Jeremy Harris

The specs for icynene shows that it does have a fairly high decrement delay for foam insulation, as far as I can see from the limited data available. It's certainly got a decrement delay that's maybe double that of PIR, just based on the published lambda and an assumed value for specific heat capacity based on similar spray foam materials.

You're still missing the decrement delay issue. If you take two flat structures, with an identical thermal conductivity (U value), but where one has a decrement delay that is, say twice as long as the other, then heat will take twice as long to travel from one face to the other of the high decrement delay structure. The U value is immaterial in this specific case, what is significant is the dynamic effect of the rate of change of heat transfer, when conditions change, as they do during the day as the sun's position varies, cloud cover changes or at night when radiative loss changes with cloud cover. Roof slates have virtually no effect at all, as they heat up to a high temperature in the sun and re-radiate from their rear surface - that's simple physics. That means that the surface under the roof covering still gets very hot. I have a couple of DS18B20 sensors under our PV panels (which are in-roof, with a ~50mm air space behind them) and despite the fact that the PV panels are removing around 18% of the energy and turning it into electricity (which is another good argument for using PV, it helps keep the roof cooler in sunny weather) I still see temperatures on the top surface of our sarking of over 40 deg C pretty regularly. That air space is well ventilated, too, with vents at the top and bottom to help reduce the temperature build up and ensure that the sarking boards stay nice and dry. This has nothing to do with MVHR, it's just simple physics as applied to structures and heat flow through them, by a mix of conduction and radiation in the main, although air flow does result in some external convective effects, hence the impact of wind speed, which results in accelerated convection from any exposed surface. As mentioned before, when the air temperature outside is higher than you want the air temperature inside, opening windows will just make the house warmer. Opening the windows at night, using a night purge system as is fairly common in hotter climates. does work well, though, when the night time temperature is low enough. This takes advantage of the long thermal time constant of well-designed house, by extracting heat from the interior during the cooler night time, so lowering the internal temperature, and the house then takes a long time time to warm up the next day, because the structure has a long decrement delay.

Should be OK, as there's a slight ridge in the valley mouldings to catch any rain or snow that gets blown under the slates/tiles and I can't see an obvious reason why just trimming the panel side flashing to fit shouldn't work, although it is really designed to go underneath slates/tiles, rather than over them. The only problem to solve would be how to secure the flashing neatly. With slates/tiles, they sit over the top of the flashing fixings, the same as for the Kytun dry verge, so they stay hidden.

It will, and I reckon it would be easier to laminate thin ply around an internal cylindrical mould, with a long scarf joint on the inner joint to ensure a dead smooth internal surface. I made a hovercraft duct up this way years ago and it worked OK, with the smooth inner surface true to within a couple of mm, as the fan only had around 3mm tip clearance. It'd need good quality ply, as the cheaper stuff may well have internal voids that will stop it taking an even curve, but wouldn't take long to make. Probably take longer to cut out three or four MDF discs to make the form than it would to bend and glue the ply around it. The ply needs to be held in place with ratchet straps whilst the adhesive cures/dries, and it's best to use a slow curing/drying adhesive to allow time to gradually tighten up the ply laminate. It also needs the sheets of ply to be scarf jointed together before hand, to make a strong continuous length, but that's easy to do with a long scarf jig and a plane.

Sorry, but have you looked at the impact of decrement delay, as it seems you're focussed on one aspect of performance, thermal conductivity? For a comfortable, well-insulated and airtight house decrement delay is very important. Cold storage facilities aren't designed for comfort, and decrement delay isn't at all important, but thermal conductivity and airtightness are. House are not cold storage facilities though; if they were then they could be designed and built the same way. FWIW, our upstairs rooms don't get hot at all, and they are room in roof, with vaulted ceilings, and that's because I made sure that we used a high decrement delay roof design.

How are you going to get a roofing SIP to take the bending loads without internal rafters between the skins? I asked all the SIP manufacturers about this, as we were originally going for a SIP build for our house (I spent over 6 months working with a SIP build company, and we went through loads of options), and none made roofing panels that didn't have internal rafters. Not surprising, as the shear strength of the foam core is pretty low.

There has to be a side flashing, which is black aluminium as standard, like the Kytun dry verge (we used this, it's very good). At a guess, I'd say that the side flashing is around 150mm wide, with an overlap over the edge of the tray that may be around half of that. I think it should be possible to fit the GSE trays and flashing pretty close to the verge, or you might be able to get Kytun to add a lip on the inner edge to replicate that on the GSE side flashing. The only other things to sort out would be how to raise this custom dry verge up a bit so that it sits neatly over the edge of the GSE frames, and how to resolve any interface above or below with the roofing slates/tiles.

We had this discussion a few years ago on Ebuild, before its demise, and I remember calculating the losses through the repeating thermal bridges from the required internal rafters on 190mm deep roofing SIPs, the thickest then available, and it was pretty grim. They were also a lot more expensive than other options.

It does seem to be a common failing, having really crap trades. The biggest problem, by far, seems to be the deeply engrained resistance to change, and blind refusal by so many in the construction industry, to accept new, and significantly better, ways of doing things. Adding that to the chaotic way so many businesses in this industry seem to work, and the ineptitude of so many people who call themselves tradespeople, and it seems we are doomed to keep going around this circle of woe.

SIP's as roof panels aren't a good idea, IMHO. They inevitably need internal rafters between the skins, as SIP panels are strong in compression, as walls, but nowhere near as stiff in bending, and the internal rafters tend to create repeating thermal bridges. Also, SIP insulation tends to have a pretty low decrement delay, so for room-in-roof designs there will be a tendency for those rooms to overheat more rapidly when the sun's shining on the roof, and conversely cool more quickly during a cold night. I'm not sure the importance of decrement delay is getting across well here, but it is key to making a living space comfortable; more important than the overall insulation and airtightness spec in many ways.

Looks very nice. Laminated ply might be an option for the circular lining. Not too hard to do, with a carefully scarfed joint on the inside edge to blend it in. Could even just give it a varnished or oiled finish if you use ply with a nice surface veneer. There was a curved staircase on one of the house build TV shows a few years ago that used laminated ply and it looked very nice indeed when it was finished. IIRC they used a boatbuilder to make it, as laminating curves is bread and butter work for them.

Both mine are fitted to the PRVs with about 100mm of vertical pipe to straighten up the flow. I wouldn't say they are "readily visible", as one is right in the rear corner of our services room, so you need to bend over to see it, but building control were happy when they checked it.

On the topic of pavers, we used Bradstone Infilta to meet the local authority requirement for permeable paving, under the SUDS rules. Just the same as laying ordinary pavers, except the sub-base must be Type 3, not Type 1 and instead of laying them on sand they are laid on coarse grit, with more coarse grit brushed over to fill the slightly wider joints. Not significantly more expensive than ordinary pavers, as most of the cost is in labour, and there's no more labour laying permeable pavers than there is laying normal ones (if anything, permeable ones are easier to lay, according to the chap that laid ours).

Sadly I suspect that there is a nice, cosy, relationship involving mutual back scratching going on between the SE and his "preferred" building control body, and by opting to use LABC you've undoubtedly thrown a spanner into their little profiteering racket. I opted to coordinate and take responsibility for the full plans building regs submission, and that involved collating information from two different SEs, plus ensuring that all the information needed was available either on the plans submitted, or in additional documents I provided. I did get everything cleared by LABC before we started work, so the inspections went very smoothly in the main, right up until a change of inspector right at the end (long story, but it was a personal issue with a twat, "old school", part time inspector). FWIW, I didn't find putting the building regs submission together too onerous. I'd had to read and absorb all the regs before I decided to design the house, rather than use an architect, so most of the stuff needed was pretty fresh in my mind at the time. LABC were pretty damned good I found, with only a couple of glitches, one being that they were unable to provide me with an electrical installation inspection under Part P (they are obliged to, but apologised and said they didn't have anyone competent to do it) and a question that arose over accepting Irish accreditation rather than BBA accreditation, which was solved by an email with a link to the agreement that Ireland and the UK have over accepting each others accreditation.

I bought some really cheap little thermometers on eBay. Not super accurate, but good enough, and they come with a remote probe that's waterproof, so you can stick it under a running tap (and wait, as they only sample the probe every 30 seconds or so) or you can tape the probe to a pipe, under a bit of pipe insulation. The batteries seem to last a couple of years. These are the ones I have scattered around the place, I only really used them when setting things up; they are cheap enough to be considered disposable, really, at £0.99 each from China: https://www.ebay.co.uk/itm/Digital-LCD-Fish-Tank-Aquarium-Thermometer-With-Waterproof-Probe-Black-631F/123254839016?epid=2292557761&hash=item1cb28f96e8:g:-3cAAOSwg0hbXPzc:rk:2:pf:1&frcectupt=true If you search for aquarium thermometer or fish tank thermometer you will find others that are similar, and some from UK sellers that are a bit more expensive.

The safety concern isn't being able to see the tundish, it's the air gap at the tundish that allows an emergency release of pressure if the pipe D2 gets blocked, so preventing a cylinder explosion under severe fault conditions. Being able to see the tundish easily is nice, but not at all essential.

I can say with certainty that G4 will absorb a fair bit of moisture when it cures, plus it will form an excellent moisture barrier that will 100% prevent any slight remnants of moisture in the substrate from reaching the GRP. The only thing to watch is to make sure that the third coat of G4 isn't allowed to fully cure before you start to apply the first coat of resin. The G4 needs to be allowed to cure until it's just tacky on the surface, then roller the first coat of resin on. Not only does this ensure a good bond, but it also allows the final coat of G4 to absorb any slight trace of moisture that might remain on the surface from condensation, and will draw moisture out of the styrene in the resin, ensuring that the critical first coat remains moisture-free.

Good move. Nothing should leak if the PRV remains sealed. That photo isn't of a faulty tundish, it's evidence of a leak above the tundish that's been dripping water over it for a long time.

The snag is that it may take a year or two for the GRP to start breaking down internally from trapped moisture, so a quick test may well not do very much. If there's enough moisture around a quick test will show the problem, as the first coat of resin will take a long time to partially cure and will remain a bit like a gel for a long time, but it's hard to judge this unless you've done a lot of polyester GRP work and seen what moisture curing inhibition looks like. Some resins will blush if there's too much moisture present, which is a good clue, but it's not a guarantee that it is all OK, unfortunately.

The problem is as @Nickfromwales has said, the outside drain, D2, can freeze, get blocked with snow or whatever and then there is no where for the explosive pressure build up to go, if something like the boiler stat fails and the tank overheats. A leaking PRV should be fixed as soon as the leak is spotted, not "fixed" by taking the tundish out so the leak isn't visible any more.

Absolutely not, I'm afraid, as PVA has a lot of water in it, which then has to dry out as well as any residual moisture in the substrate, so it makes the problem potentially worse, by partially sealing damp in. GRP doesn't adhere that well to it either, as PVA is slightly moisture permeable when it's dried, so moisture trapped underneath (because the PVA only dries from the exposed surface) will gradually migrate through it to the GRP. Using a moisture-curing primer will help to draw moisture out of the substrate as it cures (and G4 cures, rather than dries), which is one reason G4 is used such a lot as a primer before applying GRP.

I hate to say this, but if someone is injured as a consequence of this fault, then your insurance is almost worthless, other than perhaps covering your legal fees (and that's debatable). You cannot insure against an unlawful act, and if someone was hurt the charge would most probably be criminal negligence.

My part-time job is investigation work for insurance companies, in the main. Much of it hinges on liability. One thing I can say for sure, with no doubt whatsoever, is that now you know that the installation is non-compliant with the safety regulations you will be liable. The installer may well be able to mitigate the case against him (assuming something goes wrong) by claiming that the client asked him to remove/not fit an essential, critical, safety device. What's more, as a landlord you have a higher duty of care than a house owner, so the chances are that any court will consider that you carry an enhanced liability, even before the evidence that you knew about the non-compliance is presented. The bizarre thing here is that if you had no knowledge at all of the regulations, and would not have been able to tell from just looking at the installation that there was anything wrong with it, you would have had a good argument for mitigating circumstances, and it would have been the installer that would have been almost wholly liable. Bear in mind that this non-compliance will be uncovered in a years time when the annual G3 check is due, too, so at best you're only putting this off until then.

I think I'd just set it up for what you want, including delivering DHW at the temperature you need, run it through the winter until the end of the heating season and see what the running costs look like. You can do as I've done and fit a small energy meter in the ASHP supply if you wish (no more than £20 for a meter and small enclosure to fit it, and a DIY wiring job if you fit it downstream of the ASHP isolating switch), as that then tells you exactly how much electricity the thing has used. Tracking this will give you a feel for the running cost and you can then make a decision as to whether to make things more complex or not, based on how acceptable those costs are.

Three coats of G4 may do the job. I've used it to prime concrete a few times and it's moisture-curing, so it may well be OK for a roof. CFS say it's OK as a wood primer: https://www.cfsnet.co.uk/acatalog/G4_Primer_Sealer.html