“sub-floor radiant heat” Heh. Me too! Elevated house in Australian arid area. Steel stilts about 1.5m long, steel beams, and above that steel tray with 100mm concrete thickness. No embedded insulation. No hindrance to airflow under house. Not my design! During heatwave our air conditioners spend all their effort pumping out the heat that arrived 2.41 hours ago (by calculation) on the other side of the stack-up. Thermal camera shows roughly where the beams are.
1 Like
LOL!
I’m in southern California. Everything’s wood here… I added the subfloor radiant between floor joists myself, stapled up against the subfloor. Then added insulation below. I’m trying to figure out how to do the controls… Right now I have dumb bang-bang on/off with a temp threshold. I tried to do PWM (with a period of several minutes based on the time it takes the water to make it through the pipes) but that causes too much noise due to heat expansion/contraction 
. I think I need some PID but I don’t know how to approach that given how long it takes to try anything out. Also, lots of windows and older house, so wind has a huge impact on the heating requirements and especially on any predictive formula (e.g. “reach temperature T by 8AM”).
[Oops, I hope Gabe is OK with our brief hijack of his build log thread…]
1 Like
An example is here:
A good practical guide:
Your system is slow. But if you follow the tuning method, you should be able to tune it in 3-4 cycles. Systems are different, so you should tune it back so it is not aggressive (low D part), so your control will work on all systems
1 Like
Maybe I should not be procrastinating, but I found a link with more details. Even a self-tuning algoritm:
1 Like