View Full Version : need radiant help
11-14-2007, 11:17 AM
So, even though I am super busy trying to beat winter and get a roof up, I know I have to be ready to get my heating system in place soon.
We are planning on using underfloor radiant heat with Al plates. We want to use a Munchkin Contender Boiler (or something similar) and have indirect tank for domestic hot water.
My engineer that did my energy Calcs came up with a minimum load total of 68,439 BTU per hour. Using Slantfin, I came up with 50,000 BTU's per hour.
We use a woodstove to provide much of our heat. It is rated at 60,000 btu's per hour. We would like to set (and leave) the radiant and a low temp (just comfortable) and then use the wood heat to bring the warmth to where we want it.
I have talked to a few companies and I have received many different answers. They range from using 1/2" pex with no plates to 7/8" pex with plates. Many also quoted me with tankless water heaters, even though I asked to be quoted with Mod/con boiler and indirect DHW.
I am uncertain if my heat load calcs are correct (the seem high). It is a two story house with the master bed and bath in the loft. It has a large vaulted dining and living room. It is 2025 sqft, basically a 28'x44' building with about 2/3rds of second floor a loft. We have a whole bunch of large windows. I am using R21 in the walls, R38 in the ceiling, and R19 in the floor. The second floor will not be heated and I want to use a single zone.
I guess I am looking for some direction. I will be doing the majority of the work by myself. What would you do in my position?
11-14-2007, 04:30 PM
Personally, I would not use PEX ... and just two days ago, a long-time plumber who does a lot of radiant heat in commercial buildings told me some much better things have become available over the past five-or-so years. He had mentioned "mini slant-fins" and had talked about various ways to install things, and he told me to mention his name and go talk with a hydronics engineer at a local plumbing supply house.
Overall, good information and advice seem to be difficult to find, and sales reps seem only interested in what they have on hand.
11-14-2007, 05:36 PM
There is nothing at all wrong with using PEX. It is real common too use 1/2" PEX. For heating, make sure the PEX has an oxygen barrier. How you install it (Al spreader sheets, etc.) depends on what the floor is made of and the covering material and how you mount the tubing. Embed the PEX in concrete or gypcrete and you need nothing else. Gypcrete needs some sort of protective surface. On OSB with wood or something on top spreaders would be appropriate.
I suspect that your operating strategy will cause you problems. You will have to control the radiant heat to compensate for thermal mass and make it stop at a temp less than what you want to maintain with the wood stove. You may have to put some work into making two independent heating systems cooperate well enough to keep comfortable temperatures.
Regarding heat loss calcs; perhaps you should sit down with the engineer and compare the base condition definitions and heating models and work out the correct answer. There are so many assumptions in the process that can make the differences you have seen.
11-14-2007, 07:07 PM
You mentionned lots of big windows - that might be why the load calcs seem high. I'm no expert, but I've played around with some softwares, and I'm always surprised at the impact of windows.
11-28-2007, 08:57 PM
In my opinion (I'm no expert) I would use Onix tubing over the subfloor and then Gypcrete or similar. The thermal mass of the system will be key for that even low temperature you are looking for. For a DIYer, the connections are simple - barbs and hose clamps vs. expensive crimpers or expanders for PEX connections. I'm guessing the supply temperature for a system like that would be less than would be required if you were trying to heat through a wooden subloor. I heat almost 2000 ft/sq AND supply domestic hot water with a 104k BTU oil-fired water heater. I keep my domestic water separate via a heat exchanger. This is only possible of your required supply temp is low enough, otherwise a more expensive boiler is required. If you keep the loop lengths equal, you can further save by not installing balancing valves on the manifold.
There are a lot of resources available for radiant heating. When I was designing my system, I found the Mr. PEX design software heat loss calc. very accurate. It's free for a month:
One interesting thing I learned is that tube size has less to do with the amount of heat tranfer capability of a given length of tube than with the maximum length of a loop due to head loss. Bigger tubing requires bigger pumps and more energy to heat the higher water volume. With the price of oil ever on the rise, efficiency should be top priority.
11-28-2007, 10:01 PM
A couple of points. I am not sure from the post above if you are suggesting bare Gypcrete or not. I do not believe Gypcrete is a suitable wear surface. I would not use it as such. If you are going to leave the embedding surface exposed use some form of concrete. Or tile it. Something. Include crack controls appropriate to the substrate and covering material.
Using the hot water heater temps to run the floor, you need to understand the temperature requirements. DHW temperatures may be too high for the floor; requiring some sort of tempering.
The thermal mass of the floor does tend to allow heat to flow out over the surface. It is more the conductivity of the medium (for example aluminum sheets fitted to the tubing) rather than the mass. Mass will slow the response in both directions and make it more difficult to integrate two control systems.
Depending on building heat loss, water temperature, water flow rate, tubing size, floor surface material, insulation under the floor, rugs, your floor may or may not feel particularly warm. Many combinations of parameters and materials can be implemented to put heat in the floor. Some may work a whole lot better and be more or less comfortable than others. Personally I would not (and did not) wing it. I strongly suggest working the numbers and understanding the whole process before comitting to a solution. It is not rocket science; but it is also not a handyman project without learning what matters and how to do it. Be cautious of internet radiant sales sites. Research the validity of the site. Some are very very bad.
Balancing cocks are not necessary because the various loops are not the same length. They are REQUIRED to balance out the heat loss in the various areas, and that would be the case even if, and probably especially if, the loops were all the same length regardless of the heat requirement for each room. And if all loops are the same length, they had better be configured for the room that would be the coldest, or that room will not heat properly
11-30-2007, 08:16 AM
First off I would take the quotes from the companies that did not quote what you asked for and throw them in the rubbish. If you E mail me I have a simple excel spreadsheet that does a heating load calc. It has no bells and whistles but its fairly accurate and will easily allow you to change design conditions to see how it changes your heating load. Most heating loads are done for worst case and then some scenario. So most of the time you will not need nearly that amount of heat, which is why modulating boilers rule. I have aGB142/24 by buderus and love it. Its capable of 98% efficiency at 100 degree water temperature.
Second. I agree with the others above about how little design changes can make big comfort differences. I would advise anyone to have a professional design the controls and the loop layouts. Then you can do the grunt work of installing the tubes. A radiant floor with some mass can be wonderful but it can also be a nightmare if improperly designed