Mike

Yes, if you're not into electronics and don't know someone who is, then monitoring is out - but if it did show a problem then it would be better to resolve it than to wait for structural damage to set in. However I'm not suggesting using it as a way of pushing the boundaries, only for curiosity or reassurance. I'm doing it for both reasons - WUFI suggests that my ceiling will be OK but, as it's a highly unusual situation and I still have a few reservations with WUFI, I want to validate it. Then option 1 on my list is probably going to suit you best, if you still want to consider IWI. That is, go to the technical department of Pro Clima or the like and ask if they'll do an evaluation for you. If you add your proposed rain screen (and it's not airtight), the chances are it will add to the drying capacity of the wall and be in your favour. Although if you can add a rain screen, maybe you can add external wall insulation, which is going to be preferable. There are many properties with cavity fill and joists embedded in the wall without problems so, if you forget the IWI and just fill the cavity, then you should be fine. However make sure that the rooms are adequately ventilated and you make the wall airtight, particularly at the joist-wall junction. At ground floor level, add plenty of ventilation to the sub-floor void (if that isn't already the case) and insulate the floor well - there are other threads on here about both.

From experimenting with WUFI, the place that experiences the highest humidity is (unsurprisingly) the junction of the IWI and the original wall, so ideal for a temperature-humidity sensor. There could be a moisture build-up behind a VCL (if not very intelligent), so perhaps another there. If a cavity wall has joist ends sticking into it, I'd be interested in what's happening at the cavity to joist-end junction too. You'd expect to see a build-up of moisture during the winter that declines again in spring (cold moisture is generally OK, warm moisture in contact with sensitive materials definitely isn't), without any year-to-year build-up. I've added a handful of sensors on my current project, but at various points above my very non-standard unventilated insulated ceiling, rather than for the IWI. I'm using pairs of DFRobot SHT31-F sensors & LTC4311 I2C range extenders, feeding into a Raspberry Pi via a pair of Adafruit TCA9548A mutiplexers. Until finished I'm just running occasional spot-checks: currently inside = 21.63°C & 49.8% RH, outside = 3°C & 90%, ceiling* = 10.02°C 79.53% - which is good for January, though it has been relatively mild. *above 300mm-ish of hemp insulation, no VCL, directly below timber decking + clay tiles in a very windy loft. Ubacus and similar software just shows a point-in-time, so of limited use. WUFI can throw up some very odd results too, with an incorrect internal environment. Modelling a single element in WUFI doesn't take long, so might only cost a few hundred pounds (but leaving a question mark over the output). I'd be surprised if whole-building modelling cost less than 2K. For someone doing a full-building thermal upgrade the cost would only be a fraction of the total renovation cost, so easy to justify - I can imagine some Architects insisting on it, or doing it in-house. I would guess that it's used too by Housing Associations and the like who have multiple homes to upgrade and/or their own taff to carry out the assessments. Probably not used so much by those on this forum, though there must be some.

Yes, those are key areas to focus on. It's a really interesting topic if you have the temperament for it; plenty of scope if you want to turn it into a PhD, or to do some DIY humidity monitoring on your eventual solution. FWIW the most extreme experiment I've come across so far is the French Réhafutur 1 project in Pecquencourt - but solid walls again, no filled cavity. They used eye-popping thicknesses of internal wall insulation, with 'intelligent' VCLs (not for the hempcrete) + very good airtightness + MVHR. Of course the documentation is virtually all in French, but I posted a bit about it a couple of years ago in this thread mainly in relation to hempcrete. The images in the linked video give some idea. Disappointingly, a long-term report doesn't seem to have been published (one day I'll get round to asking), but after year 1 it was looking positive (with higher than desirable moisture in the north-facing elevation thought to be due to moisture from the build still drying out). Tip: don't take things this far. Keep us posted.

As you may have gathered, what seems like a simple question is actually quite complex. Dew point is important, but dew point calculations don't take into account several key factors, such as the abilities of the particular building materials to move moisture through the structure by capillary action, nor the absorption properties of some materials (the ability to remove vapour molecules from the air and bind them within the natural fibres such as hemp and wood - not the same as absorption - before it condenses to liquid), then desorb it again. External factors are important too - hence my point about the condition and exposure of the external wall and the state of the DPC - because, as mentioned by @Gus Potter, insulating the wall changes how and when it retains and releases moisture. Then you have factors such as the presence or absence of vapour control layers, vapour permeable paints and MVHR... Anyway, to answer your question, I can only think of 4 ways of deciding on an insulation thickness: 1. Major manufacturers of vapour control layers will run it through their model and come up with a recommendation based, of course, on using their VCL, which must be installed to very exacting standards. You'd be wise to include a service void with a VCL to cut the risk of damaging it (now or in the future). 2. Find a consultant to carry out a WUFI analysis. However WUFI is highly dependent on the data used for the building's internal environment (variations in temperature and humidity throughout the year); the default data options provided are very finger-in-the-air and don't take into account factors such as the presence of MVHR, which can significantly cut moisture levels. That situation can be improved by using their companion software to model the environment of the specific building as a whole, but that's going to rack up the cost significantly. I'm not entirely confident in the standard external environment files used either, in view of the expected climatic changes, though at least those are more easily assessed and changed and have less impact on the output. 3. Take a cue from research such as that carried out by Historic Scotland. If only there were more of it available to cover a greater range of buildings and locations. 4. Go no lower than the more-or-less accepted 'safe' U-value limit of 0.5W/m².K

premature posting - hang on...

Yes, I was using the same version in 2008 and had a similar experience with the climate data, though I don't recall trying anything as exotic as Vlissingen! At the time I didn't think about getting data for the future climate; nobody then was expecting much of an impact from climate change this century, at least not among the general public. And the forecast data probably wouldn't have been available anyway.

This was something the Danes realised decades ago, at least for wind turbines. Their introduction, back in the 70s & 80s, was partly driven by local energy co-operatives, allowing local citizens to invest in and profit from their local wind turbine(s). I just went looking for some stats and, from here, in 2016, more than half of Denmark’s installed wind capacity was owned by citizens, rather than private companies. That's a lot of citizen involvement in a country where nearly 60% of electricity generation comes from wind power. There are similar schemes in France, allowing local citizens to invest and profit from local solar farms.

From a mention of GB0013a-Hemsby here, this is historic data, not forecast data (which likely also means that they're based on an average year, not a heatwave year). It looks like PHPP 10.6 included updated climate files, but can't spot anything about how they changed. Your PHPP manual may say something about using .epw files. They seem to have become something of a standard, so I'd hope that they could be easily used now. As opposed to the climate data sheet in my PHPP, which just says 'To enter new data, fill in yellow cells'! Alternatively, it looks like Meteonorm sells PHPP-ready future climate files.

My PHPP version is ancient, so I can't take that as a guide, but I wonder what data their current climate files are based on. Is it still a historic analysis, or a forward projection? The projected climate of the 2050's will look shockingly different to historic data from the 1990's. I think that PHPP uses EnergyPlus weather files (.epw); if so then I'd suggest using the University of Bath forecast data; see my post here:

+1

It's also susceptible to cracking & edge damage if something heavy gets dropped on it. I have seen a couple of negative posts about it in France - impact damage and surface scratching. There's no perfect worktop, but honed dark granites come the closest, though expensive.

Unfortunately the chances are that this won't work well when you most need it. Passive stack ventilation is driven by the difference between indoor and outdoor temperatures; on a hot summer day that difference can be small, resulting in very little stack effect - so very little cooling. In fact the outside temperature may exceed the internal temperature, causing the stack to operate in reverse and drawing in warmer air. For it to have a chance of working you'd also need to crack open some vents or windows, so that the air you hope will go out the Velux has somewhere to come from. I'd strongly suggest a chilled slab or aircon instead. If choosing the former then, as @JohnMo says, don't over chill the slab; in France the regs say that floor surface should go no lower than 18°C to avoid condensation.

@RedSpottedSev I should have mentioned another safety point on this topic, if you choose to go down this route. The wires held by screw terminals on contactors (as well as RCBOs and the like) must be the same size, so that they are properly secured in place. So, if you were to use 1mm² contractor control wires (black in my image above), you'd need to solder short stubs of 1mm² wire to the end of the contactor leads before applying the heat shrink (unless they're already 1mm² - I can't remember). It's interesting how this is viewed in the reverse direction in France. Here the wiring regs don't allow the mixing of 230V and ELV wiring in a domestic consumer unit so, as new installations have to certified by an independent inspectorate, all the control gear needs to be 230V! I'll be adding the patch cable to the Shelly afterwards.

It's good in principle, however by using a contactor you bypass that function - it will only tell how how much power the contactor itself is using, not how much power is passing through the contactor...

As @JohnMo says, if you can use a time switch that's the easy and cheap option. For the Shelly solutions, the main practical difference between the Pro 1PM and the 4PM is that you need 1 x 1PM for each device (heater, in your case) that you want to control. The 4PM can control 4 devices (and has manual controls in addition to smart controls) and would work out cheaper. Here's a diagram (modified from my installation to cut out some bits) that shows the wiring; your electrician may prefer to use RCBOs instead of the breakers & RCCB shown here. The latest version of the 4PM has a single live supply terminal, not 4. I'll be wiring the snubbers directly across the contactor terminals, as shown. The one Shelly sells has bare leads, which would need covering in heat shrink sleeving for safety. I don't know the T-Smart, but if it only works over WiFi and by using their proprietary app, then it wouldn't be my choice - that's all their brochure mentions.

France certainly can't build them cheaply. Flamanville 3 was due to be completed in 2012 for €3.3 billion. It opened at the end of 2024 and cost €19.1 billion (in French). In Finland, Olkiluoto 3 was expected to open in 2009 for €3 billion, but opened in 2024 for an estimated € 11 billion.

I prefer the 4th one you posted, which seems to be a pale blue on my monitor, for a little more contrast.

Not structural. Most likely the relic of a 1970's renovation.

Jointly awarded the Nobel Prize in Physics 2022 "for experiments with entangled photons, establishing the violation of Bell inequalities and pioneering quantum information science". Clearly not an expert in climate science then. Then there's: Alexis Carrel - Awarded the Nobel Prize in Physiology or Medicine 1912 "in recognition of his work on vascular sutures and the transplantation of blood vessels and organs". Believed in telepathy and eugenics. Richard Smalley - Jointly awarded the Nobel Prize for Chemistry 1996 "for their discovery of fullerenes" Believed that evolution could not have occurred. There are others...

A very good point. Since maintaining vapour-permiability will be essential following my current renovation, I'll be inscribing that, together with a few other key points, on a metal plate that will be screwed next to my MVHR unit for posterity - it may be worth considering. In theory I shouldn't need to do that here in France because a mandatory 'house manual' is now required as part of the conveyancing process. It has to include, in detail (product names, technical characteristics, serial numbers, etc.), all the materials and equipment related to insulation, thermal performance, energy use & generation, etc., from initial build (if built after the law came in) through all subsequent modifications. It must also contain instruction manuals, servicing requirements & maintenance details - for example the need to use vapour-permiable paints.

Science is a process. Over time, through repeated experimentation, testing, review and criticism it develops explanations of reality, often based in mathematics, that best fit the currently available evidence. If new evidence is gathered and/or better explanations are developed, then the consensus best explanation of reality eventually changes. Opinions are subjective personal judgements. Individual scientists may hold them, but that doesn't turn opinions into science, nor science into opinions.

If you have a cavity that's certainly the best solution.

I briefly mentioned using cork as a spacer earlier; I'd use continuous strips, not just spacers every 600mm; without that kind of firm support 45x45 would be too flimsy. Personally I'd increase it to 45x75 if I was using timber.

Yes, it will be a big help Installing MVHR, would help too. It keeps internal humidity lower than it would otherwise be, cutting the moisture entering the structure in winter and allowing it to dry out quicker in spring. Making the building airtight should also be a key priority and to accompany that you need an effective ventilation strategy, be it MVHR or otherwise. There are several threads on here about both topics.

Nor me, and £86 million is certainly worthwhile. Just a shame that their parent company isn't British too - their turnover was €17.3 billion (the 4th biggest company in Denmark).