I would agree on sensing the room temperature for comfort. Though as a scientific experiment, I think some added sensors are OK. The thermal coefficient of concrete is a fairly standard number, along with the skin coeffient for a flat surface of given texture, color, etc. The thermal effects of the materials under the concrete are suspect. It's not know what the moisture content will be, and that will change for some time after construction. That will be an item of interest, along with edge effects. Back to structural engineering. Concrete does shrink as it cures (a chemical reaction, not a drying), for a slab free to move on a frictionless surface, this could easily be 1/4" per 20' length. The friction of underlying material will reduce that slightly, but everyone has seen 1/8" or bigger cracks in concrete every 20'. The cracks occurs when the concrete shrinks, goes into tension, and concrete is weak in tension, and exceeds it elastic limit. Also the concrete does have thermal coefficient of expansion like most substances that can contibute to the tension and cause more cracking. If the concrete can be placed during cold weather (above freezing) then there will be less cracking. Also, since the concrete gains strength with age, if the temperature can be kept down there will be less cracking. Since concrete curing is an exothermic reaction, and can be accelerated, more heat can be generated, and more cracking can happen. For a 4" slab, my experience has been in climates like Northern Ohio, there should be control joints every 17' to allow the concrete to shrink both thermally and shrinkage. The control joints can be saw cut 1/3 the depth of the slab or formed with a bond breaker to the next section. Allowing the heating element or bare sensor cable to cross a control joint is asking for long term issues due to movement. And we haven't addressed issues related to the movement due to loading. Note that different shrinkage can happen top and bottom of the slab, causing curling where the edges lift up, and the slab can actually rock when a vehicle is driven across. Nate Duehr wrote: > Seems to me like the KISS principal should be applied. > > Keep the water temp below a temp that will cause any problems with the > concrete floor. > > Measure the air temp in the rooms above the floor. > > Control zones appropriately. > > Why bother measuring the temperature IN the concrete at all? Isn't that > just overly complex for no good reason? Who cares how hot the concrete gets > if the system is designed from the start not to damage the concrete? > > A sensor (or two or three for redundancy) UNDER the concrete for that one > requirement of seeing if it is allowing too much heat into the ground below > the floor, seems to cover that one... > > Why measure in the concrete itself at all? Just a guarantee that eventually > a sensor will fail, and won't be easy to fix. > > Nate > > -----Original Message----- > From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf Of > Volker Soffel > Sent: Thursday, January 29, 2009 9:43 AM > To: piclist@mit.edu > Subject: RE: [EE] Burying a Temperature Sensor in Concrete > > At 06:20 AM 1/29/2009, you wrote: >> From: "Andre Abelian" >> >> Why do you need to put temperature sensor in concrete ? >> I am just cereus? Suppose you know the temperature ? then >> What are going to do with it? >> >> Thanks >> >> Andre > > The purpose is to measure different heating zones of a solar powered > radiant floor heating system > for a single family home. Area of concrete approx 4200sqft. The > temperature readings are used > to adjust the hot water flow through the different zones and to monitor the > performance of the overall system in general (there will also be > temperature sensors > to monitor ground temperature (below concrete - to verify > effectiveness of insulation > of concrete floor to underlying ground essentially), water > temperature in the entry and return loops > and outside air temperature, as water temperature also needs to be a > function of the > temperature delta between outside and inside temperatures. > > So for the concrete based sensors you are looking probably at > something like 20 sensors > with cable runs of up to 200 ft to get to the central collecting point. > > > best regards > Volker > MicroController Pros Corporation > microcontrollershop.com > The World's Largest Embedded Tool Selection > +1-408-776-7992 Phone > +1-215-243-6071 Fax -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist