Jeremy Harris

If I were doing a build again, I'd forget about ring finals, and run radials for all power outlets, probably grouped as one radial per room (subject to the limit on the number of outlets), maybe two or three radials for the kitchen (just because kitchens need a lot of outlets and may well have several fairly high peak load appliances in use). I'd go for the biggest physical CU I could fit in the available space, as there is nothing worse than trying to work in a CU that's packed full of wires. I'd also fit compact DP RCBOs on every circuit, rather than a split board. I've recently made up a sub-board using these single module DP RCBOs, and, apart from the need to loop in the L and N supplies, rather than use bus bars, they make for a neat installation: https://www.electricalcomponentsdirect.co.uk/acatalog/Compact_Mini_RCBO_Circuit_Breakers.html

A development just up the road from us has internally fitted bat boxes, that are sealed from the main loft space. There's a name for them (may be a trade name), I'll see if I can dig it out.

In practice any additional, pretty small, potential heat loss is almost all due to the increase in volume of heated air, and results from the additional ventilation heat loss. The fabric insulation can be fairly easily brought up to a level where the loss from the increased surface area is small enough to not worry about too much. Ventilation heat loss can be mitigated by using MVHR, plus there's no need to heavily ventilated the warm loft void, anyway, as there's no moisture or CO2 generation up there, and diffusion should mean that the warm void ends up at equilibrium with the rest of the house.

1 is a warm roof, though, where the void is sealed inside the heated envelope.

From a simple physics perspective, prevention means: 1. Preventing the temperature inside the void dropping to the local dew point, or: 2. Preventing the humidity of the air in a cold void rising to a level where moisture could condense out under the range of temperatures it might be exposed to, or: 3. Having no void at all 1 is relatively easy to achieve, 2 is near impossible to practically achieve long term I suspect, 3 is pretty much the same as 1 in practice. Hard to see how any other option than 1 is likely to work reliably under all anticipated UK environmental conditions, for the expected life of the building.

I opted to use armaflex pipe insulation primarily because it goes around bends easily and generally doesn't need taping in place much. The only downside with it (apart from the price) is that it can be fiddly to remove the backing from the self-adhesive areas in tight spots. The adhesive seems incredibly sticky, too, so you don't want to remove the backing until the stuff is neatly in place.

Yes, IIRC I mentioned this over on the other place around 10 or more years ago, and I think we had a discussion there about it. I did the work to improve the old house around 17 years ago now, which is when the condensation problem became apparent. I suspect I only spotted it because I'd opted to use the perforated plastic bag wrapped fibreglass, if I'd laid unwrapped stuff the condensation would have ended up soaking into the insulation, and not being visible until it stained the ceiling, perhaps.

A significant part of the problem may relate to the amount of energy available to allow condensed moisture to phase change to vapour and then move out of the structure by ventilation. Increasing ventilation may increase the amount of heat energy available to enable the water to change phase, or it may not. It's a variation on the dynamic interstitial condensation risk that may impact other areas of construction. It doesn't need to be a new build, either. I opted to improve the airtightness and loft insulation level in our old house. I sealed up most of the tiny spaces that were leaking air into the loft space (made up a home made blower to find them), and, because I don't like handling fibreglass much, opted to lay an additional 200mm layer using the stuff that comes wrapped in a thin, perforated, plastic bag. In theory, the perforated bag is vapour permeable, but in practice it is not liquid water permeable (the holes are too small, and surface tension stops water flowing through them). After a while, I started seeing puddles of moisture on top of the insulation. I thought it must have dripped from the underside of the sarking, but it was clear that it hadn't, the upper surface of the insulation had just dropped below dew point and water had condensed out. After measuring the humidity and temperature up there I reckon that the problem was that I'd unwittingly created a very effective one way dew trap. The period during which the top surface of the insulation was below the local dew point was longer in winter than the period in which the same surface was above the dew point, for any 24 hour period. The result was that more and more moisture collected. To make things worse, as puddles formed they tended to stay cold for longer, because of the relatively high heat capacity of water, so they then needed more and more energy in order to evaporate. I fixed it by adding loads more soffit vents, and by stuffing short lengths of 32mm waste pipe into the sarking felt overlaps between every truss to provide a bit more ventilation. I'm not convinced that it was the increased ventilation alone that was the fix, but that increasing the ventilation allowed the loft space to warm up more quickly during the day, perhaps. The problem may well have been exacerbated by the roof ridge running more or less North/South, so the roof pitches didn't get much energy from the sun during winter.

So am I, if for no other reason than the fact that Tesla's attract geeks who like nothing better than to find flaws in their systems. The fact that the car runs on Linux, though, and that much (supposedly all) of the code is open source means that there are a lot of people playing around, amongst both the white and black hat fraternities, so as long as the white hats are better than the black hats things should be OK. As an aside, I've just finished getting a Raspberry Pi Zero W to work as a USB drive, that will automatically dump TeslaCam files (both DashCam and SentryMode) to my home file server as soon as the car is within range of our wifi network. It also stores all my music files, and keeps them synced to a directory on the file server, so the car will always have an up to date version of my music library onboard. Premature, really, as I don't get to test it until I get the car on Friday...

That's how Tesla's lock, unlock and start now. They detect BT at a range of about a metre, and unlock, set the driver profile, etc, so by the time you've opened the door and climbed in everything has booted and is running. They do the reverse when you get out, detect that your BT device is outside the vehicle, shut everything down and lock the doors. This needed a BT enabled device ('phone, tablet etc) either Apple or Android, but they've now released a hands-free remote that does the same thing. It seems to just have a very low power BT device that works the same way as a 'phone, with the same sort of encryption, that's just paired to the car, like any other BT device. Supposed to be secure and immune from a relay attack, although whether this remains true is anyone's guess, knowing how good hackers are at busting into stuff like this.

I went to see a demo of screw piles being put in, as the first plot we looked at needed them to avoid damage to underlying archaeology (strictly speaking, it was a cost mitigation measure). Very impressive to watch piles being driven in with nothing more than a big hydraulic "screwdriver" fitted to the jib of a digger, in place of the bucket. It took more time getting the pile set up dead square initially, and in the right spot, than it did to drive it to depth. IIRC, the initial price we were quoted was way less than £1000 per pile, too, but this was back in 2012.

We bought a house in the late 1990s that had the loft converted into a large bedroom, complete with dormer window looking out over the rear garden. The survey revealed that this was boarded out with hardboard (covered with wallpaper), that the electrics were definitely DIY and dodgy and suggested that our solicitor checked the position wrg to planning and building regs. There was no record of either having been obtained. The vendor sorted out the planning with a lawful development certificate, we refused to accept this was a habitable room, and the vendors estate agents negotiated with their client to change the description of the house so this was no longer a bedroom, together with a price reduction. I was planning to rewire the house, fit a new kitchen, bathroom etc, anyway, so we went ahead and bought it. I never got around to making the loft room habitable, but this didn't matter in the end as we sold the house to an architect who was going to knock the house down and build another one on the plot... I would guess that stuff like this probably only comes to light when a house gets sold most of the time. As above, I just doubt there are the resources to go around looking to enforce breaches of building regs. In the case of the house we bought, the uninsulated loft bedroom had been used for two children, who'd since grown up and left home, so it had been like that for well over a decade.

The spec for our ASHP gave the dimensions for the concrete base, with two options, either two cast concrete piers or one larger, single, one. It so happened that the dimensions of each pier in the two pier option was very close to the size of a block laid on it's side, so the base for ours is made up from two piers, each consisting of two blocks laid on their side, bedded on a concrete foundation. I don't think you'd be able to secure it to paving slabs for a couple of reasons. Firstly, the ASHP really needs to be raised above ground level a bit, both to allow easy access to the condensate drain connection underneath (this is easier with the thing on two piers) and also because it may tend to be noisier if mounted on slabs (although fitting a rubber anti-vibration mount under each leg might make that work OK). I find that, at this time of the year, the ASHP tends to draw dead leaves in around it, so having a decent gap underneath it to sweep this area clear is handy. If doing it again I think I would opt to raise the ASHP up a bit more, maybe using three, or even four, blocks per side, mainly to make cleaning around it easier.

I believe these are supposed to go at the top of a stand pipe, with a trap at the base. There needs to be a length of vertical pipe below this fitting, and before the trap, to allow them to work correctly, I think.

We have MVHR. I can say with confidence that even with no cooker hood we get no cooking smells outside the kitchen, ever. What smells are created when cooking dissipate quickly as the air gets changed via the MVHR. If we feel there may be a smell problem we just hit the boost button to clear the smell more quickly, but most of the time we don't bother. Right now there's cauliflower and broccoli being cooked in the kitchen, and the door to the kitchen is partially open and around 6 feet from where I'm sitting. There is no smell here at all, absolutely nothing, and the MVHR isn't even on boost. I did put provision in the ceiling to fit a hood, both wiring and a bit of thick plywood behind the plasterboard as a strong point, but I never got around to installing one, as it seems it's not needed. Came as a bit of a surprise, TBH, as in our old house we were plagued with cooking smells permeating the whole house, even with an extractor hood above the hob, that was ducted directly outside.

There's a lot of misreporting re: Tesla. There was never, ever, any plan to locate the European Gigafactory in the UK. Tesla never even looked at the option, or had any discussions with anyone in the UK about siting a Gigafactory here. It's been a badly kept secret that the European Gigafactory was going to be located in Germany for months now. The decision Tesla announced this week regarding the UK was about the siting of the European R&D facility, not the European Gigafactory. There was a slim chance that the UK might be chosen for the location of the R&D facility, but once it became clear that (largely for logistic reasons, plus, I suspect, some incentives) Tesla were going to locate the European Gigafactory in Germany, it made a lot of sense for them to also locate the R&D facility there too. This really should be no suprise, as the UK just isn't a volume car exporting nation. Germany exported over $260 billion worth of cars last year, the UK exported less than $60 billion. Bit of a no-brainer to locate a factory somewhere with very good transport links (no ferries, for example) if looking for a central European manufacturing hub. The fact that Germany also has some of the worlds finest automotive robot manufacturing companies, and that Tesla Gigafactories are more automated then any other car manufacturer, was probably a pretty significant factor too. I'm sure that Brexit uncertainty was (and still is) on the Tesla risk register, but I'm also pretty sure that part of the deal with Germany will have included the wording of the press statements. It's very clearly in Germany's interest to make the point that Brexit played a part in the decision, even if we all know full well that, at least as far as the big bit goes, the location of the European Gigafactory, it wasn't a significant consideration.

FWIW, the instructions for our unit indicated that fan speed 1 was only for use when the house was unoccupied, speed 2 was the normal background ventilation rate, speed 3 is the normal boost speed (humidity controlled in our case) and speed 4 is 100%, used infrequently for purge ventilation. I adjusted speed 1 right down to about 10%, enough to keep things fresh with no one at home, handy for use when we go away. I initially set speed 2 to meet building regs, then backed it off later, so that it runs at around 0.4 ACH (seems about right for us). Speed 2 is something like 30% speed, I think. Speed 3 is set to 75%, which seems fine for normal boost. I think we've only used speed 4 once, when we didn't manage to turn the MVHR off when someone nearby had a bonfire, when the extra ventilation after the bonfire smoke had died down helped to clear the air in the house quickly.

I have the Salus auto balancing actuators. Best thing since sliced bread in my view. They just work perfectly, quicker to actuate than the wax ones, ensure the flow to the UFH is maximum when the system first switches on, for a fast warm-up, and then they automatically throttle back to maintain a constant temperature difference between flow and return. This temperature difference depends on the flow temperature. Below a flow temperature of (I think) 30°C they maintain a 4°C differential, above that I think they maintain a 7°C differential. We keep our flow below 30°C, as the 4°C differential seems to work very well for our system. I had ordinary wax actuators originally, but these were really slow to open, and meant that the system had to be balanced using the flow control valves, a fiddly and time consuming process to get right. We initially had problems with our kitchen getting too hot, that took ages to try and get sorted by adjustment, and when I did get the adjustment right the kitchen then took ages to warm up if the heating had been off. The Salus actuators fixed that, with no manual adjustments being needed at all. Also quick and easy to fit. The only downside is that they are a lot more expensive than conventional wax actuators.

The issue here is that initially this company was promoting a product that didn't work in the way claimed at all. Caused a stir and a lot of suspicion, as they had claimed that the multifoil element could provide far greater levels of insulation than it can in practice, and consequently there was a fair bit of mistrust (well-founded, given the chequered history of the introduction of multifoil products, with their ludicrous and easily disproven claims). They changed the product completely, so that now it works almost entirely as a conventional insulation material, the foil layers do very little, other than, possibly, reducing air movement a little bit. There's no long wavelength IR to be worth being bothered about inside the fabric of the structure, so the foil is now really only there to differentiate this product from competitors. It's performs at about the same level as most other fluffy-type blanket insulation materials, and they are quoting a sensible value for λ of 0.033 W/m, pretty typical of a fibre-type insulation (rock wool is around 0.032 W/m.K, wood fibre is around 0.037 W/m.K, PIR foam is around 0.022 W/m.K) . It's no better, or worse, than most other fluffy type insulation, but that's wholly down to the honeycomb fluffy stuff, not the foil layers. It does have the advantage all fluffy blanket type insulation has of being relatively easy to fit well between frames, and an additional advantage in that it is as clean and comfortable to use as wood fibre (so much nicer than rock wool). The downside is that it has a pretty short decrement delay when compared to wood fibre, or even rock wool.

Be interesting to find out how much functionality a smart charge point has with no connectivity. Hopefully there may be some ability for it to work as a basic charge point without any connectivity, but it could be a bit of a nuisance if it ends up being disabled if it cannot connect. Some of the more basic smart charge points may be a better bet than something more sophisticated, as I get the feeling that the basic models may have had the required smart functionality installed as an add-on to an existing design. If so, then it seems likely that they may well work with no connectivity without a hitch. One option is to just get a commando outlet installed, then use a portable charge point. Some can charge at 32A (the Tesla UMC allows this) and a portable charge point is a useful thing to have (although most EVs come with one, normally limited to 10 A maximum). You can buy a portable charge point that will deliver up to 32 A and plugs into a commando outlet, for about £250 or so (I've seen cheaper ones around, just can't find them right now). Wiring in a 32 A commando might not be a very costly job.

A smart charge point usually connects via wifi, although I believe some can (rather dodgy in my view) accept an Ethernet cable. If there's no connection, then there may be be no way to turn the charge point on and off, schedule off-peak charging, etc, as I believe it's managed via an app. I have concerns as to whether security has been adequately addressed with smart charge points, TBH. At least one system is so deeply flawed that it created a bit of a stir when this was exposed (although I believe it's being addressed now, although this isn't much consolation to the few thousand who already have the Raspberry Pi version): https://www.pentestpartners.com/security-blog/pwning-a-smart-car-charger-building-a-botnet/

No it doesn't, but a condition of the OLEV grant is that you have to have a charge point installed that does have external data gathering and charge control capability, much like a smart meter. You can't get a grant-funded installation without this feature. My charge point has a (manually switchable) option to charge when there is excess PV available, but the rather limited J1772 signalling protocol used for both a safety interlock and signalling maximum available charge current is a restriction. The minimum charge current is 6 A (and most cars are very inefficient when charging at 6 A), so any PV system needs to be generating in excess of this before car charging can start. My implementation takes a pragmatic view that partial charging from excess PV is always beneficial if the cost of charging is lower than the off-peak rate. I used to adjust the charge current on the fly to allow for this, but that was flawed, because the car onboard charger was both slow to respond and had a fairly long latency (it didn't respond quickly to the control pilot increasing duty cycle to signal that more current was available). I've ended up with a fixed solar charge current setting of 12 A, together with a bit of code that keeps the car charging for as long as the cost is less than 8p/kWh. If the charge cost stays above 8p/kWh for 20 minutes, then the car stops charging until either the PV starts to generate more, or until the next off-peak time period.

I've DIY'd three charge point installations now, including building the charge points themselves. The OLEV grant only applies to "smart" charge point installations, and like all subsidies it seems that the installers are absorbing the grant money as a bit of additional profit. The grant is 75% of the installed cost up to a maximum grant of £500, and guess what? No installers claim less than the full £500... There are a few charge points around for ~£200 to £250 or so that work just fine. On top of that there will be the cost of a run of 6mm² cable (maybe 10mm² if it's a long distance), plus an enclosure with a Type A RCBO, a couple of henlys and lengths of tails cable. Not wise to connect a charge point to a spare way in an CU, IMHO, as it will draw 32 A for several hours on end, and things may get a bit warm in there. Another thing to consider is whether you want to install a tethered charge point or a non-tethered one. Tethered ones are easier to use, just unhook the cable and plug in, but they look a bit uglier, as there will be a few metres of hefty cable coiled up around the thing. A socketed unit means getting the lead out of the car every time you want to charge, and plugging it in both ends, but does tend to look a bit neater. Personally I'm not a fan of smart charge points, for a few reasons. Firstly, on a point of principle I object to data being collected about my car charging habits, for several reasons, not least of which being that a failure to charge for a few days would be a reliable indicator that I'm not at home. Secondly, users of various makes of smart chargers are reporting that, like pretty much all app controlled kit, reliability can be a bit flaky. Tales about of cars not charging as expected, charge rates being randomly reduced, user interfaces falling over and some charge points refusing to play nicely with some devices. The bottom line is that a charge point is just a switched power outlet, and doesn't need anything other than a means to turn it on and off. I prefer being able to do this easily and reliably with a switch, as it then needs no 'phone, internet connectivity, sanction by whatever authority controls the smart features, etc in order to work. I want to be sure that my car charges when I want it to, not when someone else decides I'm allowed to.

The only caution I'd raise with using bales would be rodents. Rats and mice really love straw; within hours of stacking a barn the little buggers would start moving in en masse.

May be, I've seen a few odd things happen, but nothing quite as clear cut as this. Hopefully you can just file the corners off the pegs to get them to fit.