Ivar Uni-MIx UFH Blending Valve

[Rob]

Anyone using one of these on their UFH Manifold?

If so, are you controlling the blending valve via Loxone?

Cheers

Rob

[Duncan]

if its this one:

then you can unsdrew and replace the white thermostatic control with a standard trv type 0-10v actuator such as loxones own, then add a temperature sensor eg 1-wire to the hot side of the manifold and let loxone take over easily

ive done exactly that with a different make (wundafloor) but the same type of thermostatic trv controller and the loxone control works perfectly

[Rob]

Yep, thats the one.

Thanks for the confirmation Duncan, I assumed it would work like that but it just seemed too easy..!!

Rob

[Duncan]

you will need to use the mixing valve controller linked to your 0-10v actuator, and you will probably need to fiddle with various parameters to get things working optimally

the values of Ki and Kp will need virtual inputs attached so you can change them in real time

subtract the actual manifold temp from the target and display using a virtual state with some logging so you can create a graph of the difference

i would also suggest using virtual inputs for the offset (N), gain (S) and load gain (G) of each ITC 

ideally you want the water temperature to be enough to heat up your ufh slowly over  time without any significant overshoot, and the water temperature to go higher when the outside is colder so the heatup time stays roughly constant

the recommended values for Ki and Kp are usually too large

i found 

for hard floors my offset is around -3, gain is 0.3-0.5 and G (increased temp with increasing load, based on temp difference and floor area) needs to be reduced to 0.3 or so

for upstairs carpeted areas with a higher heat loss (ceilings and walls) and more floor insulation then an offset of 0, gain of 0.6-0.7 and again G of 0.3 but with all the virtual inputs, time and patience you can find a reasonable compromise for your hydraulics and constructon

i also discovered that perceived outside temp (including wind chill) worked much better than real temp, as heat loss is affected by the wind significantly

[Rob]

Thanks Duncan

We have standard radiators upstairs which seem to cope ok with AIR actuators but the downstairs UFH has been struggling in the sudden cold snap we've had so I wanted to try and control it a bit smarter and automatically vary the manifold temperature very much as you describe, based on climatic conditions.

Thanks for the gain "lesson", it's not my strong point as not had to get too involved before but when the wife complains it's cold then it's time to do something, hence changing to a new Ivar pump set.

I'm assuming you don't have a variable speed circulating pump (or do you?) Seems like it might be something to play around with pump speed and flow rates too.

Cheers

Rob

[Duncan]

i dont use a variable speed pump (that is to say, the pump has 3 speeds but i have quite a few circuits on my manifolds and always leave the pump on max, as when all circuits are open the system just about delivers the loop flow designed by the ufh company

if you use the As input (service modes 2 or 4) of the IRCs linked to a memory flag, you can set all the IRCs / zone valves fully open at the same time to hydraulically balance your system according to the design flow specs - thats the time to adjust the pump speed to meet requirements., then reset As to 0 for normal operation

if you log a rooms target temp vs real temp (again using subtraction and a virtual state) you can turn up the flow temperature to improve warm-up rate, but stop/turn down if you get any overheat (but make sure the overheat isnt just the presence of people in the room after its got to the desired temp)

its worth checking on the Min and Max (temperature parameters) of the ITC to ensure you get no surface damage - my upstairs is chipboard with carpet  so the max is set to 50 deg, whereas downstairs with engineered wood its set to 32 (this is hot water in, the surface temperature will be lower) but the problem with a very low setting is the flow water may not achieve sufficient to compensate for very cold weather, resulting in partial compensation and a slower warmup period in really cold snaps

instead of limiting the max temperature too much, it might be worth lowering the offset so the heatup period will be slower but more consistent across a wider range of outdoor temperatures - there is no 'right' for these settings at all

 if you keep the target temperature for a room constant 24/7 for the heating period, the target flow temperature can be much closer to the target temperature, particularly for hard surfaces