient for that volume of water. If I break the main floor into multiple zones, I am not sure how I would do that. I will have to reconfigure the distribution panel for 3 zones and figure out zones that make sense with the layout of the house which may not be easy / will require some compromise. Since the house is well insulated maybe heating the great room - living room - kitchen as one zone and counting on leaving doors open is sufficient during the day. This is going to require some thought.

    gusfhb -- can you provide more information on overdriving?

  4. gusfhb | | #4

    Just meaning if you have too much heating capacity the system will go over setpoint.

    You do not need to zone excessively. The manifolds will have valves on every loop. IF a room runs cold do to higher loss in that room [say a bigger window etc] you throttle down the other loops until the 'cold room' is warmer.
    When I did the system in my former house some 20 years ago, I ended up with two upstairs manifolds for 3/8 tubing in a wood floor on 8 inch centers, about 1300 sq ft . It was a primary secondary loop with the temperature adjustable for each zone, but not a outdoor reset variable temperature setup.
    All of this stuff is math, and you should have all that math figured out before you put tube in the floor.
    I got a lot of help from heatinghelp.com, and the install contractor from that site.

  5. Expert Member
    DCcontrarian | | #5

    >All of this stuff is math, and you should have all that math figured out before you put tube in the floor.

    Yeah, I probably wasn't doing rtrask any favors with my post #1 without more background.

    Here's the key thing to know about a heated floor: the heat output is determined by the floor surface temperature.

    There's a pretty narrow band of temperatures that are acceptable for a heated floor, anything below about 75F and you can't feel it doing anything, and anything above 90F and it starts getting uncomfortable to walk on. The ideal floor has some warming any time heat is needed, and when it's at its maximum puts out enough heat to keep the house warm on the coldest days you see. So theoretically you'd want a control mechanism that keeps the floor at say 75F when it's 55F out, and goes to 90F when it's as cold as it gets where you are. If 90F doesn't actually put out enough heat on the coldest days, that's not terrible, you just have to have some sort of supplemental heat that kicks on when the floor can't keep up. At an extreme you could heat the floor to 80F or so when it's 55F, and leave it there and use supplemental heat at any temperature below.

    So controlling the heat is all about controlling the floor temperature. There are basically three ways of doing that: modulating the water temperature, modulating the water flow rate, and turning the flow on and off. They can be used in combination. I'm not a big fan of modulating the flow rate, what I like to do is modulate the water temperature and fine tune with a thermostat that turns the zone on and off.

    Everything else -- the boiler, circulator, piping, valving, etc. -- is all about being able to control the floor temperature, and get heat into the floor at the rate you need. Because all of the heat flows are the same -- whatever the heat flow out of the floor into the room is, it has to be the same into the floor from the tubing, which has to be the same as the heat carried into the tubing by the water, which has to be the same as the heat put into the water by the boiler (or whatever your heat source is).

    So the design is a reductive process. You figure out how much heat the floor has to put out. You figure out how hot it has to be to put out that much heat. You figure out how closely the tubing has to be spaced and with what water temperature to put out that much heat. You figure out what flow rate and temperature drop you have to have to put out that much heat, and what kind of circulator you need to achieve that flow rate. And you figure out what kind of boiler you need to put out that much heat.

    So how do you figure out how much heat you need to put out? This article: https://www.greenbuildingadvisor.com/article/replacing-a-furnace-or-boiler shows an easy way of estimating to pretty good accuracy your heating load, based on historical energy usage. That gives a number for the whole house. What you really want is a room by room calculation. There's a process for doing that, it's called a Manual J. If you know the whole-house number you can sanity-check it by seeing if the rooms add up to the same number.

    The rule of thumb is that a heated floor puts out 2 BTU/hour per square foot for every degree difference between the surface temperature and the room temperature. So if a room is at 70F and the floor is at 85F, that's 15F in difference and the floor would be putting out 30 BTU/hr per square foot.

    The hardest part about designing a heated floor is estimating what water temperature will give the desired floor temperature. Usually what you end up doing is making your best guess, and having enough leeway to adjust the water temperature once it's running and you've taken measurements.

    1. Expert Member
      DCcontrarian | | #6

      Bump.

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |