Jeremy Harris

Yes, although the main issue is when there is an imbalance between the phases, caused either by one phase feeding a greater load, or, perhaps more likely in this scenario, harmonic distortion causing a high neutral current in combination with a high N to PE impedance. When there is a combination of an imbalance and a high impedance in the N, the voltage on the least heavily loaded phase can approach the phase to phase voltage if things get really bad. This would, perhaps, explain why the phase feeding the house seems to have had a major over-voltage event. Checking Ze on the incomer should determine whether there is a DNO side problem and checking Zs at the consumer side, perhaps at the ASHP (as I strongly suspect that may be the heaviest load) should show whether there is a N to PE impedance problem somewhere else in the installation. On its own, having an imbalance in a 3 phase system isn't really a problem as far as the consumer is concerned, as all it does is lead to a higher neutral current in a mixed 3 phase/single phase installation. If the installation is correctly wired and has no faults then there shouldn't be a problem, as the distribution network is sized to deal with phase current imbalance and the consequent current flow in the neutral.

True, my suspicious nature leads me to suspect this is probably a fault on the consumer side, though, given that the incomer can only be around five years old.

Do they have a copy of the EIC/Part P certificate that was lodged with the building control body? They can get this from the accreditation body for the electrician/company that wired the house and it may be useful, in as much as it will give some of the measured values from the installation testing. It's not comprehensive, in that not every measured value is recorded, but key ones are, including Ze and the highest Zs for each circuit, and as the house is 5 years old one would assume that there is still a warranty of some kind. If it turns out to be a long standing fault in the installation, then it may well be possible to try and claim on the warranty (although frankly I suspect that's a lost cause).

Too true, and that's despite every one having a label on telling the installer to check the tightness and, I very strongly suspect, loose terminals being the cause of a fair few CU-related fires, hence the amendment 3 metal cases. The latter seems to be treating the symptom, not the cause, to me.

Nothing surprises me any more, either, particularly with relatively new builds.

OK, PME = TN-C-S in this case (strictly speaking it's not exactly the same, but PME is often used to describe what's really a TN-C-S installation). No RCD on the 3 phase supply is another clue, as if there had been, and if N really is 30 VAC above PE (Protective Earth), then the RCD would certainly have tripped long before the fault condition got to the point where equipment has been damaged. Number one priority would seem to be to measure Ze, to see if there's an earth loop fault. If the N isn't sitting at the same potential as E, then it sounds as if there could be a problem with the PEN in the head (Protective Earth and Neutral) or the CPC (Circuit Protective Conductor) from there to the installation PE. Normally, the PE are N and connected together (hence the term PEN) at the head, and the incoming N conductor is also the PE conductor. There will probably be other earth paths within the installation, particularly from things like the ASHP, where the chances are that its earthed metal case will be bolted down to ground. However, the chances are that the other incidental earth paths may have quite a high impedance, so if the PEN is really just N then this could explain the 30 VAC measured.

How do you solve the softener valve problem though, Nick? I tried a check valve between the softener and the accumulator and it didn't solve the problem, I had to move the accumulator to the inlet side to fix it. Probably not a problem with a single cylinder, non-metered, type softener, that regens overnight when there's no water demand, but my experience suggests it is definitely a problem for the non-electric twin cylinder type softeners.

The relatively high voltage measured between N and E is a possible clue. If this was a true measurement, then it indicates a range of possible fault conditions, but the one that would concern me is if Ze, the earth loop impedance, has risen to an unacceptably high level for some reason. This should be a low value, less than 0.35 ohms if the house earth protection scheme is TN-C-S, or PME (should be if the house is only 5 years old). Do you happen to know what the earth system is? Alternatively, is the house in a very rural area, where the DNO may have been unable to provide a TN-C-S supply, so the house could be using a TT system? Clearly there seems to be an electrical installation problem somewhere, either on the consumer side or the supplier side, as everything seems to point towards an event that has caused the supply to go outside an acceptable range for some reason. Having a 3 phase supply introduces some other possible causes, as 3 phase will be 400 VAC (in harmonised terms; it's really still 415 VAC as we fudged the tolerance to comply with the EU requirement, as we did with 230 VAC single phase). As the 3 phase ASHP has failed and now trips the circuit, do you know if it's tripping the 3 phase RCD or the MCB? Your chap can check Ze quickly and see if that odd 30 VAC measured between the house N and the incomer PEN (I'm assuming this is what was checked) is real or not. I suspect that, if the 30 VAC measured is correct, then Ze can't be within limits, as if the installation is TN-C-S then with with the max allowable Ze of 0.35 ohms, 30 VAC between N and E indicates an earth fault current flowing at the time of the measurement of just under 86 A, which would be producing a host of other symptoms. I'm with @ProDave on this (I've typed most of this before he posted) but would suggest that it may well be worth getting the whole installation inspected and tested, for reassurance as much as fault finding.

I've got a spreadsheet that works out the heat input to a house from a given floor area that's fitted with UFH on another machine. I could have a go later at combining this with the heat loss sums and tidying it up a bit to make it easier to use. I've also got a simple U value calculator sheet, which will work out the composite U value for a floor easily enough, so may try and incorporate that, so someone could just select the insulation type, thickness, floor slab thickness, and UFH heat output needed and it could then give the floor heat loss to the ground.

That's odd, I wonder what went wrong with that attachment link? Here it is again (as before, rename the file to change txt to xls): Heat loss calculator - Master.txt

That's a very good price for water testing, much cheaper than the nearest commercial lab to us. That's also really hard water you've got! The LA have a legal obligation to provide a water testing service for private water supplies, so cannot dodge it. Private supplies that feed more than one house, or that feed something like a bed and breakfast or hotel, have to be annually tested, I believe, and the LA also have an obligation to do this (and charge for it). I found that our LA were reluctant to do a water test, though, as the government have capped the actual test price they can charge to a level that is less than the LA has to pay a lab. As a consequence, our LA will not accept samples of water for testing, they insist on coming out and taking the sample themselves. By doing this they can charge for taking the sample, which is the lion's share of the £125 we paid. BC wanted to see a certificate from the LA Environmental Health people to sign off our build, so we didn't have any choice by to pay.

Depends on the softener, but all will cause a small pressure drop, although not enough to make any significant difference if the supply pressure is high enough. All I know for sure (from personal experience) is that the twin cylinder softeners (like the Harvey, but these come with several different badges on) do not like having a slightly higher pressure on their outlet than their inlet. It disrupts the way the internal water pressure operated valves work when they change over and can cause salty water to get into the house plumbing. It happened with our system a couple of times, when I had an accumulator on the outlet side of the softener. The first time it happened I flushed all the pipework out and fitted a non-return valve on the softener outlet, but it happened again, so I shifted the accumulator over to the inlet side and it has never happened since. There's also the problem with running an unsoftened supply to the kitchen tap if you fit an accumulator after the softener. There's a recommendation that if you fit an ion exchange water softener you should run an unsoftened supply to the main potable water tap, due to concerns over excessive sodium intake. I, personally, don't 100% agree with this, as there is little evidence that the small increase in sodium intake from drinking ion exchange softened water is potentially harmful. In our case, the amount of sodium in the softened water is pretty small, and only makes up a tiny fraction of the recommended daily intake. It's also a lower than the level that's allowable in a drinking water supply. I think the key thing is really being aware that there is slightly more sodium in softened water and, if you're concerned about it, take steps to reduce your sodium intake from other food and drink to compensate. In our case, my daily sodium intake is around half the maximum recommended, so the little bit that comes from drinking softened water doesn't cause me any concern.

Don't you have to have the concrete polished first, to get it dead flat and smooth? If so, then I think that's a fairly expensive job on its own. @jack would know, as his ground floor is polished concrete.

There's no limescale problem from having hard water in the accumulator that I'm aware of. Pretty much every domestic borehole supply in a hard water area will have an accumulator just after the pump, and before any water treatment system.

Be very wary of fitting the accumulator after the softener if you are looking to use a Harvey or similar twin cylinder softener. They do not like having a slightly higher pressure on the outlet than the inlet and will reward you for doing this by filling your plumbing with salty water every now and again (ask me how I know this...). I had to re-plumb our system so the in-house 100 litre accumulator was on the inlet side of the water softener to fix this problem, as my initial fix of fitting a non-return valve on the softener outlet didn't work, probably because the high outlet pressure is still "seen" by the internal valve system within the softener that changes over from one cylinder to another during regen. If using a single cylinder softener then make sure there is a non-return valve on the outlet (single cylinder units usually have one, or specify that one is fitted). You also need a double non-return valve on the incoming supply, before the accumulator, so that the accumulator doesn't back feed the main when/if the incoming pressure drops. Normally there will be one of these on the incoming water main, next to the water meter or external stop cock, but older systems may not have one.

There's a pretty good thermal imaging camera available from Banggood (when they have stock...), that's worth a look, especially for the price (~£320): https://www.banggood.com/HT-A1-Handheld-Infrared-Imager-300000-Pixel-3_2-Inch-Full-View-TFT-Display-Screen-Thermal-Camera-Dig-p-1331615.html?rmmds=buy

The floor depth depends almost entirely on the maximum span of the joists and the distance between their centres. For example, we have a maximum span of around 4.5m, with 253mm deep Posijoists at 400mm centres. If you have a greater single span then you'll probably need deeper Posijoists. 450mm seems very deep to me, and implies that you have a pretty long span to cover. Worth checking the joist spacing, too, as if this 450mm depth is at 600mm centres you can probably reduce the depth of the joists by going to 400mm centres.

That model has a fairly short (3 minutes) anti-short cycle timer, plus the normal fan over-run time of 30 seconds after shut down, according to the manual available to download here: https://www.worcester-bosch.co.uk/professional/support/literature/greenstar-cdi-classic-combi-installation-instructions So, the boiler is presumably OK if it fires up around ten times an hour (assuming that it would stay on for a couple of minutes or so whenever it fires up). I'm not convinced that deliberately making the boiler short cycle as a way to modulate output below the lowest level is a smart thing to do, for a couple of reasons. The obvious one is wear and tear on the boiler; the less often it fires up the longer it's likely to last. Another point is efficiency. When firing up initially most boilers are pretty inefficient for the first 20 to 30 seconds or so. If you are near the flue when a boiler fires up you can detect this, as there will often be a slight smell from incomplete combustion on start up; it takes a short time for the flame to start burning cleanly.

In all probability the boiler already has anti-short cycling built in, in the form of a time delay that stops it firing up within a given time since it was shut down. I know that our combi has this feature when it's in heating mode, and it's a pretty old model, so I'd be surprised if newer models weren't similar.

You can still get long radius elbows if you hunt around: https://www.jtmplumbing.co.uk/pipe-fittings-c433/endfeed-fittings-8mm-28mm-c117/jtm-endfeed-endfeed-long-radius-elbow-p3576

They don't, as the cost had risen to almost the same as getting the standard Part P just before they stopped it, plus they put restrictions on it whereby lecturers who wanted to join Just 8 had to be in full time employment; they wouldn't allow retired part-time lecturers (like me) to have joined.

Flow restrictions from a reduction over a very short length of pipe aren't that great at normal household flow rates, the problem mainly arises when you have a narrow pipe over a long run. For example, I have a deliberate flow restricting orifice on our incoming supply from the borehole pump, a 3mm diameter venturi at the ozone injector. That flows at around 10 litres/minute, which is enough to keep the accumulators topped up at around 3 bar. Although it's good to run the largest bore pipes you can to reduce losses, there is a law of diminishing returns. We run with an incoming pressure of between 3 and 3.5 bar, and an acceptable pressure at the taps is between 2 and 2.5 bar (anything higher than this tends to make the taps splash when fully open, unless flow restrictors are fitted). The highest flow rate on the supply to the house we're ever likely to see is around 20 litres/min, most of the time the flow rate when using taps etc is around 6 litres/minute. If we had a 10m run of 15mm copper pipe as the main feed, then at 20 litres/minute flow rate the pressure loss along the pipe would be 0.8933 bar. So, for 3 bar in we would get 2.1067 bar out at 20 litres/min, which is perfectly OK for pretty much any domestic use. If we had a 10m run of 22mm copper pipe as the main feed, then at 20 litres/minute flow rate the pressure loss along the pipe would be 0.1255 bar. So, for 3 bar in we would get 2.8745 bar out at 20 litres/min, which is again perfectly OK. If we look at what happens if we put a very short length of 15mm copper pipe into the system, say a 0.5m length of 15mm, with the remaining 9.5m in 22mm, then for the same 3 bar in we would get a pressure loss of 0.0447 bar along the 0.5m of 15mm pipe, plus a pressure loss of 0.1192 bar along the 9.5m of 22mm pipe, giving a total pressure loss along the whole 10m of mixed pipe of 0.1639 bar. So, for 3 bar in we would get 2.8361 bar out at 20 litres/min, which is hardly any different to having the whole run in 22mm. The above is for straight runs of pipe, and bends, particularly elbows, will likely have a greater effect on flow rate and pressure loss than a short length of smaller bore pipe.

Yes, I fitted it as a backup, but never used it, other than to test it. It's been removed now, and the space it occupied on the wall has been taken up by the Sunamp control box.

OK, I'm going to do write up, with photos, on our experience with first a Sunamp PV, and since the beginning of this week with a Sunamp UniQ 9 eHW, as a blog entry this weekend. The short explanation is that a Sunamp heat battery is like a compact thermal store (a lot smaller than a water filled thermal store for the same capacity) with significantly lower heat losses than a typical water-filled thermal store (my experience is that the losses are only about 20% of those from a very well-insulated water-filled thermal store). The UniQ 9 that we have now loses about 0.75 kWh per 24 hours and has about the same storage capacity as a 210 litre unvented hot water cylinder. The water-filled thermal store we had lost around 2.5 kWh per 24 hours, and that was double spray insulated, plus I added an additional 50mm of PIR insulation around it, to try and reduce the losses (it was over 3 kWh per 24 hours before I added extra insulation). These very low losses effectively increase the capacity of the Sunamp over a 24 hour period, plus the way that the phase change material works means that the temperature of the hot water it delivers stays pretty constant until the Sunamp is depleted, whereas for a water-filled thermal store the delivered hot water temperature tends to drop steadily as heat is drawn from it.

My experience when I rang NAPIT was pretty positive, they were far and away the most helpful of the Part P accreditation companies I spoke with, so my guess is that they would see this for what it is.