UFH Prob

[looneyfitter]

Went to a job yesterday, cust says UFH stopped working. I cant see that it ever worked.

20m2 conservatory single loop from a blended manifold. He says it had 100mm kingspan in, 120m 16mm uponor and 130mm of CONCRETE over the top of it. It had a flow temp of approx 45deg. Also he said he always had it set to timed on off.

The blender that was being used was just a standard undersink blender 15mm 18lm tmv3 ( i think this should be ok but never saw one used before) what you think?

Im tempted to alter the manifold too as at present the mix port on the blender also goes to the return of the boiler and flow of the UFH circuit.

The pump is setup so that it sucks through ufh return but then mixes with the blended water and the boiler return which doesnt seem right. Ill draw a pic in a min.

Now in my opinion the setup itself will never have worked. I have suggested he needs a higher flow temp (not sure on the max on the tmv3 and to use a setback programmer, but even then its gonna take forever to heatup.

What would you suggest the min flow temp should be and also what is the heating output of 120m of 16mm?

I cant find any charts

Cheers Gareth

 

[Agile]

45° is already more than usual for UFH flow temp.

Whats the return temp? Its that which is more relevant so you can see the average.

Also the pipe spacing, if needby worked out from the size and length of pipe ( who told you its 120m ? )

With 130mm concrete above the heating pipe its going to need being on 24 hrs and the heat transfer might be losing quite a bit too much as well.

Whats the heat loss of the conservatory. They are usually far more than a habitable dwelling.

You need to have it on for 24 hrs and measure the surface temperature and the room temperature and outside temperature during the evening when its cooler.

Then you can get an idea of how useful its going to be.

Tony

 

[looneyfitter]

cheers for the reply mate, the cust laid the pipe himself, he said it was a 120m coil and used it all in the floor.

Im going back tommorrow so ill fire it up again and get my return temp. I really dont think its gonna work tho. I recon it more like 180mm centres.

Sorry i should have said the temp he run it at was 30ish. I want it to run in the 40's[/quote]

 

[Agile]

To be meaningful I think that you are going to need to measure the return not just initially but after a few hours as the initial heat output is likely to be higher than after when it has warmed up a bit.

I suggest you measure flow and return intially and then say the next day as well.

Dont expect to get paid much for a no brainer job like this.

Tony

 

[looneyfitter]

all sorted mate, i altered the manifold so the pump pumps directly off the blender to the flow. Its running at 40 deg. Cust says the room is toasty now.

What i did find was that with the setup he had too, was that the pump is now putting negative pressure (sucking) on the primary return, Not sure weather i should fit a NRV on it?

The prob it had was that the pump in this orienation seems to be robbing the primary supplies to the 2 rads he has in there too? I wondering weather to re-pipe the rad primaries so it is fed off the manifold. I know it will then be blended but the rads are only there to add a little extra heat.

What you think? Still on the UFH learning curve

Cheers Mate

Gareth

 

[Rhondo]

The cold feed to the TMV should have a NRV on it otherwise when the primary pump operates the flow will reverse back through the tmv.
I usually fit the low resistance types.

 

[looneyfitter]

Yeah the cold to the blender has a NRV in it. What I am saying is the way it is setup, the pump will also suck on the primary return. Doesn't seem right to me? Thinking fitting a NRV here.

 

[mysteryman]

The UFH circuit pump should pump away from the mixer, and the return should branch into what is the cold port on the mixer. The mixer then blends return water from the UFH into the flow from the boiler. This gives you the lower flow temp for the UFH without raising the return temp to the boiler.

 

[looneyfitter]

Yeah I know that mate, what I am saying I that the pump is also suckin on the PRIMARY return as well which I don't think is correct.

 

[mysteryman]

Yes, I think it is.