We had intended to bury a radiant piping system in 2 inches of a concrete screed.

This system in on a first floor, which is supported by joists.

The structural Engineer has indicated that the loading of 2 inches of a concrete screed would impose too much of a load, so he will not allow it.

My question is there any other material we can bury the piping in, and will not impose an excessive loading and also have a good retention of heat.

Would sand be a good substitue? it has a lower density and a fairly good specific heat?

All comments and advice appreciated.

why is mass necessary?

Not a fan of sand. It's not a particularly good conductor. It can work, but I would rather use aluminized products for better conductivity, or gypcrete if that lowers the load enough to use it.

Cement based screeds are normally used in the UK

Certainly I would use a gypcrete if that lowers the load, what is the density, specific heat  and composition of Gypcrete?

What are the aluminized products available?

Regards

I'm not sure about those characteristics of gypcrete. I'm not sure why heat capacity (density * specific heat) is really relevant here either; are you attempting to engineer a heat storage device, or a heat delivery device? If you're trying to engineer a delivery device, storage is more or less irrelevant and conductivity is a key element. Some heat storage isn't necessarily a horrible thing, but it's not the primary goal of selecting a radiant emitter.

I have no idea what brands are available in the UK, sorry, but there are a lot of products with aluminum as a layer available for radiant floor installations.

Why 2" instead of 1.5 inches which would decrease your dead load? Most floor systems are over engineered and would easily hold up to the additional dead load of 18.5# per sq ft at 1.5 inches of normal concrete or a light weight if available. I poured a 12" I joist 10 years ago with a 15' span this way and it meet all necessary requirements.

Dave

Rob, the radiant floor installation is heated by a geothermal heat pump. This heat pump runs on half price electrical current during the night.

So the covering over the pipes need to have a good Specific heat value, so that as much heat as possible can be stored during the night and released during the day. The material has to be light weight, so it does not impose excessive loading. What is your best solution?

I like where this discussion is going. 'why is mass necessary?' Easy: It increases the size of the 'radiator'. What were old 'radiators' made of? Heavy Iron/Steel. What are current baseboard 'radiators' made of? Heavy Steel. Hmmm, where is the mass? The water, combined with the Steel!

Does a below floor system('Staple Up') have a 'radiator'? Nope. Wood is an insulator. The only time that wood is radiating heat is when it's burning.

So, I find it interesting that there are those that think 'Radiant' heat can exist efficiently without a 'radiator'.

I also find it interesing that people think that Radiant Heat should be used as an 'on demand' system. It's not. Any that think so, are trying to degrade Radiant into a forced air or electric baseboard system. And, why would anyone pay a premium price for a system whose performance is no better than a less expensive alternative?

The name says it all: Radiant - It must have a 'Radiator'!

First, to 410a, then yes, you do need specific heat. You are charging a mass to radiant later, or "time shifting" your output. I would recommend a tank of water, which makes a great storage mechanism and is easy to charge without having to worry about overheating the space.

Second, to PC, you're mixing things up. Mass is mass. Mass is not "size", nor it is important to being a "radiator". Anything with appropriate emissivity and a higher temperature than the objects/space around it will radiate heat. The floor itself, in any radiant heating system, is by definition a "radiator". Otherwise, if you put wood over the top of your mass, how would all that radiant heat heat the room? The wood, tile, or carpet still has to radiate it or it's not radiant.

Furthermore, the wood, tile or carpet you have on the floor surface doesn't know, nor does it care how much mass is mounted underneath it. If the top surface of the wood is X temperature, it will radiate Y heat, regardless of how that heat got into the wood in the first place. You could pump that heat in via aluminum plates or via concrete under mass or, indeed, even by suspended tube in the joists and the output would be the same as long as the amount of heat imparted to the wood is the same.

What DETERMINES how fast that heat is imparted to the wood is the conductivity of the path between water and finish floor. As it so happens, concrete has fairly good conductivity, so it works at fairly low water temperatures. This has nothing to do with its mass. Aluminum has excellent conductivity as well, so it's a good choice too. Suspended tubes, however, don't even have a conductive path, so their output is much weaker. Likewise, sand is not a very good conductor (lots of trapped air) and so its output is weaker as well. "output is weaker" means "you get less output for the same water temperature".

Nothing mentioned so far is helped by mass at all. Mass does one thing, and one thing only; it STORES AND ABSORBS heat. This can have a "bufferring" effect, which in some cases can be of benefit (such as to 410a and his "time shifted charge" system) but has downsides as well (temperature swings in some cases, slower response).

However, you do not have to have that mass at the emitter if a buffer effect is useful to you. You could instead have a tank of water, which provides exactly the same buffering benefit, without effecting the performance of your emitter. Now that doesn't always make sense either and it's not always necessary, but if mass is actually of some benefit to the particular system you are considering (usually, to avoid short-cycling a heat source), then you don't have to have that mass at the emitter is all I'm saying.

And in no way, shape, or form is mass always a requirement for a high performance radiant system. Conductivity is king, first and foremost. an appropriate amount of mass is useful during low load conditions (relative to heat source output) for some heat sources, on some systems, or to "time shift" your demand. But that's it.

410a, for pure heat storage you can't beat water, except with maybe some exotic products like molten sodium. I was looking at a heat storage system for the radiant system for the house I'm getting ready to build. I considered a tank full of rocks in a water bath. After looking up the specific heat of rocks and water both, it turns out that on a volume basis water will hold about 3 or 4 times as much heat as will rocks. Concrete is similar to rock in density and specific heat. If you need to store heat take a hard look at using a water tank to do so.

If you need me too I can look for the data and calcs I did. I don't have them right at hand at the moment.

Everyone is right, but I think it misses the original point. If one is using the radiant heat during the day and want some kind of storage at the same time than mass is the way to go. If you have the storage capcity available you could store the extra heat receieved during the day in the water and then expend it during the night. In the end it is a balancing act of how much storage in mass as in concrete, stone, etc. and maybe a source such as water. Having less of one means more of the other. Not enough of either and then you are talkng a backup source such as electric/gas/etc.

Its a numbers game. Good discussion. I am learning a lot from these forums.

good thread.some brainpower here.