Radiant Tube Sizing

Discussion in 'HVAC Heating & Cooling' started by Drewski123, Sep 16, 2013.

  1. Drewski123

    Drewski123 New Member

    Messages:
    47
    Location:
    Seattle
    My contractor has recommended using 3/8" tubing for hydronics that are going to be installed under the floor along with the Wirsbo Joist Track heat plates. My initial plan was going with ½†tubing. He says that there is no difference in heat transfer between 3/8†and 1/2†tubing with using the plates. There is going to be 1â€-2†space between the plates and the insulation. The tubes will be installed 8†on center. I live in Seattle with winter temperatures ranging from 30-45, with majority of 42 degree days. The radiant heat is going to be the only heat in the house. Here is some more info about the house and the project summary:
    • Heating area is about 1,300 sq.f
    • 3 Zones (2 zones for the 1,300 Sq.ft; and 1 zone prepped for future finished basement that will have radiators installed
    • Total Load: 20,600 BTU/H
    • Total Radiant Load: 13,821 BTU/H
    • Total Supplemental Load: 6,779 BTU/H
    • Boiler: Navien CH-240ASME; Min: 20,000 BTU/H; Max199,900 BTU/H; 4.5GPM; 91% AFUE
    Also, the contractor has recommended using ½†tubing hanged with 8†wide heat plates, 16†on center. He said this is something new that supposed to be good too, but less costly.
    Based on the project summary, which option would you recommend?
    Thank you for your help.
  2. Tom Sawyer

    Tom Sawyer In the Trades

    Messages:
    3,230
    Location:
    Maine
    I never use 3/8" the runs have to bemuch shorter in order to maintain decent flow. 1/2" with joist track will perform better
  3. Drewski123

    Drewski123 New Member

    Messages:
    47
    Location:
    Seattle
    Thank you. Would you consider using 1/2" tubes with the Ultra-Fin plates 16" on center instead of 1/2" tubes with Wirsbo Joist Track heat plates 8" on center?
  4. Tom Sawyer

    Tom Sawyer In the Trades

    Messages:
    3,230
    Location:
    Maine
    Depends on the heat lad and the available area to install.
  5. Drewski123

    Drewski123 New Member

    Messages:
    47
    Location:
    Seattle
    Tom,
    Here are the calculated heat loads. Basement was not included in this calculation, but I am prepping the system to accomodate future baseboard installation. In terms of area, there there is enough space to accomodate 2 runs in each cavity.

    Project.jpg
    Last edited: Sep 16, 2013
  6. Dana

    Dana In the trades

    Messages:
    2,924
    Location:
    01609
    Ultra-Fins are convectors, and require higher water temps, and the response time is on the sluggish side to boot. The heat is drawn from the tubing into the Ultra Fin via conduction, but to get from the Ultra Fin to the floor is only via heating up the air surrounding the Ultra Fin, with an air convection loop in the joist bay cavity. This takes a fairly big temperature difference to deliver the heat at a reasonable rate, but it also has the down-side of higher losses through the joists to the space below, and potential for losses to thermal bypass air movement laterally in the air cavity between the insulation and the sub-floor.

    JoistTrak plates do the heat transfer primarily via conduction, which gets heat out of the tubing and into the floor at a faster rate at any given temperature difference between water & floor, and you can snug up the insulation to the Joist Trak, limiting the thermal bypass and joist conduction losses.

    You're looking at about 12 BTU/hr per square foot of floor, which through 3/4" plywood + 3/4" oak (R1.7) a you can get with ~95-100F water in the JoistTrak (depending on the R-value of the subfloor + finish floor + floor coverings). See the nomograph on the last page of this document. But takes 120F or more to get the same BTU/foot-hour with the Ultra-Fin. That's a 20-25F higher operating temp, which has efficiency consequences. If your floors are thicker or more insulating than the 3/4" plywood + 3/4" oak paradigm case, the diffferences in water temperature requirments grows proportionally- if your flooring stackup is more like R2.5, you'd need 120-125F water with Joist Trak, but out of condensing range with Ultra Fin.

    The apps where Ultra Fin might make more sense is with non-condensing boiler, and BTU per square foot requirements would be over the 180F specified operating temp limit of PEX using suspended tube alone, and it would improve the response time of suspended tube.

    It goes in quicker & cheaper than extruded plate. But if you wanted to cheap out, some of the better sheet-metal plate systems can deliver 12BTU/ft and higher at condensing temps.

    For less money and hassle you can get pretty good comfort out of low-temperature panel radiators, and run it at a single water temp high enough that this micro-zoned heating system won't short-cycle the boiler into an early grave. At 130F out/115-120F return a single Biasi B-24.71 delivers about a third of the min-mod output of your oversized boiler, and has enough thermal mass to keep the burn times almost reasonable. The fact that your total radiant load is significantly lower than the boiler's min-fire output is a problem when you cut it up into even smaller zones, as discussed on your other thread.

    There's no need for the 240ASME- the smallest in the line 180ASME is MORE than enough for both your heating and hot water loads, at Seattle's worst-case incoming water temps. I'm using a non-condensing tankless with only ~150KBTU/hr max-fire (about the output of the 180ASME) as a combi heating boiler (with a buffering external heat exchanger) in my house with nearly 2x the space, in a lodation 20F colder than yours, and a heat load of nearly 35KBTU/hr, and I could still take endless showers without the tankless breaking 120KBTU/hr firing rate. In practice, with a drainwater heat exchanger and the 48 gallon buffer it never breaks 60,000BTU/hr output (measured), with all zones calling for heat and the teenager stuck in endless shower mode. The 180ASME is already overkill in your application unless you absolutely need to run two full-flow showers simultaneously on the coldest day of the year with all zones calling for heat simultaneously. The 240ASME is just more overkill at best buying you another ~1gpm of hot water performance in mid-winter.
  7. Dana

    Dana In the trades

    Messages:
    2,924
    Location:
    01609
    If I'm reading the thing right you're going to need up to ~23BTU/ft on some of those rooms, which is what will determine your operating temperature and outdoor reset curve need to be. He's estimating that JoistTrak you can get there with 110F water (implying a ceramic tile floor throughout or something?) which is very a reasonable condensing temp. With Ultra-Fin you'd be somewhat outside of the condensing temperature range, getting maybe 86-87% efficiency.

    None of it solves the short-cycling on zone calls issue though.
  8. Drewski123

    Drewski123 New Member

    Messages:
    47
    Location:
    Seattle
    Thanks Dana.
    What would you recommend in order to minimize the chances of possible short-cycling in my situation?
  9. Tom Sawyer

    Tom Sawyer In the Trades

    Messages:
    3,230
    Location:
    Maine
    I'd put the entire thing on a single zone with a floor temperature sensor instead of a thermostat.
  10. Dana

    Dana In the trades

    Messages:
    2,924
    Location:
    01609
    With the whole floor on one zone the output of the radiation is more than half the output of the combi-boiler's min-fire output, but still ~75% of it's min-fire output, which isn't bad. It'll never modulate on heating calls, only cycle, but there's enough thermal mass that a 25% undersizing of the radiation to the boiler output isn't the disaster that chopping it up into 3 micro zones delivering less than 1/4 the boiler output is. Cutting it into two equal 6900 BTU/hr zones would probably work if you had a smaller boiler with a 15K min-fire output, but do the math on the thermal mass of the water in the tubing against the hysteresis of the boiler output temp around it's setpoint.

    Going with floor thermostat-only approach delivers a fixed output to the radiation, which will lead to overheating during milder weather or undershooting when it's cold out. While combined floor & room T-stats exist, and would provide somewhat higher comfort & faster response, that's something to consider only if you're willing to take the time set it up and tweak it for max comfort. It's more appropriate when using high mass radiation like slabs that have a significant thermal time lag- much longer than a JoistTrak & PEX setup.

    In some systems it's possible to fine-tune the outdoor reset curve and run it constant flow without thermostats, letting the outdoor temp determine the floor temp & output, but those take a lot of tweaking to get right, and outdoor temp is a fairly crude model of heat load- solar gains, electrical loads, and wind-washing/infiltration have significant second order effects on instantaneous heat load. Even having 5 people over for dinner would deliver a real fraction of your design-day heat load in body-heat, and sun shining on south facing windows during periods only slightly clearer than the famous Seattle "bright cloud" days can more than carry the entire mid-day load on a 40F day.

    Bottom line, room thermostats probably are going to be necessary here.
  11. Drewski123

    Drewski123 New Member

    Messages:
    47
    Location:
    Seattle
    I've started planning each loop layout for the hydronic heating and was wondering if the return section of the orange loop can be used and installed in one of the bottom yellow plates. The bottom two yellow plates are under a closet and partially in the hallway. I was wondering of there still be enough heat to emit toward that area. The entire loop is within 200 feet using 3/8" tubing with joist track plates. I would like to use the return sections in my entire project as much as possible in order to use the space and heat areas that have a small amount of heat loss comparing to standard rooms. Please, advise.

    House Joists 2a.jpg
  12. Raspy

    Raspy New Member

    Messages:
    11
    Location:
    Northern Nevada
    You are on the right track with your tube layout. Down and back in each bay works well. But you don't need the fins. Just run the tubing and mount it to the side of the joists near the underside of the floor. Then insulate and leave a space above the insulation where the tubes are.

    Never use 3/8" tube because the runs have to be shorter. 3/8 is typically 200' max and 1/2 is typically 300' max. Remember, the size measurement is the ID, not the OD, so don't confuse 3/8 with 1/2.

    Fins add cost and possibly noise, and mounting them to the underside of the floor makes the tube vulnerable for flooring nails or even carpet strip nails.
  13. Drewski123

    Drewski123 New Member

    Messages:
    47
    Location:
    Seattle
    Thanks Raspy. I've decided on joist track plates with 3/8" tubing, all within 200'. Another questions I have is about manifolds. I will have 8 loops on two zones. Which manifolds would you prefer, Uponor TruFLOW, or the EP ones. Also, I was considering actuators, but I am not sure if they are necessary. The water supply temperature between the two will differ slightly, but not much. For the controls, I am planning to use the wireless Climate Control from Uponor that includes a C-55 (base unit), an I-75 interface, as well as two T-75 thermostats. What's your opinion on the control systems? Is there anything better then that? I do not mind spending more money for something better. Would you also consider a floor sensor. I am definitely installing an outdoor temperature sensor.

    Thanks for your help.
Similar Threads: Radiant Tube
Forum Title Date
HVAC Heating & Cooling Floor joist radiant heat system Jul 13, 2014
HVAC Heating & Cooling Hydronic radiant floor help please Jul 7, 2014
HVAC Heating & Cooling Forced air in bathroom with radiant floor? Mar 13, 2014
HVAC Heating & Cooling Radiant Heat Under Bathtub Aug 13, 2013
HVAC Heating & Cooling Check valve for open-direct radiant heating system May 14, 2013

Share This Page