I am looking at adding a hydronic snow melt system to concrete pavement. Any recommendations for design guidance? Typically I would design the pavement using the PCA, Thickness Design for Concrete Highway and Street Pavements.

Any guidance is greatly appreciated!

You might check with the guys who do in-floor radiant heating for houses.

The radiant installers are great at providing details/feedback in regards to the snow melt, but have not provided any industry guidance in regards to reinforcing/jointing requirements for the concrete slab design itself.

I'm in Hawaii, so I have no experience with these systems, but as a guy responsible for concrete mix design and quality control, I'd be concerned about the potential for curling due to temperature differential between the bottom and top of the slab.

We frequently see a 40-50 degree top to bottom temperature differential in a concrete pavement during the summertime. Do you expect something more than that?

Tom...I don't know if the winter temperature gradient would be greater than the summertime gradient because a number of variables are in play. I expect there would be a large difference between what temperature must the bottom of the pavement be heated to thaw the surface when them ambient temperature is 25° (small gradient) vs. if the ambient temperature is -40° (large gradient), and the slab thickness would also make a significant difference. Other factors that contribute to curling may accentuate the problem in summer or winter, or they may moderate the effect of temperature alone. If your winters are cold enough, the winter differentials certainly could be greater than summer.

Also, the summertime temperature differential's effect on the pavement may not be a good predictor of the same amount of differential in the winter because the any curling would be in the opposite direction. A pavement warm on the bottom, cold on top would tend to curl the edges upward (away from edge base support), but the summertime differential would create forces in the opposite direction (toward edge support, away from support in center area). I would guess that the wintertime type curling would be more likely to produce longitudinal cracking due to reduced support under loading than would an equal differential of summertime type curling.

Hoakapohaku,

I doubt putting the heating elements in the bottom of the slab would be effective. Around here concrete pavement starts at a minimum of 8" and goes up to 16" or so. I was thinking the heating elements would be more like 2-4" from the upper surface.

After I wrote that last post, the thought came to mind, "what makes you think the coils will be at the bottom of the slab?" Must have been some website I looked at. Or maybe it's because I imagine that coils embedded within a concrete pavement subject to continual loading and unloading just sounds like trouble.

May I ask what is the size and intended purpose of the pavement?

I can't give much guidance from the structural slab design, but...

I would give a lot of thought to special designs for the non-heated areas adjacent to the heated section. From my experience, this is where you will get the most failures. The extreme temperature difference (particularly in the basecourse and subgrade) can cause lots of uneven settling/heaving during the winter and spring. You also can get issues with the melted water from the heated areas draining onto the surface of the colder sections - this water gets into seams, etc. and freezes causing more damage.

For instance, it's typical to have a heated sidewalk adjacent to an unheated curb and gutter. In this instance, it's best to try to extend your heating elements into the curb and gutter or even into the c&g basecourse. You should also consider carrying out your basecourse past the extent of the concrete and slowly decreasing your basecourse thickness as you go out further. This will minimize the direct contact from heated basecourse material to cold subgrade and therefore minimize differential settlement between the two materials.