Replies

1. 
   johnnyd | Apr 02, 2003 08:34pm | #1

   Have you considered a low mass system?  Quite often better in a retrofit like this over floor framing, no wieght to worry about and much more reactive.

   Good advice from this forum:

   http://www.radiantpanelassociation.org/i4a/pages/index.cfm?pageid=1

   Here's another one:

   http://www.taunton.com/finehomebuilding/pages/h00028.asp

    

1. 
   Accelar | Apr 02, 2003 09:36pm | #2

   By low mass do you mean the gypsum system referred to in the article ?

   I am based in Toronto, and cannot find a local contractor for the mix shown.  Are you aware of any other similar products ?Gavin Pitchford

    

1. 
   johnnyd | Apr 02, 2003 10:03pm | #3

   Low mass would be using:

   PEX hung in the joist spaces underneath the subfloor with insulation below. OK if it's easy to get under there, but you have to heat the airspace between the bottom of the subfloor and the top of the insulation.  Probably requires 180 degree water. No floor build up.

   Variation of above except the PEX is snapped into aluminum plates that are fastened to the bottom of the sub floor. That way you get direct transfer of the heat and lower water temperature. No floor build up.

   "sandwich" method ON TOP of subfloor using 3/4 " OSB or lumber sleepers however wide your PEX spacing will be.  Put aluminum plates encasing the PEX either underneath or over the sleepers.  Best heat transfer, lowest water temperature, but there will be floor build up, which may cause a problem with doors, cabinets, and threshholds.

   There is a product called "Warmboard" that basically uses a pex/aluminum/wood sandwich that also provides the subfloor.  Mostly used in new construction where labor savings in applying the subfloor and heating system in one step justify the cost of the product. But does speak to the viability of low mass sandwich versus adding gypcrete to a framed floor.

   I think the FHB article goes into explaining the various methods in detail.  Also, if you have time, the Radiant Panel Association link has lots of good information in the threads that is written by pros (which I am NOT...just have done alot of research in the area)

   Main point is, there are viable options for effective radiant floor heat that do NOT use a slab or gypcrete.  Some of the companies offering products for low mass floor heat (aluminum plates, pex, pumps, manifolds etc) now advertise  in FHB.

   Another thing that will come up is the boiler that currently supplies the radiators in the rest of the house.  Is it a pressurized system? What is the fluid temperature? You may have to add a mixing valve to get the water temp right for the in-floor, if you will be using the same boiler.

1. 
   Accelar | Apr 02, 2003 10:40pm | #4

   Thanks for all the suggestions - but a few problems

   I am replacing the subfloor anyway - and it is not accessible in this area from below as there is a maze of radiator feeds underneath.  I expect to add a seperate controller to the boiler - brand new Veissmann bought with using RFH in a few places in mind.   Cannot accept the build up that the warmboard would produce - must be level at the thresholds - weird design but there are 3.   

   The boiler guy suggested the countersink method, but as I looked more closely I was concerned about the added weight and the possibility that a temp induced shift in the concrete mass might kink the pex as it went through the joists, or develop cracks in the subfloor under the tiles - unless I separated the tiles from the subfloor and let it sit independent of the subfloor mass.

   Gavin Pitchford

    

1. 
   johnnyd | Apr 02, 2003 11:00pm | #5

   Why not just use the same countersunk method except use plates and plywood as the heat transfer mechanism instead of concrete?  Set the sandwich between the joists on ledgers, and put the subfloor and tile over? If the tiles are separated from the heat transfer, whether slab or wood, they won't act as good conductors.  Remember that the floor surface itself shouldn't be above 80 degrees, and I think the goal is to balance the flow, tube spacing, and water temperature to heat the room comfortably with the lowest possible fluid temperature, avoiding the contraction, expansion, and comfort problems that a high fluid temperature will induce.  You might want to pose this same question on the RPA board. 

2. 
   Accelar | Apr 03, 2003 01:09am | #6

   I will post it on the other board and see the responses - but the problem is that plywood is a better insulator than conductor, and I can't fit the pipe to the underside before I install the subfloor and then install the subfloor unless I cut the tops of the joists which I really don't want to do - and the lack of access from below precludes running the pex afterwards.

   What I was really hoping for was somone who could tell me what concrete product would be best - and whether the load would exceed safe limits for the 2X10sGavin Pitchford