Underfloor heating

[mointainwalker]

@deuce

I am sure that your results are being falsified by your room-height and possibly by correction factors in the calculation software but that is a guess.

In order to be comfortable, you are only really interested in heating the bottom two metres of your room, however radiators at 70+ C will be heating a limited amount of air to a high temperature and this will rise (uselessly) into the roof. How it acts exactly as and when it cools and starts rolling back into the room I cannot say.

UFH on the other hand should ( by regulation) not exceed 30 C at floor level. Obviously with this, the air above will be heated to a lesser extent and rise less. UFH companies claim that this gives a temp of approx 20 C at around 1.8 m i.e. head-height. What happens above that I don't know/recall and don't know what heat you need to stop convection currents etc.

UFH circuits should not generally exceed 100 lm and at 10 cm centres that equals 10 m2 of floor. If you want to cover 60 m2 at 100 W/m2 you need 6 circuits which means two x 6 port manifolds ( feed and return).

If the budget is an absolute limit, it is not possible.

 

[looneyfitter]

HI Mountain walker i think were crossing lines, i am reffering to secondary flow temps and not floor temps.

 

[deuce22]

Hi mointainwalker.

That's the exact way that I look at it.

I think that it would feel warmer if the whole floor area gave out a less amount of heat as apposed to almost 3x the amount from 5 radiators, but only heating the area around it.

What is the reason behind using no more than 100Lm of pipe per circuit.

Is the pipe used similar to normal speedfit or is it a different type. I'm guessing it's different as I can't see it being able to bend that tight.

 

[looneyfitter]

Hi mointainwalker.

That's the exact way that I look at it.

I think that it would feel warmer if the whole floor area gave out a less amount of heat as apposed to almost 3x the amount from 5 radiators, but only heating the area around it.

What is the reason behind using no more than 100Lm of pipe per circuit.

Is the pipe used similar to normal speedfit or is it a different type. I'm guessing it's different as I can't see it being able to bend that tight.

The reason for not going over 100lm is because of excess heat loss within greater lengths of pipe.

This is why we create circuits. If you imagine the floor is zapping energy from the pipe so the futher round it gets the more heat it loses.

Have you thought about control as yet? as for pipe i would recommend uponor or polypipe.. I personally dont like speedfit as it tends to be a nightmare and kinks

 

[big-all]

as long as you are aware that underfloor heating is designed to be left on for extended periods and has a long time lag and will not operate in the normal way iff you only require heat twice a day for an hour or so
underflor heating would take one or two hours extra to warm a room up to the same extent

 

[mointainwalker]

The reason for the length of 100 lm is not as looney says :

The reason for not going over 100lm is because of excess heat loss within greater lengths of pipe.

That is just plain daft The whole purpose of the pipe is to lose/transfer heat and if it were possible to arrange for it to lose all of its heat you would do so because that would save cost and wear on driving the pump. a lot of effort and cost goes into covering the pipes as perfectly as possible with the screed and the reason for this is to eliminate voids which impede heat-transfer.

Reason for pipe length is the reduced efficiency of the pump being able to overcome the significant drag when pumping through a very long (relatively) small-bore pipe.

I am not familiar with UK brands, sorry, but the pipe required has to have an oxygen-barrier and is referred to as PEX pipe. It will probably look just the same as other pipe and it is difficult to bend and hold in place. When you get to tight bends, you allow the pipe-run to balloon out and then come back in to (say) your 10 cm spacing.

I've got to empahasise again that you cannot do this for 500 pnds. I guess you would pay at least 50p/lm for the pipe, so thats at least 250 to start with and then there is the insulation, wire-mesh to attch pipe, manifolds and more.

 

[looneyfitter]

Now that is ridiculous!!! Yes I agree with increase pump duty and pressure increase, but the length of pipe does have alot to do with heat loss too!!!!

It would be a great waste of money also having extra pipework carrying little or no heat for heat transfer. We are all aware that ufh can/does take along time to come up, laying stupidly long lengths also would make it take longer to come up to temp.

So yes we want to use all generated heat but to a point.

The pipe we use here is called PolyPipe, Uponor, JG SF etc.

You can also clip direct to insulation etc

 

[mointainwalker]

@Looney.

Can I ask how long you have been fitting UFH ? This is because in this one thread you have said numerous things that strike me as very odd.

The reason for not going over 100lm is because of excess heat loss within greater lengths of pipe.

As already said, i think that's just plain wrong.

To up your w/m2 output you can always up your floor temp slightly to 40-45deg.

Advising someone to have a floor-temp 10C - 15C above the regulatory maximum seems highly unprofessional.

One method is to lay foil insulation over the area to be heated(screw fix sell 20m2 at 100notes

An unusual recommendation. Foil insulation is very controversial and none has been thermally certified by the National Physical Laboratory because they get a miserable rating in the Hot Box Test (EU standard testing method for insulating materials).

That apart however, even if you believe their claims, they all state that you must have a 25 mm air-gap on both sides of the foil.

Can you say how you can achieve that in the layout you have suggested to the OP ?

Obviously if you install a system with inadequate insulation you commit the customer to a continuing waste of money because the heat goes into the ground instead of into the floor, the heat-input calculations will also be wrong and the system will never be able to reach its designed capacity .

 

[DeltaT]

I agree with you mointainwalker, but what you forgot was the fact that with UFH design, air changes are of little or no consideration and heat-loss through the floor is also not considered.

Perhaps if the OP carried out a further heat-loss calc. with this in mind, his figures would be more realistic.

 

[DeltaT]

'looneyfitter', you clearly have had more experience than mointainear and I like you have seen UFH systems that you'd think would never work - but they do!!

 

[deuce22]

Hi Guy's.

My situation is this.

The original bungalow, which is roughly the area that I'm looking to heat Is still in the process of being completed.

I am living in the extension that I built onto the back (even though this is not 100% complete yet).

As work is still ongoing I can adapt certain things as I go along.

I have calculated that it is going to cost around £600 for the 5 radiators plus copper pipe and extras. I would be willing to pay an extra £200 (£800 in total) if I could successfully install U/F heating.

However if it is going to cost much more for it to work properly, then I will have to revert to the radiators.

Although the floor area is around 60m2 in total, I will only need to cover between 40/50m2.

100LM covers 10m2 @100mm centres. Does this mean that I could use a 4 way manifold or could the lengths be connected together to cover 200LM and get away with using a 2 way Manifold.

If I were to use a Polystyrene type insulation what is the least amount of thickness I would need to use for it to be efficient.

I obviously want this to be as good as I can possibly get it, but I will not go ahead with it if it is going to a waste of time and money.

I know it seems like I'm trying to cut corners, but as DeltaT said he has seen UFH systems that you'd think would never work - but they do!!

One last thing.

Would the electric U/F system work and how much more would it cost to run.

I have found someone selling the electric matting for £70 per 5m2.

It is rated at 130w per m2. I would need 8 of them at a total of £560, giving me £240 left for the rest of things I would need.

Thanks.

 

[Agile]

The reason for not going over 100lm is because of excess heat loss within greater lengths of pipe.

Although he has not explained himself very well, he is correct !

If the flow temp was 45° and the loop was excessive at say 300m then the heat given up to the floor would have reduced the return temperature to say 25° which is far too low to give out enough heat.

So the length of each loop is kept to about 50-90m so that the return temperature is still high enough near the return to give out useful heat, perhaps with a differential of 5-10° between flow and return.

I am still suspicious of his calculation program and would like to see the same calculation used with a more common program. It would be odd for 100w/m² not to give out sufficient heat particularly with good insulation in the house. But that assumes at least 50mm of insulation in the floor UNDER the UFH pipes.

Tony

 

[deuce22]

Hi.

I have just gone to this guy's house with my friend to find out exactly what his setup is.

He has linked into the flow and return on the exact same circuit that is feeding the radiators to the rest of the house with speedfit pipe. He has connected 2x100LM of pipe together and has run this throughout his floor area in one complete circuit.

He only has 25mm of polystyrene under the pipes and none where he has fitted timber to the concrete floor.

He has then fitted his laminate floor directly on to the timber joists without ply boarding it first.

He told me that he only has the heating on during the colder months, but when he does put it on he keeps it on constantly at about 30/40 degrees.

He said that he has tested between keeping the boiler on constantly at a lower temp and having it come on 2 times per day at a higher temp and there's hardly any difference between the running costs.

The temp in his room is fine all throughout the winter, but he did say that the floor becomes really hot if the temperature on the boiler goes much higher than about 50 degrees.

The pipe is connected directly into a normal radiator valve, which can be turned on and off when he wants.

I mentioned about filling in the voids with sand to try and spread the heat more evenly and he said that he wished he had done that. Although the temp of the room is fine, there are certain spots on the floor (where the pipe is directly beneath) that are hotter, however you only feel that with bare feet.

I'm more confused now as his setup does seem to work, but is this OK to do and is this anything to do with the BCO.

I'm trying to make my mind up with what to do as I'm buying a load of materials on Friday and need to have a decision on which way I'm going.

Thanks.

 

[Agile]

You should have noticed that he is running his BOILER at the required feed temperature of 40-50°.

That makes it incompatible with normal rads or heating a water cylinder.

A 200m length of UFH pipe will mean that the temp differential is likely to be greater than ideal but that depends on what pump pressure/flow is able to achieve. A differential of 5-10° is about right.

What he has done is NOT wnat a professional would do as we follow normal guidelines which have been worked out to make the most effective use of UFH.

As an example of what NOT to do, I once saw a DIY job where he had used 200m of 10mm tube over plywood. That was then covered with an expensive hardwood floor. Not surprisingly there was little flow and the heat was mostly lost in the first 20m.

Tony

 

[deuce22]

Hi Tony.

In no way was this guy a professional. Although what he had done seemed to work.

What I'm trying to understand is the heat loss because of you using a long length of pipe.

The setup in my property is.

22mm pipe FLOW and RETURN both going up to the ridge height of the ceiling and running the full length (15.5 metres) and then reduced to 15mm feeding each radiator in each room.

There's a distance of about 10m from the boiler to the furthest 2 radiators and in total about 200LM of pipe.

The radiators don't seem to have lost any heat, so what is the reason for the heat loss of the pipe laid on the floor.

I need to understand all these variables, so I can make a more informed decision of whether to do it or not.

Also what do you mean by the rads not being compatible with a temp of 40-50 degrees.

Although I'm not a Plumber I do have an idea of how things work and knowing all of this makes it better for me, so that I can't have the wool pulled over my eyes by Plumbers I use.

Do you think it would be easier for me to use that electric matting I suggested earlier.

Thanks