All Activity

For the second winter in a row, my Wilis UFCH is operating manually, but by next winter it will be more sophisticated. A Shelly Pro 4PM will be capable of scheduling the days and times to switch on and off & providing remote control. Upstream of the Shelly there will eventually be a Raspberry Pi which will calculate the heat required, monitor temperatures and command the Shelly (which will become a fall-back in case the Pi is out of action). Either way, the Shelly will send control signals to a pair of 25A contactors (one for each* Willis heater) and will directly control the pump (since that draws little current). A Hager 60060 load-shedder will temporarily cut the contactor control signals if the mains supply is under heavy load, but in your case, @Antonb182, that would be the place to put a thermostat. However, note that with UFCH isn't as reactive as radiators - there's a significant time lag between heating the floor and the air reaching the set temperature that you'd need to take into account to avoid overheating. Schema + sketch below. All the this (excluding the Pi) will take up one row of my consumer unit. *Although I could heat my place with 1 Willis heater, I've chosen to install 2. That's partly to have the option to use both to load the floor with heat during off-peak hours, but mainly so that I can switch to the second if the first should fail.

My money is on the PRV in the outdoor unit. It is a known weak component and you would not see the outflow.

Hi all. Having purchased Ecosash windows, we are now looking to buy french doors for our build. We have gone to 4 suppliers Timber Windows Rom Valley ASG We origionally wanted Timber but the costs are so high for them. Here are the figures. What are peoples opinions on TimberLook? Trying to see some french doors from Timberlook is very very difficult... Timber Windows Redwood Front door £4,094.80 £818.96 £4,913.76 French door set 1 £6,316.30 £1,263.26 £7,579.56 French door set 2 £6,316.30 £1,263.26 £7,579.56 French door set 3 £6,363.14 £1,272.63 £7,635.77 French door set 4 £6,259.73 £1,251.95 £7,511.68 Single door with sidelights £4,841.21 £968.24 £5,809.45 Back door £3,552.57 £710.51 £4,263.08 Timber Assured (Deposit and Warranty protection policy) £83.93 £16.79 £100.72 £37,827.98 £7,565.60 £45,393.58 Discount 15% £5,674.20 £1,134.84 £6,809.04 Total £32,153.78 £6,430.76 £38,584.54 Timber Windows Evolution Front door 3,559.49 711.9 4,271.39 French door set 1 5,128.65 1,025.73 6,154.38 French door set 2 5,128.65 1,025.73 6,154.38 French door set 3 5,128.65 1,035.76 6,164.41 French door set 4 5,012.45 1,002.49 6,014.94 Single door with sidelights 3,573.83 714.77 4,288.60 Back door 3,018.37 603.67 3,622.04 Total 30,550.09 6120.05 36,670.14 Rom Valley Estimated breakdown of costs, only have a total figure R9 French door set 1 4,234.17 846.83 5,081.00 French door set 2 4,234.17 846.83 5,081.00 French door set 3 4,241.67 848.33 5,090.00 French door set 4 4,138.33 827.67 4,966.00 Single door with sidelights 2,950.83 590.17 3,541 Back door 2,492.50 498.50 2,991 Total 22,291.67 4,458.33 26,750.00 AGS Estimated breakdown of costs, only have a total figure Timberlook Front door 4,234.17 846.83 2,557 French door set 1 4,234.17 846.83 3,684.00 French door set 2 3,070.00 614.00 3,684.00 French door set 3 3,075.00 615.00 3,690.00 French door set 4 3,000.00 600.00 3,600.00 Single door with sidelights 2,139.17 427.83 2,567 Back door 1,806.67 361.33 2,168 Total 18,291.67 3,465.00 21,950.00 Without front door 19,393.00 I think we would go with a timber front door but the rest will be either R9 or Timberlook. The cost of TimberLook is almost too good to say no. They also do Chrome hinges which R9 do not do. Thanks Nick

Do not add insulation in the roof void under a warm flat roof - it creates a hybrid flat roof that will most likely have condensation problems. By all means pack the roof edges with mineral wool to both minimise air leakage and ‘link’ the wall and roof insulation to minimise thermal bridging (& in a fire safe way).

If it's trailling across site, make sure it's HO7RN-F as that will stand up to mechanical stress and weathering.

Plus 1 to that. Once you take into account the build sequence then agree second option is more practicable. I know you have just asked about the inside but can you just check the weathering detail on the outside? This is a critical detail for longevity. Ask your Architect if they know whether the Larsen truss has been designed so both the inside and outside flanges require structural support. @Dunc Does your cladding require ventilation behind, always worth checking the fire protection detail around doors , windows and at the wall head if this is the case.

If you get something like this, get a bigger box! The spacer on this one has to be trimmed down as it won't fit when you have your stuffing glands and cabling installed and the mcb & relay shouldn't be installed next to each other due to heat! It's a very tight fit with all cables installed.

Bit of an update on this shower tray installation. Long story short but the original Mira tray I had was warped so ended up sending it back and picked up a nice Stone Resin Duplach tray which I believe are made in Spain. This tray is perfectly flat and straight so will be getting put in shortly. I have a question though about the waste trap that is suppled with it, not seen one like this and is quite shallow. No idea who makes it or what flow rate it has. No markings on it and nothing in the manufacturers literature about it. The hole for the waste is 55mm and the recess for the cover is about 72mm. I understand that European pipe sizes are different to UK ones so an adaptor is supplied to join the trap to pipe (not shown). I like the supplied rubber seal as it seals to the trap and passes through the tray flange then is bolted down with the metal disc shown. Would people be happy installing this trap or going for a more UK recognisable one, Mcalpine for example? A bit limited by the 55mm waste hole, my waste pipe is 50mm and will be solvent welded. Joist depth is 200mm and tray is sat on 18mm ply. Any recs for a suitable trap for this situation?

Wait, what!?

Took my insulation at the wall edge out as far as the inside edge of the skirting. So bottom image but wider. I would do 200mm PIR and increase screed depth.

What If Your Lighting Simulations Are Wrong? New research raises questions about photometric accuracy in industry-standard formats For more than two decades, lighting specifiers have relied on the same fundamental approach to predict how installations will perform. Lighting design software has become the trusted standard, backed by industry leaders and used daily by thousands of professionals worldwide. Photometric data flows from manufacturers into calculation engines that promise accuracy within tight tolerances. Projects worth millions of dollars proceed based on these simulations, and nobody questions the underlying assumptions — until the lights turn on and reality doesn't match the render. But a peer-reviewed study published in LEUKOS last month suggests something troubling: under certain conditions, lighting simulation tools may not be as accurate as expected, due to legacy data formats that reduce complex luminaires to simplified point-source representations. The consequences aren’t purely academic. Some designers may already be encountering unexpected dark zones or glare problems that weren’t predicted by their simulation tools. ARTICLE CONTINUES BELOW The Reach of the Problem The researchers used DIALux and RELUX software applications — widely used in Europe — to model three types of luminaires and compare results against calibrated lab measurements. While the study focuses on these two platforms, the underlying issue lies in the structure of standard photometric file formats. That raises an open question: if AGi32 and Visual in North America also rely on IES file formats with similar geometric assumptions, could comparable simulation errors be present across all major platforms? The LEUKOS paper doesn’t test those North American tools, but it does highlight fundamental limits in how traditional file formats represent spatially complex luminaires. As such, it’s reasonable — though unconfirmed — to consider whether the challenge extends across the board until enhanced data formats are more broadly adopted. A Closer Look at the Errors The research, led by Marek Mokran at Slovak University of Technology, tested three luminaire types against laboratory measurements. For compact fixtures, DIALux and RELUX delivered relatively accurate results — deviations stayed within 2–6%. But for luminaires with multiple spatially separated light sources, errors rose sharply. In one tested configuration, deviation exceeded 30% at a 1-meter offset — a scenario not uncommon in certain classroom or office settings. Even standard linear luminaires showed consistent 6% deviations at one-meter off-axis positions, which often correspond with task surfaces. To be clear: not all typical installations will suffer large errors. The most significant deviations appeared in near-field regions (within ~1 meter) and with specific luminaire geometries. At standard mounting heights around 2.5 meters — typical in offices — the study reports that deviations often fell to within 1–2% for linear fixtures. But the potential for larger errors remains, particularly when designers rely on simulations involving complex luminaires at close range. Above: Excerpt from "Simulation Inaccuracy in Lighting Design Caused by Geometric Assumptions in Luminaire Data" The Format Problem Nobody Talks About The root cause isn’t a software bug — it’s a structural limitation of widely used photometric file formats. Both the IES (.ies) and Eulumdat (.ldt) formats were developed decades ago under assumptions suited to simpler lighting technology. These formats typically assume far-field conditions and are not designed to capture the spatial complexity of modern LED luminaires with non-continuous or modular emitting areas. As a result, real-world luminaires with non-emitting gaps or physically separated modules may be flattened into a single, smoothed abstraction. The software appears to treat these as uniform surfaces, producing renderings that look more seamless than the actual luminous environment. Researchers describe this as a kind of “pseudo-near-field correction,” though the study does not confirm the specific algorithms in use. At mounting heights of 2.5 meters or below — a common scenario — this simplification can break down, especially in luminaires with more spatially distributed emitters. Whether these approximations are built into the software intentionally or emerge from file limitations remains unclear. What the Study Does (and Doesn’t) Say The study avoids alarmism. It recommends that designers “be aware of possible inaccuracies in near field regions and reflect this in the design process.” It does not suggest the entire industry is designing blindly or using tools that are universally unfit. Rather, it offers targeted evidence that in some design scenarios — particularly with non-uniform luminaires at short distances — simulation tools can deviate in meaningful ways from real-world results. It’s worth asking: are current workflows accounting for this possibility? Or are designers assuming a level of fidelity that isn’t always there? The study invites scrutiny, not panic. Questions Around Liability and Simulation Integrity While the LEUKOS study doesn’t weigh in on regulatory or legal implications, its findings raise questions. If lighting simulations lead to unexpected glare or non-uniformity, and those conditions compromise code compliance or visual comfort, where does responsibility lie? Could there be cases where narrowly passed simulations lead to real-world conditions that fall short of performance requirements? The study also mentions that luminaire definitions in data files can involve larger emitting surfaces than physically present — a practice that, if widespread, could affect UGR calculations. It stops short of suggesting intent or identifying specific manufacturers. Still, the potential for discrepancies between modeled and real-world glare may prompt designers to ask whether simulated UGR ratings can always be trusted at face value. What’s missing, researchers note, is transparency. Many software platforms don’t openly document how they handle spatial data distribution or near-field approximations. That opacity makes it hard for specifiers to assess which luminaires and which tools might be more susceptible to simulation mismatch. Where the Industry Goes from Here Enhanced photometric data formats that incorporate near-field spatial measurements already exist — but they’re not yet standard. Until that changes, designers face a decision: continue to rely on familiar workflows that may miss near-field inaccuracies, or build in validation steps to confirm that simulations match reality in critical applications. The authors of the LEUKOS study suggest a middle path: deliberate oversizing in simulation when near-field conditions apply, and empirical checks for projects where uniformity or glare control are especially sensitive. The study doesn’t claim that the sky is falling. But it does shine a spotlight on a blind spot. Whether that becomes a call to action — or just another unsolved quirk of lighting design — remains to be seen. https://inside.lighting/news/25-12/what-if-your-lighting-simulations-are-wrong

You may as well go with bottom drawing. No need for floor finish to go any further than the line of your finished wall and looks like your wall is in from the timber frame?

Hopefully a quick one: When installing the PIR floor insulation a 50mm perimeter upstand is specified. In one of the architects diagrams the upstand stops level with the screed and the final floor covering oversails this to butt up aginst the wall. In another diagram the perimeter upstand extends high enough for the floor covering to be separated from the outisde wall. As this is a larsen truss frame, the inner sole plate is effectively insulated from the outside as it doesn't span the full width of the wall. The blockwork underneath the sole plate has a course of Marmox blocks below the top block and this will overlap with the 220mm PIR, so the top block is thermally broken from the ground....so I *think* it's OK to oversail the upstand? Top to bottom: Tiles 50mm screed with UFH. Top of screed is level with the bottom of the sole plate slip membrane 220mm PIR 150mm reinforced slab.

Via the same search, but direct wording from the Emneti website. The Tecno-Varia vent valves allow for automatic air evacuation, but with the patented Tecno-Varia version and the extra manual vent, it is also possible: - monitor the valve's regular operation; - accelerate air evacuation in specific conditions (e.g., system filling). Those were my words

Thanks. Was that an AI answer, or from your own experience? Ive been testing out ChatGPT recently and found it’s not to be trusted. Confidently states things as fact, which turn out to be nonsense. It has its uses, but it’s basically a more advanced search engine imo. AI in the literal sense it is not

You really should do an image search and let AI tell you Operation: For normal automatic operation, the top cap should be left slightly open. It also includes a manual vent function, which is useful for rapidly bleeding air when filling a new system. So is left slightly open to self bleed. Side to quickly manually bleed.

I'm about to have my cast stone stooled cills fitted into the face brick exterior wall and I need to know if it's best to use 100mm DPC underneath, or a larger stepped DPC. What are the benefits if any of the stepped DPC as per my sketch? My partial fill cavity wall is 100 face brick, 50mm air gap, 100mm PIR, 100mm block. The cills are 65mm in height x 175mm deep so that i can have them set back into the cavity by 25mm, which then allows the windows to do the same. Rear of stone cill to front outer face of PIR=25mm. I planned to install a trimmed cavity closer under the cill, so a stepped angled DPC would interfere with this unless folded at 90 degrees.

I used 1000's of these from Screwfix to fix Cedar and Douglas Fir Cladding. Worked well, good value, a little delicate if you want to keep removing them. Need to be square on. Described as stainless steel construction.

I’ve not come across these AAV’s with 2 caps before. Usually just the one. Can anyone explain the function of both? cheers

Im constantly amazed how builders seen unable to grasp the concept of insulation. If air is free to circulate both sides of the insulation, clearly it cant be any use as insulation anymore. I appreciate that things can get technical quite quickly with regards moisture, condensation etc, but surely, not having the same air both sides of insulation should be understandable by even the dumbest person?

To me this makes perfect sense. But the way the OP's roof is done seems common practice, ie, just leave a void. Why? I just dont get it.

This might help if anyone else searching for stainless steel screws for chestnut cladding / decking etc in what I found to be a significant minefield with massive cost and clearly quality differentials! A good few hours down the pan! Eventually found these and placed order today - proof will be in the pudding so no comments on quality etc as yet but feeling quietly confident. So not a recommendation - let's hope it will be................ You can also speak with a normal human who knows what they are talking about! https://www.endurastainless.com/stainless-woodscrews/decking-screws---countersunk-stainless-steel---torx-drive/decking-screw-countersunk-torx-tx-drive-stainless-steel-a4-marine-grade-316 also very interesting: https://www.fastenerusa.com/blog/tell-screw-stainless-steel?srsltid=AfmBOoogzyXC-QceFYTVxyxBdILjVskzH34yHohaOgi9x-y49Rw3NaQG

Whats the insurance or safety implications of the 240v lead being run through your site? Could you use one of the big daddy transformers instead or does that not improve the amount of power taken down 1 110v lead?

Thanks all. This is all very helpful. I did some electrical engineering at school, just as an extra subject. My teacher couldn't believe how bad I was at it when i was good with other sorts of engineering. so thanks for the patience. We are supervising so can ensure that a reel is unwound. The reel, has the advantage of easy relocation, either out of the way, or to suit room arrangements as they get built. We can brief the workers too and label the ends with instructions. The point about checking the spec is good, as some give very few details. I can see that a non-reel somehow seems more credible. and could be fastened to a wall: but I'm amazed that I haven't found any ready-made. On this basis we should locate the outlet fairly centrally, then an adjacent transformer and no other sockets. shortest distances for 110 cables from there. this is £3/m cur to any length, so a reel of 25m at either £50 or £30 seems surprising. we have plenty of transformers, I think through often needing them in a hurry then back to store after use. @ProDave what type of plugs would be best for your way? Or domestic one ended , round the other? CEE seems to be the term, and they look somehow more credible to the amateur (which our demanding, but seemingly corect) joiners are in this regard. are they straightforward to connect. I've done armoured and surely anything is easy compared to that? what does one do with the multiple L and N cables? Just join them in as if it was a 3 core cable? Id like to buy the cable in blue as it is clear what it is and easier to see, but can't find any. can I just use 2 of this and join them? OR How about this for a quick fix. I buy 2 of the reels shown above. Thence they have lots of capacity and can be set in different spots. or 3, with one extending off the other to the other end of the building? Call 'silly' if it is so!

My worry would be it's 2.5mm2 copper coated aluminium. Still probably better than the 1.25mm2 copper ones mentioned earlier, but nowhere near 2.5mm2 copper.