Electric UFH
we're about to install electric resistance UFH in a small area (washrooms) ... and haven't bought the sensor or controller to go with the wire ... the idea being that timed intervals should be sufficient - ie: Cortex decides when appropriate (presence, button-press, time of day, whatever) & switches it on for maybe ten or twenty minutes (enough to heat, whilst leaving some margin relative to the don't exceed time) ...
any thoughts, anyone done this, by any chance ??
Typically the floor sensors are thermistor types so should easily connect into a QAI, then you'd need some relay module to switch power to the heating wire (check wattage to choose which relay but if in doubt a DRH, SRH or QRH). In principle you could then set up a heating control loop around the floor temperature using an HVAC object, or you could do it more crudely just using QAI thresholds to control temperature and various inputs gating the operation period.
I had thought about this for our ensuite UFH and thought the easiest approach would be to wire in a relay module after the existing controller.
So you retain the existing controller to read the floor thermostat and maintain the set temperature but set its timer to be always on. The idratek relay then switches the heater output on or off based on Cortex control.
Otherwise you have to interface not only the floor heating element but the thermostat yourself. The standard fitting UFH controller has logic in it specifically designed to manage UFH elements, preventing overheating and the like, so there is some reason to continue to use it.
By retaining it you then have the fallback of reverting to the standard control by setting the idratek relay on and using the controller to set heating periods as normal.
The worry I'd have with your approach, Chris, is that the pad might overheat if it was just switched on without any floor thermostat sensor to check on its local pad temperature. Karam's approach of having a floor sensor, but using Cortex to manage the heat pad would be safer, but then you have the calibration issue of getting the thermistor integrated into Cortex.
Last edited by achapman; 14th October 2011 at 03:19 PM.
yep ... take the points ...
when we bought the kit, the controller & sensor came separately, and doubled the cost ...
so instead, to do without both, my thought was to first measure how the floor temperature rose with time, then allow Cortex to switch it on for only (say) half as much ...
the washroom is a combined shower-room with WC etc & a small laundry area, with WM & drying area, and will be used mainly when the adjoining area is occupied, and we could maybe contrive to have some short warning of that ...
and any heat generated will soon find itself distributed around the house, courtesy of the MVHR system ...
As Achapman has suggested is exactly how we have wired the heating in our Kitchen and plan to for our Hall floor. They are slate so we have them come on an hour or so in the morning before we get up with the times varying depending on working day or weekend.
As far as I am aware the controllers are all the same so you can shop around. As long as they are rated for the load of the element. When we bought the hall kit it actually came with remote control and I was going to investigate using IR to control it but that is a long way down the list on the refurb at the moment. Racing to get house ready for rendering before winter sets in!!
Regards
Allan
thanks again ... hmm, we bought one of these :
http://www.warmup.co.uk/uk/loose-wir...g-system.phtml
going for the loose-wire type because of the floor-standing sanitary-ware & the long-narrow room shape ...
and the choice of control units might therefore be these :
http://www.warmup.co.uk/uk/warmup-un...rmostats.phtml
and since Cortex would do the clever bit, I guess the basic electro-mechanical version would be the one ...
OTOH, using just Cortex with a time-limit might be more reliable, long-term, and would mean no problem in finding a place to site the thermostat, in-line with room zoning requirements (tricky in our case, due to bath & shower etc locations) ...
OTOH, using any one of them would eliminate the need to calibrate ...
Last edited by chris_j_hunter; 15th October 2011 at 03:16 PM.
I have electric UFH, wire type rather than mat type. I installed it in such a way that I have very easy access to the wires, I knocked the back out of the back box and made the hole about 2 cm deeper, just in case I needed to put a relay or something else in there when i eventually get round to automating it.
The controller I have is made by Aube model number TH132-F (the F is for floor temp sensor). Its pretty good so at the moment automating it is on the back burner.
There are 2 things I really like about it is that sometimes rather than coming on at 100% it may only come on at 25% so this saves energy in itself. Also it has an early start function so the on time acts more like time for floor to be at 20 degrees.It looks at yesterday and then decides do I need to come on earlier today or later or the same time as yesterday.
Aube have some models that also have a room temp sensor as well as a floor sensor.
IF YOU CAN'T FIX IT WITH A HAMMER, YOU'VE GOT AN ELECTRICAL PROBLEM.
www.casatech.eu Renovation Spain Blog
looks good ... there seem to be three versions, A & F & AF, with the first being without floor-sensor, which seems odd - no way to avoid overheating, if the element should be undersized for the room-temperature demanded !
PS: actually, that Aube looks even better than I thought - appearance & user interface look good, plus it gets through power cuts (not sure if there's a limit) without needing a battery ...
the smart early start function you mention might save a lot of macro writing in Cortex, but it & the automatic scheduling would operate only to a regular routine, which generally wouldn't work for us ...
it also has a manual mode, however, in which it just maintains temperature, which might give safety & reliability & comfort & efficiency while allowing Cortex to do the when-needed / timing bit ...
that PIA mode you mentioned (varying power according to need) also looks v.good, for better temperature control ... presumably it works in manual mode ?
hmmm, Googling shows some reasonable prices, too ...
on top of which, it can be linked to Cortex via their TH134 - meaning it can be left to manage things on its own, while being open to timely commands from Cortex, albeit the instruction set is quite limited !
http://www.aubetech.com/products/pro...=37&noLangue=2
http://www.aubetech.com/manuel/2/TH132-A-F-AF-230.pdf
http://www.aubetech.com/manuel/2/TH134.pdf
Last edited by chris_j_hunter; 16th October 2011 at 05:18 PM. Reason: added PIA point & TH134 point
I've seen this description of different % on the Aube unit (or maybe some other) and was intrigued by what it might actually mean. By this I mean: to maintain a given floor temperature set point given a particular environment temperature then you will have to feed a particular amount of power into the heating element (which in practice will be delivered by the controller in a pulsed manner - i.e. heating element switched on/off or PWM'd at suitable intervals). So in this context I didn't understand what these different % values mean. Does it effectively mean that the unit sometimes automatically changes the setpoint? For example I set it to 30C but it decides to deliver 25C instead? Any ideas?Originally Posted by toscal
Regarding other posts, yes I think if you were using Cortex to do the control then it would be a good idea to have a floor sensor. These are normally thermistor types and their parameters can usually be easily obtained and plugged into the existing Cortex analogue input thermistor mapping equation. I didn't think they were very costly but a while since I looked and they tend to come as part of a package as it were. For just floor control I think the accuracy required is not really exacting so I don't think you'd need to calibrate the sensor further beyond its stated % accuracy level.
If you are using the electric floor heating as the main heat source for room temperature control then for the room temperature loop you'd be better off using an IDRATEK temperature sensor. In other words you would have two control loops - one doing the floor and making sure that could deliver as much power as possible when required but without overheating (so that you could respond to changes in room set points as quickly as possible) and another doing the room temeperature control which needs much better accuracy and resolution sensing.
I think switching on the heating element for a given period of time (or even pulse width modulating it at a given ratio) without temperature feedback is plausible but at best is only going to be useful for taking the chill out of the floor on a morning for example and not for maintaining a steady floor temperature in the face of changing conditions. To reduce the overheating risk you will probably end up catering for the worst case scenario meaning that not much of the chill will end up being taken out on cold days. But that's just my opinion - I haven't tried to find out what you could actually get away with in practice.
Posting Permissions
Reply With Quote