Jeremy Harris

The Sunamp PV doesn't cycle the heater, it has a pump that circulates water through a primary charging circuit to the heat exchangers inside the cells and the heating element relay is on all the time that PV power is detected. The pump controls allow for an over-run when the heater turns off, if there is still power to the unit from the always-on supply. The snag is that if the always-on power to the unit fails, then the pump doesn't over-run to cool the heating element and it gets very hot from heat soak, which can then trip the over-temperature cut-out switch. The other snag with the Sunamp PV units is that they have an ultrasonic flow sensor, that can be upset if there is any air in the system, which causes the unit to shut down. Once the air is out of the unit and it's been running for a while, it's unlikely that air will cause any problems, but it can during commissioning I found.

The 2800 w is an EU rating anomaly I think, 2 x 2800 W at 230 VAC, is 2 x 3047 W at 240 VAC, and 240 VAC is still our nominal mains supply voltage, despite it being officially labelled as 230 VAC. We fudged EU compliance by creating an asymmetric supply voltage tolerance, of +10%, -6% on the nominal 230 VAC that the EU demanded, so we didn't have to go around changing all the local distribution transformers or adjust their taps. When calculating cable sizes it's better (and safer) to ignore the 230 VAC power rating for resistive loads like heating elements and assume that the supply is still really 240 VAC, as most of the time it is. This makes a small, but significant, difference to the current drawn by any resistive load, plus it's sensible to allow for the current drawn by the heating element under worst case conditions, especially in an installation where there is PV fitted, as that will tend to increase the local supply voltage above the nominal 240 VAC. Our supply often limits at the 253 VAC upper tolerance voltage when it's sunny, for example, as all three houses on our supply phase from the local transformer have PV systems. The 2800 W at 230 VAC rated element in the Sunamp has a nominal hot resistance of around 18.9 ohms, so if supplied with the normal 240 VAC supply voltage it will draw around 12.7 A and if supplied with the maximum supply voltage (say when a PV system is running well) then the current drawn increases to about 13.4 A. The 2800 W at 230 VAC element will run at around 3386 W at the maximum supply voltage of 253 VAC, or around 3047 W at 240 VAC (so the heating element is, in reality, a 3 kW one). If it were me (assuming this is an all RCBO set up, with no RCDs in the box), then I'd just fit a 32 A RCBO and a run of 4mm² T&E to a small CU with two 16 A DP MCBs near the Sunamp installation as the isolators. It's then fine to run 2.5mm² cable to the Sunamps from there.

Firstly, the RCBO is primarily there to protect the cable running from it, so needs to be a lower current than the maximum rating for the cable, allowing for derating factors that are determined by the mounting and location of the cable. A ring final is not a good idea, or normal practice, for a heating circuit, it needs to be a radial. Two immersion heaters will draw less than 30 A in practice, under any condition, but 30 A is above the rating for 2.5mm² T&E cable when clipped directly (maximum is 27 A if there is no insulation around the cable). 4mm² T&E is rated at 37 A when clipped directly (installation type C), but that drops to 27 A if it is behind plasterboard that has insulation against the cable (mounting type 100), so before determining the cable size the mounting method needs to be known. If you can run the cable using either type A, B or C mounting, (in conduit, enclosed in conduit in a wall or clipped directly), then 4mm² is OK, but the RCBO needs to be 32 A, not 40 A, as 40 A isn't a low enough rating to protect the cable.

Frankly it's worryingly easy to get the paper ticket needed, without any real on-the-job experience. It's no wonder there are a few "qualified" people around who are, on paper, "competent", but who in reality aren't up to the job. I'd rather have someone who's served a proper apprenticeship, and got lots of experience of working on a wide range of jobs, than someone who's just done a short course at a technical college and come away with enough knowledge to pass the exams. One big problem is that the accreditation bodies are just businesses, keen to recruit more people so they can increase their profits, which has led to the introduction of shorter "training" periods and exams that can be passed by those without any real in-depth experience. Back when I used to teach electrical engineering science to apprentice electricians, all of them were doing a three year apprenticeship and doing their exam work on day release. They learned far more on the four days of the week they were at work then on the one day a week they spent in college. Even in college around 50% of the course was practical work in the labs, where we had just about every wiring installation system they would ever come across mocked up, so they could put the theory learned in the classrooms into practice.

I doubt you'd get a reliable joint, so would suggest you just GRP the whole area, to make one integral, watertight, composite roof with no joints. Joints are always going to be the Achilles heel of any waterproofing system, and the really big advantage of GRP is that there are no joints.

You can't mix solutions, as that's a recipe for failure. If going for GRP, then create a single, integral, GRP "tub" over the entire roof area, including the edges and upstands, with no breaks or joins.

Yes, in the link I gave earlier there's even one that will take gravel on top, from the look of it: http://www.duracomposites.com/industrial/fibreglass-grp-grating/micro-mesh-grating-drainage-grating-walkways-grating-catwalk-grating/ Mounts on pillars, too:

You could have a look at using the steel grating in place of the pavers. Some of them look pretty good, and it would be one less expense. There are systems made that have only narrow slots, that can even be walked on in stilettos, should you be minded to...

Properly filtered rainwater is OK for toilet flushing, but I'd not want to use UK rainwater for clothes washing, mainly because lots of wash cycles are too low a temperature to kill off the faecal coliforms that will be in the water from the inevitable bird poo that gets into it.

You could probably raise it up on high density concrete blocks, or brick piers, if the thicknesses work out.

I've got a 30+ year old galvanised anchor that suggests otherwise! Even my old galvanised mooring chains, would wear before they corroded out. I think there is a problem with a lot of supposedly galvanised steel not being proper hot dip galvanised, which may well be a part of the problem, but we had galvanised grating walkways running over concrete ducts at work that had been there thirty years or more and still looked OK. Failing that, the Duracomposites stuff I linked to would probably last forever. IIRC, it's very similar to the stuff used on oil rigs.

Looks ideal, especially with that high load rating. Probably a good idea to liberally grease it when fitting, although it shouldn't get wet under a disc of ply. Just need to make some sort of pedestal to raise the whole thing up to a good working height.

Something like this 300mm diameter bearing might do it, supporting a disc of thick ply that would take the greenhouse: https://www.ebay.co.uk/itm/12-Heavy-Duty-Lazy-Susan-Rotating-Swivel-Turntable-Bearing-Hold-30kg-Desk-Table/173292001547?_trkparms=aid%3D555018%26algo%3DPL.SIM%26ao%3D1%26asc%3D54618%26meid%3De2dedc85fc4c4534a94a7ab652b63b33%26pid%3D100005%26rk%3D5%26rkt%3D12%26sd%3D222648701219%26itm%3D173292001547&_trksid=p2047675.c100005.m1851 Or, if 30kg isn't a high enough load capacity there are some bigger ones here that will take up to 120kg: https://www.ebay.co.uk/itm/Round-Rotating-Tray-Bearing-Turntable-Turn-Table-Heavy-Duty-Swivel-Plate/142972195351?hash=item2149ce9217:m:m6Ms5uEO8tC4t1STyL9TEaQ:rk:18:pf:0

I agree with him, GRP would make the detailing around the awkward areas at the edges and upstands a lot simpler to implement, and likely be significantly more reliable in the long term. I also think that the idea of a galvanised steel subframe, that's free draining and which supports the pavers is a good one. You can then just ensure that the GRP "bathtub" underneath is properly drained, so water runs off easily, so reducing the risk of any water getting close to the junction of the roof lights and the top of their upstands. At a guess, standard walk way galvanised grating, like this, may well do the job, fitted to stand off legs with load spreading plates underneath them: https://www.themetalstore.co.uk/products/galvanised-steel-walkway You could opt to just leave the grating exposed, even, or go for a GRP grating like this stuff: http://www.duracomposites.com/industrial/fibreglass-grp-grating/

I thought the EPDM was over cement board and then the concrete was over the top of that? Anyway, GRP can be laid over concrete, just needs the concrete to be primed with G4 first.

We have loads of 75mm OD MVHR ducting going through Posijoists. The vast majority of our ventilation ducts run through them. From a DIY perspective, the 75mm semi-rigid ducting has to be probably the easiest stuff to use.

No problem, it has significantly better load bearing properties than EPDM.

EPDM and glass fibre laminate are not that different in terms of their coefficient of linear thermal expansion, plus I doubt there would be much of a temperature change underneath that layer of reinforced concrete and pavers, anyway.

That issue of compressibility, together with the greater robustness of GRP, was what had me puzzled from the start. I will admit to being a fan of GRP roofing, having done a couple and seen just how tough and durable it is.

Didn't we have a discussion about using GRP for this some time ago? GRP certainly has a lot of advantages, especially when it comes to getting a really robust seal around those upstands, and at the more complex edge detailing, like that by the gate/entrance, but it does need dry weather, and ideally needs to be done when the air temperature is reasonably warm, and consistently so for the whole of the cure time. There's no doubt that a good GRP roof will be a lot tougher and take a fair bit more abuse than an EPDM roof, though, plus it's easier to patch repair any small areas that may get damaged.

And don't try hot wire cutting PIR or PUR, only EPS or XPS. PIR and PUR don't cut with a hot wire, plus the fumes given off it they did would be very unpleasant.

In 1990 there would have been no specific ventilation requirement, an opening window would have been OK. Same goes for sound insulation, and the integral garage would have had a ceiling to the remainder of the house (as a garage) that exceeded the requirement for a habitable room. Wall insulation wasn't a requirement in 1990 either, the cavity wall built 3 years earlier when the house was built would meet the regs. I'm pretty sure that the only elements that would need approval would be the fabric of the wall blocking off the old garage door (cavity wall as per the rest of the building would be required) and the means of escape from fire (an approved fire escape window with a low threshold to the floor internally, or an alternative escape route that doesn't involved going via another habitable space).

As a guide, I fitted unswitched outlets behind our built in ovens, but fitted a pair of isolating switches in the adjacent cupboard. Not hard to do. For the boiling water tap, where I wanted easy access to the switch, I made up an extension lead that runs behind the units (from the dishwasher double gang outlet that's accessed via a hole in the top rear of an adjacent unit) and that runs to a double pole switched fused connection unit in the cupboard under the sink (high up, near the front) that supplies the boiling water tap. This keeps things adequately accessible, whilst being relatively unobtrusive, and remains compliant with the regs. BTW, if you feel competent you can now legally modify an existing circuit in a kitchen, without needing a Part P sign off.

Yes, there has to be an accessible switch or circuit breaker for built in appliances. It's widely ignored, but not hard to comply with. For our built in cooker and microwave I added switches in the cupboard next to them to turn them off. For the dishwasher, I cut a hole in the top rear of the adjacent cupboard to access the switch and socket. It's extremely useful to do this, as it makes any future electrical installation inspection a great deal less hassle. Anything that allows easy access to outlets and switches generally seems to make life a lot easier. As above, the requirement to have easy access to switches for all appliances is very commonly ignored, though. I've seen built in ovens fitted with a normal switched outlet tucked right behind it, with the oven screwed in to the unit preventing access. One has to wonder how the power is supposed to be turned off quickly in the event of a fault.

The key is, in part, the date of the conversion. Regs have changed a great deal in the past 15 years or so, so conversion undertaken some time after the original build will need to show that it complied not with the original regulations at the time the house was built, but those that applied at the date of conversion. Examples would be the Part M changes that altered the minimum acceptable height of wall sockets and set the maximum height of wall switches. Other compliance issuse would be means of escape in the even of fire. That is a regulation that doesn't apply to a garage, but does to a habitable room. If you can prove that the conversion too place in 1990, with evidence that will satisfy the solicitor for the purchaser, then the only building regs requirement that I think you need to prove compliance with is the fire regs, specifically the means of escape. Is the window a regulation size escape window, with the lower edge of the specified minimum escape pane no higher than 1100mm above the floor? Is there another safe means of escape directly, via a single door, either the outside or to a communal area with a single fire exit door?