Any underfloor heating installers here, familiar with JG Speedfit installations?

[GRC]

Title says it all; friends moved into a home with an extensive underfloor system, but trying to figure out what they're missing about operation.....

 

[GRC]

Owner feels that UFH is not heating the place at all. The central manifold installation is built primarily with JG Speedfit components; there are two separate manifolds, each with their own pump and what I gather is a blending valve.

The pump and blending valve assembly is very similar to this;

The pipes coming in to this assembly are piping hot, the pipe leading from the top of the pump to the inlet section of the manifold still hot, but not piping hot, and the flexi piping leaving the manifold to each individual circuit seems barely warm.

Q; Does the blending valve control the overall temp of the whole system? This is currently set at a smidgen over 50 - assuming this to be farenheit. for it's not clear, that's around 11C ....

Is it just a case of turning the blend valve higher to coax the whole system to a higher temp?

 

[robinbanks]

The pipes coming in to this assembly are piping hot,

If UFH is the only heat emitter in the system then the flow temp may be excessive

the pipe leading from the top of the pump to the inlet section of the manifold still hot, but not piping hot,

It's part of a circuit that mixes the colder return from the UFH with the hot supply. UFH shouldn't be run at the same temperature as radiators are

and the flexi piping leaving the manifold to each individual circuit seems barely warm.

Yeah but the flexi piping is plastic. Invest in an infrared thermometer rather than going by the feel of it

Q; Does the blending valve control the overall temp of the whole system?

Not the whole system, just the pipes that run through out of the manifold top rail, through the floor and back through the bottom rail, through the pump and back into the top rail

This is currently set at a smidgen over 50 -

I wouldn't necessarily go by any numbers I could see on the valve

assuming this to be farenheit. for it's not clear, that's around 11C ....

I wouldn't assume it's Fahrenheit; an 11C circuit wouldn't feel warm, because your hand is typically warmer than 11C, so it would feel cold

Is it just a case of turning the blend valve higher to coax the whole system to a higher temp?

No, it's not just a case of that; there are probably flow regulators on the manifold that may need adjustment too. What is the floor covering? What temperature do you measure at the floor in each zone, the top rail and the bottom rail?

Flow temp might need to be raised so heat can bully it's way through carpet or other insulated floor coverings, but care should be taken for zones that have more transmissive coverings such as LVT; don't set the flow/temp so high that the floor covering is damaged

UFH typically works best in a well insulated, draught free house and run at a low background temperature; my flow temp for the UFH is 28 degrees. Running mine at 50 would cook the occupants and make the floor uncomfortable to stand on because it's LVT

Do you have the original design documents for the loop layout and flow/temps?

 

[MeldrewsMate]

Please don't be tempted to fiddle with settings until you have a clear understanding of what you are changing, and its risks and benefits, for instance the blending temperature should rarely if ever exceed 50C (that dial scale is in Celcius). The inlet pipes should indeed feel hot, but the blended flow pipes (the warmer section of the manifold) should feel cooler, though a true 50C blended flow will be uncomfortable to hold. The return manifold will have considerably cooler temperatures, feeling only tepid.

 

[Madrab]

As others have suggested, it's not just about turning up the blending valve to get the flow temp hotter.

The system needs setup properly and that is managed by understanding certain key conditions/settings/requirements.

1) UFH type and what controls are in place - concrete&screed/overlay/floating spreader plates/inserted w/ biscuit layer/etc - are there floor/air stats
2) What Floor temps are required - floor coverings are the usual deciding factor here
3) Flow rates - how much heat is required per m2 to adequately heat the space - calculated at W/per m2
4) Pump settings - once 3 is known then how much flow/pressure is required to supply the loop areas correctly on each manifold @ peak demand

So there is a lot to consider even before understanding how those requirements are then achieved. As suggested there may be a design setup doc that details what was installed and how it was set.

 

[SpecialK]

Keep an eye on the boiler as it may be cycling and going in to an anti cycle mode where it wont heat for 5 minutes.

Do you have the grundfos pumps on UFH mode? Ideally you need a thermometer to measure some temps. The feed in to the UFH should be similar to the blender valve setting. Before you fiddle with anything take a note of what flow rate each circuit is set to.

And a reminder that UFH works very differently to normal rads, takes ages to get heat in to a room. Mine can take a 1.5hours to get a 10 degree rise in the slab temp!

 

[roguetrader]

First things first, make sure the loops are properly filled and free of air. That's UFH 101.

 

[GRC]

Thanks all; solution was to re-pair the room control panels to the central control unit; unbeknownst to me, the owner had reset the system in some manner, resulting in all the wifi links being lost.