Underfloor heating - anyone used Uponor products?

[RichardS]

About to have an extension to a kitchen built (_when_ the party wall agreement finally arrives and is signed.. 6 weeks and counting now, grrr...) and we're using the disruption and opportunity to change the ageing combi and put in underfloor heating.

We're looking at a couple of manufacturers products - Hepworth and Uponor being amongst them, but wondered whether anyone had any experience of Uponor products, esp their underfloor heating solutions?

In particular, although a lot of information is available for their systems via the website, one question that isn't answered is which product to choose - PEX or MLC (multi-layer composite, I think) pipe and what the relative advantages of each are.

Floor construction is suspended timber - I'll install as much kingspan (or similar) as will fit between the joists and then fit the heat spreader plates on top laying the loops in the channels. 18mm ply on top (_VERY_ carefully screwed into position) and then ceramic/porcelain tiles on top. The new floor has to be concrete (apparently, to meet current building regs) but we'll level it to the height of the underside of the joists so that new joists can be run right across the width of the room & avoid having to undergo a screed/suspended transition on every single run of pipe. Area of room is about 25m^2, area of heated floor will be about 20m^2.

New combi will be Ideal Isar (or Icos, can never remember which is the combi...) HE30.

The other thing that I'm having difficulty getting to grips with is the area of heating controls and how the UFH control will interact with the programmable stat and the combi timer. I had thought that a fully programmable stat would negate the need for the combi timer, but heating engineer reckons not.

It would be so much more convenient to be able to control and program the heating from the hallway rather than having to open the boiler cupboard, and if a programmable stat capable of multi-zone operation can also control the UFH then so much the better.

Am I barking up the wrong tree with this?

And another thing... when running with a combi, is a pump needed in the UFH manifold, and how does this interact with the one in the combi - will there need to be a 3-way valve so that the floor can be operated independently of the heating of the rest of the house?

I thought I'd got it, but every time I think about it more questions pop up... (the Curse of the Software Developer ;-) )

[RichardS]

oops - sorry about that - still getting used to supergravity newsreader....

(meant to start new thread!)

[The Natural Philosopher]

No, but I have installed UFH so have some info. I usd polyplumb -

- so if you haven't yet checked them out, maybe worth a try.

Pass on that.

Ok so at best 2KW or so for close laid pipes.

Ditch the combi and get something decent ;-) I;ll explain why later..

I reckin yes. *If you wire it up correctly*.

Not at all.

Something like that. This is where you run off the edge of most heating engineers know-how.

I can't answer for every system possible, only describe the rationale of what I have was given by the heating engineer I used.

Firstly, especially with screed floors, there is a huge time constant - in our case it takes many HOURS to raise a cold room to correct temp. So you must have independent timing from other heating circuits.

This s exacerbated by the fact that super hot floors are simply not desirable - you get wood warping or screed cracking. UFH water temps are limited to about 50C at most. So peak input is not that large - on cold days it may only be enough to just keep up with heatloss. About 100W/sq meter s all you can realistically expect, and 50W/sq meter is more realistic.

Because of these constraints, the UFH water needs to be temperature reduced somewhat. You boler can and often does produce water up to 65C or more. The wy this is dne in te Polyplumb system, is to utilise a temperature reducing/mixer system with a separate pump. Basically the UFH is pumped as a separate circuit and enough hot water from the boiler pump circuit is 'bled' into it to keep it functioning correctly.

This pump is entirely separate from any other pump, and normally yiu simply put a motorized valve between the existing heating circuit (two way on off) and make that control the UFH pump, and run that from whatever programmable stat you want.

So up to this poit., what you have is your own timer/stat feding a motorized valve that splices the whole UFH circuit into the boiler circuit, and also firs up te UFH pump. Should the boiler happen to be on, all works well. However if its not, you need to find a way of calling for heat for UGH only. You cannot just connect the UFH pump ad the boiler and pump in parallel, because if you do that you will have the UFH pump run if some other source needs the boiler.

My solution as indicated by the engineer was to fit a relay across the UFH pump, so if that is running the relay contacts are use to call the boiler to attention. But the reverse cannot happen.

Now, as to why not a combi: for a very simple reason. Many of these - if not most - are designed for dead simple installation. You may not be able to add external valves and relays and stuff easily, if at all.

My boiler is a system boiler: It contains just a pump, and a boiler with an exit temperature thermostat to limit temperature of circulating water. I have three external motorized valves for CH, UFH and hot water (mine is a PHW tank with indirect heating) This wan't that easy to suss out, but was easy enough to IMPLEMENT from a three way timer and so on, plus that relay.

I am less sure about a combi - many have their valves internally and are a bitch to add stuff into - at best you would need a Y valve externally to go between CH and UFH - or both..integrating that with existing electrics sounds nasty.

Do be sure to understand what the boiler you want to use has available to be customised. Some simply are not amenable.

.

[John Rumm]

ISAR is the combi - ICOS is the system boiler version. You may as well have the HE35 for the better hot water delivery performance - there is not much difference in cost.

I have just a prog stat on my isar - no other timer at all.

The stat needs to be in the zone it is controlling really (unless it has some form of remote sensor capability). Hence you have one prog stat per zone.

No.

Depends on how your UFH is setup.

Note really - the pump in the combi will run whenever it is running - either pumping water round the heating circuit via any zone valves, or round its internal closed loop when driving DHW via its plate heat exchanger.

You will need a S plan system with several two port valves. That way the UFH can call for heat independently, and can run on its own without the rest of the house getting extra heat.

;-) the paralysis of analysis!

[John Rumm]

With these boilers the combi and system version are functionally very similar. They also have quite a wide modulation range that would be a benefit on a UFH system.

No need for separate timing though, a programmable stat will do the job.

Also the OPs construction technique ought to exhibit slightly more responsiveness than a solid screed based solution.

It helps if the existing circuit is made S Plan so you can have true independent control though, since you may want much longer boiler firing times for the UFH.

You don't really need to worry about the boiler pump in this setup - it is under control of the boiler and there are no external controls to it. When your UFH heating stat calls for heat, it can enable the UFH zone valve, and this in turn can call for heat from the boiler. It deals with its own pump. A secondary pump in the UFH system can be controlled from a number of places - most likely derived from the call for heat passed on from the zone valve.

Not sure why you recommend a system boiler over a combi, since for this application (i.e. ignoring DHW for the moment) they are pretty much identical. Both will give slightly less scope for interacting with the boilers pump control, but in general this is a good thing since it eliminates an area of complexity while also allowing proper pump overrun control.

I think you are worrying unnecessarily in this case. So long as the boiler allows for a simple "call for heat" input, controls its own pump, and can modulate over a decent range of output powers then you are sorted. Implement whatever control system you want around it with timers / stats etc and leave the boiler to provide the heat when required.

(you can even have a stored DHW heating zone if it takes your fancy - just because a combi has an "instant" DHW option does not mean you need to use it for all DHW needs (or at all for that matter))

I can't see what you would want to customise... the only problem would be with ones that don't support standard external stats, and instead require proprietary heating controls like the particular model of WB combi discussed a few weeks back that insisted on needing a WB TR2 controller.

[The Natural Philosopher]

That was what I had in mind. I am no expert on combis in detail,but I assumed that many were similar - one shot installations of a 'package' that even a plumber could get right ;-)

The fact that his 'heating engineer' said he couldn't split the UFH zones out to a seperte timer/stat raised alarm bells with me as well.

>

[John Rumm]

Most have a standard call for heat like any ordinary boiler. The main difference is they take care of running the pump themselves (same is true of system boilers). A small number of odd controls - but they are very much the minority as far as I can tell.

Plumbers like them because they are simple to install in their minimal configuration, however there is nothing to stop you doing more sophisticated stuff if you want.

Some of the top end ones encourage this by providing things like analogue temperature demands and feedback etc, which lets you do cleaver split temperature stuff to suit different applications like getting your thermal store up to a decent temperature without needing to run the CH hot as well.

Yup, that did sound a bit suspect.

[matthew.larkin]

The OPs system sounds much like mine, so I am interested in this "S plan system" terminology. As I've posted previously, we've got a UFH in the kitchen, with its own (dial) thermostat, and then the combi controlled by a programmable stat. So if the main stat is on, the UFH can call for heat, but not vice versa.

Can you point me in a direction where I can better understand the "S plan system", or should any competent plumber ("heating engineer") know what that would mean?

Thanks!

Matt

[John Rumm]

That sounds rather like your UFH is not on its own zone...

The simplest installation with a combi boiler is to have no zoning at all - just one heating loop that feeds all the radiators. Chances are this is what you have with the UFH connected as if it were just a rad. Not ideal, but possibly workable if the stat is in the room with the UFH and the other rads have TRVs.

Yup, try this:

heating engineer ought to know what they mean - but it is not guaranteed. They really ought to understand the concept of independent zones thought even if they don't recognise the different "plans".

[John Stumbles]

Does any of this help?

[PeterK]

"RichardS" wrote in message news: [e-mail]...

It might be worth a visit to a branch of Graham plumbers merchants ( ) My local branch sell uponor and their "plastic" plumbing is mainly Hep2O. One of the counter staff does site visits for one off underfloor heating projects ( being a plumber / account holder doesn't seem to be a requirement )

Peterk

[RichardS]

Yes - it most certainly is!

Sorry John, should have read through that first - the wiki is very good and many thanks to all who have contributed to it. Didn't think of it as it's appeared in my long absence from uk.d-i-y.

Many thanks

[RichardS]

yep - I was looking at the HE35 but the heating man thought that was complete overkill for the property - he was trying to persuade me even further down. I might reconsider and go for the 35 anyway in preparation for installing a separate shower cubicle. We've only got the one bathroom and I dont think that a loft conversion is on the cards whilst we're here.

Yep - I know that there is also a remotely mounted control option for the Isar which would at least mean that it is more conveniently (and visibly) mounted rather than in the kitchen boiler cupboard.

thanks - I suspected as much but wondered whether the programmable stats now could be aware of other remote stats and control other zones (ie the UFH). I'll be looking more closely into this - I guess the tradeoff is being able to control everything from one controller versus complexity in programming it and attempting to explain the whole setup to SWMBO!

Ta. Between the replies & the Wiki I think I've got the hang of this now. I did check that the Isar has a "call to heat" and it does appear to have - the installation manual shows L1, L2 & L3 connections with "Other heating controls" in the circuit so that pretty much indicates that it should be able to do what it wants. A bypass may be required by the looks of things though.

Indeed. I'm well known for it!

[RichardS]

Ta, will lookmore closely at them.

I've just had an email from Uponor in response to a request to their offered design service. Will specifically ask this question and report back on any reply.

Yep - reckon this will be about enough - the u-value of the new wall, roof, decent floor insulation & replacement of current sliding patio doors with new decent units will probably mean that new room will require lessheating than old one. Current rad is something a bit less than 2KW. Will supplement this with a small rad anyway, mostly for something to dry towels and things out on.

Ah, not an option am afraid. There just isn't room that I'm willing to part with for a cylinder, and the layout of the place would bump the cost of the job up vastly if I wanted to do that. Besides (and I know many may think us mad), we actually _like_ the combi's advantage of letting us just completely forget about planning for hot water at all.

yep, understood. In this case, though, decent insulation underneath and no screed should minimise thermal lag.

They reckon 70W/m^2.

Thanks - I've checked the installation manual now and it does appear to support external valves and zoning solutions. There appears to be a call to heat input, so I think I'm ok with this. Will be double checking first though!!

Many thanks

[John Rumm]

Depends a bit on how often you want baths... in the winter even the 35 can only fill them at about 15 lpm. It will *just* manage two showers at once if they only need moderate flow rates.

The bog standard control is an on/off switch and two max temperature controls - one for HW and the other for CH. Not much need to ever do anything with it.

I would have though it would make more sense to just go for ordinary prog stats in the places you need them. Then it ought to be pretty obvious which area they apply to.

Yup, mine is controlled via a Horstmann centaur stat 7. Which works nicely. I don't have it zoned however, so no need for a bypass (hall rad is controlled with two lockshields and no TRV) in that case. With multiple zones then you need to allow for the case when all the zones are off and the boiler wants to overrun for a little while.

[RichardS]

Had an email back from Uponor today about the MLCP/PEX choice:

"The MLC is the superior pipe, as it is form-stable, hence easier to=20 lay. It is also 100% resistant to oxygen diffusion, so your system=20 components are not exposed to corrosion. =20 It is slightly more expensive per metre than the older PEX system (=A31.58= =20 vs =A31.26), but the controls are cheaper so it works out about the same. = =20 I would definately go with 16mm MLCP."

I also asked about installation methods for my floor construction=20 (suspended timber): batten-supported rigid insulation (kingspan, etc) underneath with either pipes clipped directly & screed, or using the=20 heat spreader plates with the pipe running in the channels.

Answer came back "shouldn't experience problems with either method, but=20 screed may give a 'faster' floor and be more economical on under-floor=20 materials".

Note that additional support may be necessary under the insulation with=20 the screed method though (ply support, I guess).

I think I may go down the screed route.

--=20 Richard Sampson