I am studying an electric radiant floor heating, and studying precisely its regulation.
I need to make the calculations with a very short timestep of one minute.
I modelized my floor with an active layer. I did'nt look very well to the flow specifications, segmentation, etc, since I put a flowrate equals to zero, and put my regulator's output (the power delivered by my electrical wires) as a gain to this active layer.
When I looked to the results, I saw that the regulation seemed to have a strange behaviour, with a type of hourly pattern. I saw the same thing to my surface temperature. I tried to increase the timebase in trnbuild, and my hourly "pattern" became a 2 hours pattern. (See pictures attached)
I understand that the behaviour of a radiant floor is clearly linked with the timebase used. And, we can't decrease this timebase so much for heavywalls.
So my questions are:
Am I right? Is it a consequence of the function transfers, that make the surface temperature vary with hourly steps, even if the timestep is smaller?
Can someone explains me better this phenomenon?
Is it possible to eliminate this behaviour?
Are the calculations made still correct, or the energy balance is wrong with this timestep?
Thanks in advance.
Regards
Thibaut Vitte
