Wire through wall for weather-compensation sensor

[Dreadnaught]

Next I will be putting on the external cladding of my timber frame (Cedral Lap planks). Beforehand I am looking at all the penetrations through the exterior wall.

 

One penetration is the wire for the ASHP controller's weather-compensation sensor (see image below of the sensor).

 

The manual says "Connection wire is 2 x 0.75 mm²".Google says this is "2192Y" wire, which seems to be a common type cheaply available from Screwfix, etc.

 

Should I …

 

1. just buy a roll of that wire?
2. install it through the wall with a loop outside and a longer loop inside?
3. Should I use a cover plate (see below in orange)? Or instead just slope the penetration downwards and back fill with silicon?

 

This is my first bit of "wiring" so would appreciate any advice and suggestions.

 

[markc]

Cheap as chips at SF and comes in handy for lighting.

slope hole downhill towards outside and form a drip loop unless going downhill once outside.

wouldn't bother with the cover

Edited November 24, 2021 by markc

[ProDave]

I would drill a small hole in the back of the box and place it directly over the cable exit so there is no visible loop.  A squirt of your favourite sealant to seal the hole. And a blank in the bottom in place of the stuffing gland.

[Nickfromwales]

With UFH pipes in a thick passive raft, and a cellulose-blown frame, do you really think WC’p will be effective given the likely very long thermal time constant you’ll end up with? 

[Dreadnaught]

2 minutes ago, Nickfromwales said:

With UFH pipes in a thick passive raft, and a cellulose-blown frame, do you really think WC’p will be effective given the likely very long thermal time constant you’ll end up with? 

Good point. Perhaps it will be as effective as having separate heating zones within the house. 

But as a standard feature of the "SensoCOMFORT" ASHP controller (don't you love those marketing names), I suppose it can't do any harm.

[MikeSharp01]

2 hours ago, Nickfromwales said:

With UFH pipes in a thick passive raft, and a cellulose-blown frame, do you really think WC’p will be effective given the likely very long thermal time constant you’ll end up with? 

My guess is that weather compensation will be done by an AI that uses history and other factors to work out what and by how much to compensate. This is already working in some large commercial businesses.

[Nickfromwales]

3 minutes ago, MikeSharp01 said:

My guess is that weather compensation will be done by an AI that uses history and other factors to work out what and by how much to compensate. This is already working in some large commercial businesses.

Weather prediction vs compensation so to say. Interesting. My experiences of weather comp ( Vaillant )  was not great. Hopefully things have improved.

[Dreadnaught]

I don't have any details about the weather compensation, other than:

 

"sensoCOMFORT comes with an outdoor sensor which provides an understanding of the thermal behaviour in and around the property and adjusts the flow temperatures to ensure optimal efficiency".

 

Oh, and it has a pretty LCD screen so it so it must be good.

[J1mbo]

It works a treat. Takes a while to get get the curve and offset really dialled in in my experience.

[Nickfromwales]

2 minutes ago, J1mbo said:

It works a treat. Takes a while to get get the curve and offset really dialled in in my experience.

Yup. I spent 2 days back and forth ( at my expense ) trying to get the curve right. Vaillants own tech guy just told me to ramp it up and walk away or I’d be there forever……

OK if it’s your own system and you can give up the time necessary. 
I still doubt it’ll make a hot of difference in a well insulated / high performing dwelling. Definitely a place for it in lesser energy efficient builds / commercial, but when a property loses 2oC over 24 hours then you have to ask why you bothered…..

[Dreadnaught]

39 minutes ago, Nickfromwales said:

OK if it’s your own system and you can give up the time necessary. 

 

Yes, I would be happy to tinker over weeks or months.

 

37 minutes ago, Nickfromwales said:

I still doubt it’ll make a hot of difference in a well insulated / high performing dwelling.

 

What is the alternative? I wonder how else one could attempt to put the right amount of energy in to the house over a 24h period to maintain the temperature at an approximately constant level while the weather varies? I think @TerryE has a custom computer algorithm to do it linked to a weather forecast. But that's the only alternative I can recall.

[ProDave]

17 minutes ago, Dreadnaught said:

What is the alternative? I wonder how else one could attempt to put the right amount of energy in to the house over a 24h period to maintain the temperature at an approximately constant level while the weather varies? I think @TerryE has a custom computer algorithm to do it linked to a weather forecast. But that's the only alternative I can recall.

The alternative is simple room thermostats.  Yes that's really high tech.

 

Just a few simple room thermostats turn off the heating in each room when it reaches the set temperature.  The house holds it's heat for a long time, and the other side of that is it only needs a gentle amount of low level heat. So when the heating is on it heats up slowly.  This slow heating and slow cooling means it does not overshoot when it reaches the set temperature.

 

So that is how we control the right amount of heat into the house in all weather throughout the season.

 

What weather compensation will add to that, would be to run the UFH at an even lower temperature when it is mild. That would improve the COP of the ASHP in the shoulder seasons.  At the moment it runs at a fixed temperature which is dictated by the heat needed in the coldest weather.

 

I maintain you need to run the heating a whole season without weather compensation so in the coldest period you can experiment and find the minimum UFH temperature needed at the coldest time.  Then, and only then when you know that, you have one point on the weather compensation graph.  You then have to experiment to find the other end of that graph.  That is not going to be a quick process.

 

And in the real world, with a well insulated low energy house, how much will it really save with the small improvement in COP achieved in the shoulder seasons?

 

Edit to add.  @TerryE heats his house overnight using the cheap Economy 7 rate, so he as derived a system that calculates the heating needed for the next day and times the heat input during the overnight cheap period.  S slightly different situation.

[Dreadnaught]

Thanks @ProDave.

 

My own instincts on this broad subject, and based on my reading around the subject, is to have a relatively small ASHP. Set it to run at a low temperature and almost constantly. And allow it to adjust the temperature using its own controller, which I can tweak the curves for a while as @J1mbo said, but with the aim that eventually it will be "set-and-forget". 

 

Oh, and not to bother with any heating zones or indeed any automatic actuators on the manifold at all. The relative room temperatures can be tweaked as needed by manually turning the manifold knobs on a set-and-forget basis.

 

And have but a single thermostat in the living room (the one in the image above).

 

Oh, I also plan to have electric UFH mats in the bathrooms so the floor feels toasty-on-ones'-tootsies in the winter. And this can operate as auxiliary heating if needed in the coldest depths of winter. If even that proves insufficient, a dimplex or two should suffice in a severe cold snap.

 

(And use the ASHP to cool the house in the heights of summer too.)

 

Keep it really simple and aim for set-and-forget.

 

Am I nuts?

[Dreadnaught]

I'm even hopeful that the above approach might simplify the plumbing by removing the need for the usual hydraulic decoupling between the ASHP and the UFH manifold with a low-loss header or similar. I assume some form or expansion vessel would be needed. But perhaps that is all. I would certainly aim to need neither a wiring centre nor wiring for multiple thermostats. (As you can tell, my plumbing knowledge is near nil.)

 

For DHW* I am still toying between a Sunamp** heated by off-peak electricity and thus completely separate from space-heating. Or having a more conventional unvented cylinder matched to the ASHP, such as Vaillant's own UniTower all-in-one, and therefore linked in with space-heating too. If I chose the Sunamp route, I am slightly concerned that the ASHP would stand idle for about half the year, when neither space heating nor cooling is needed. My fear is that idle appliances aren't happy appliances.

 

* DHW: domestic hot water. ** probably a Sunamp eHW 9.

[J1mbo]

@Dreadnaught the Vaillant controller is more than just a thermostat - it also applies the internal room temperature to the flow temperature target calculation (when set to do so) and can self-learn to an extent the heating curve. The room temperature based adjustment shifts the configured curve up and down as a whole. Basically you want this in the coolest place of the house to provide the most responsive system. Ambisense can be added to it to have additional sensors and electronic TRVs around the place, but regardless of the inputs from those other sensors the control panel sensor will be used for flow temperature adjustment.

 

Tuning the curve does however take ages and it is quite surprising in my case the result. 1970s house, CWI and double-glazed with radiators sized for 50°C flow. Five persons and to be fair quite high electrical load, 8MWh pa (computers!). But the curve has ended up being 0.65 with a floor of 28°C. It has taken a full year to dial this in. Based on my models the HP should be within 15% of rated SCOP this year. Since it has both 4-pipe buffer tank and a heat exchanger, I'm quite happy with that.

 

Mid-season, the heat pump modulation ratios aren't as much as it appears because it recovers so much from outside. i.e. 30% compressor still gives maybe 60% rated output when it's about 12°C outside. So it will overshoot then shutdown then repeat; a lot of water is needed to slow this down.

 

@ProDave To get anywhere close to the rated SCOP, the weather compensation is definitely required. Perhaps with UFH where the design flow is 35°C it is less important though.

 

 

[ProDave]

People keep saying in a well insulated house, individual room thermostats are a waste.  I find that not to be true.  We only have 3 zones downstairs, the living room, kitchen / diner and utility room.  I find the living room heats up quickest and is the first to switch off, next is the kitchen diner, and lagging behind taking much longer us the utility room.  If I did not have the individual thermostats, it is likely the living room would overheat if forced to keep running as long as the others.  I am not trying to heat them to different temperatures.

 

This really is a simple fit and forget solution.  Anything else and you are relying on a lot of dead reckoning, and you will spend a lot of time tweaking things to get it to work properly.

[Dreadnaught]

Thanks @J1mbo. Your experience of living with the system is very helpful and encouraging.

 

26 minutes ago, J1mbo said:

Basically you want this in the coolest place of the house to provide the most responsive system.

 

That's interesting. In my case, that the coldest room in the house will be the Utility Room. We have laid the UFH so that the supply runs that traverse under the Utility Room are insulated. The Utility Room will only receive heating from the (depleted) UFH pipes that return from the large main Living / Dining / Kitchen area.

 

Mine will be a pretty well-insulated and airtight house with MVHR so I expect that all rooms within the thermal envelope will be similar in temperature.

 

Would the Utility Room be a good location for the Vaillant controller? It was actually where I was planning to put it anyway, to hide all such gubbins. 

Edited November 25, 2021 by Dreadnaught

[Dreadnaught]

15 minutes ago, ProDave said:

People keep saying in a well insulated house, individual room thermostats are a waste.  I find that not to be true. 

 

Thanks for sharing your insight @ProDave. But in your case but without multiple heating zones, without multiple thermostats, and without automatic valves on your UFH manifold linked to a wiring centre, wouldn't you instead manually dial back the flow rate on those loops that supply the zones that heat-up the quickest. Once manually balanced like that, all your zones would reach the target temperature at approximately the same time. No over heating. Simpler system. A one-off exercise to set it up.

 

In practice would that work? Am I misguided?

Edited November 25, 2021 by Dreadnaught

[ProDave]

4 hours ago, Dreadnaught said:

 

Thanks for sharing your insight @ProDave. But in your case but without multiple heating zones, without multiple thermostats, and without automatic valves on your UFH manifold linked to a wiring centre, wouldn't you instead manually dial back the flow rate on those loops that supply the zones that heat-up the quickest. Once manually balanced like that, all your zones would reach the target temperature at approximately the same time. No over heating. Simpler system. A one-off exercise to set it up.

 

In practice would that work? Am I misguided?

If you set up what you are proposing, you are doing what is generally known as dead reckoning.  Even if your system as you suggest is perfectly balanced and heat input = heat losses, how id it going to cope when you have a party and have 20 people in the house and do a lot of cooking? it will overheat.  and how will it cope when someone does not shut a door properly and nobody notices for half an hour?  it will cool down.

 

I can't see the opposition to simple thermostats and a basic number of zones, it copes with most eventualities like that.

[J1mbo]

@Dreadnaught If you're running wires for this, I would consider giving yourself a couple of options for the controller. I started with the controller in the utility but later moved it to the lounge as it happens. It connects over eBus and needs to be cited clear of thermal influence such as direct sun, heat from the ASHP system itself, and so on.

 

The outdoor sensor needs to connect to the heat-pump interface (the brains of the whole system, and also the wiring centre for the 3-port valve, sensors, eBus etc), as does the VF1 buffer tank sensor and the DHW cylinder sensor, if you are having one. On that, the ASHP will be cheaper to run than using E7 and resistive heating for DHW. You can of course use the ASHP on E7 to heat the cylinder really hot at a COP of >1. I get about 2, heating it to 65°C. It can achieve much more at lower set points, the issue being that it needs to stop heating the house to heat the water so I 'charge up' the tank overnight to avoid it doing DHW in the day. With the thermal mass of UFH, that probably is much less noticeable.

 

Re zones. Bear in mind that the system needs a lot of water circulating to avoid overshoot. In other words the more zoning in place the larger the buffer tank needed. Different rooms obviously have different external thermal influences (e.g. sunshine, cooking, electronic equipment, log burners even) so it's always going to be something of a compromise between adequate heat, lowest running cost, and quickest response.

 

Incidentally I added Ambisense mainly for the data (VR920 internet gateway and the app gives hours insight of energy consumption, environmental yield etc) as well as push alerts to any system issues. It also provides eTRVs which I have just three of to enable 'steering' of heat between parts of the house as the sun moves around basically.

[Dreadnaught]

3 hours ago, ProDave said:

how id it going to cope when you have a party and have 20 people in the house and do a lot of cooking?

 

Good point. Another similar eventuality is solar gain from the sun, especially in the late afternoon. Presumably that cannot be predicted by weather compensation, which I guess relies on temperature forecasting.

 

Going back to @Nickfromwales point about the heating system having a long time constant (because of the mass of the concrete slab), my response to your point is that there is nothing much you can do about the overheating, even with zones, as any intervention will only have a appreciable affect some hours later. I can imagine that the only remedy is to anticipate that forthcoming big-party-of 20-people in the morning before and just turn the whole heating system off in advance.

 

Equally, might the ASHP controller flip the system in to reverse cooling mode when it detects overheating. I don't know.

 

Regarding solar gain, it was (long missed)  @Jeremy Harris who highlighted the benefit of running the circulating water through the UFH pipes without a heat source, to redistribute the heat around the house.

 

Thoughts?

 

 

[Dreadnaught]

 

4 hours ago, J1mbo said:

the more zoning in place the larger the buffer tank needed

 

Yes that is what I guessed. As a fan of simplicity, I hoped that abandoning zones would make the plumbing simpler. It seems it might. Although there seems to be a consensus that no zones might be going too far.

 

By the way, what is Ambisense? Are they just smart radiator valves? Do you use them with radiators or with UFH?

Edited November 25, 2021 by Dreadnaught

[ProDave]

19 minutes ago, Dreadnaught said:

Yes that is what I guessed. As a fan of simplicity, I hoped that abandoning zones would make the plumbing simpler. It seems it might. Although there seems to be a consensus that no zones might be going too far.

A normal UFH system has all the pipe loops connected to a manifold with a circulating pump and temperature blending valve.  A standard manifold has provision for a flow meter and an actuator for each pipe loop.  Typically 2 or 3 pipe loops make up each zone.

 

If you chose not to zone it at all, all you would be saving is a £10 actuator per loop and a £15 thermostat per zone.  It would not save any plumbing.  A few actuators and a few thermostats is a very small price to pay for controlability.

[TerryE]

IMO, the whole concept of zones gets very questionable as the spec of your house approaches passive class. It is a bit like trying to maintain storage zones inside a thermos flask. The internal heat flows are an order of magnitude greater than external losses. We run our 3 floor  house as a single zone which is heated by the slab.  My control system circulates water around the slab for 10 mins every hour when not heating.  This redistributes slab heat from solar gain and enables taking an accurate average slab temperature.

 

The only real exception to this is the floor to floor heat gradient in colder months, as the 1st and 2nd floors do tend to be a degree or two colder than the ground floor, but  I trim for this with a couple small oil-filled electric heaters on the two floors and they turn on for a few hours overnight on timers and add a few kWh of space heating on each floor.  The 1st floor one sits in the doorway of my office and this boosts the entire 1st and 2nd floors. My son occasionally uses his in his top floor area.

 

Edited November 29, 2021 by TerryE
correct to Oil-filled

[TonyT]

Very difficult to retro fit an external air sensor, so I would throw in a cable even if it want needed, small junction box to make it look good, and cable marked up at boiler.