Jeremy Harris

It's really just a way of turning boost on when the shower or bath is running, or when cooking, rather than as a way to control comfort. TBH, it doesn't make a massive difference if you don't boost the MVHR when having a shower, as the room dries out pretty quickly anyway.

My local pub when I lived in Cornwall (as in 100 yds away) was the Blue Anchor in Helston. They've brewed their own beer there for hundreds of years, and their ordinary bitter ("middle" to the locals) is around 5% ABV. The idiot emmets (holiday makers) almost always came in an asked for a pint of special, which is around 6.5%, but can be as high as 7.5% at bank holidays. It was fun watching them come in on a hot day, especially on a bank holiday when the "extra" special was on, drink two or three pints and then fall over...

There are some remote control switches that use a battery-powered wall switch and a receiver that could drive a mains relay, to give dry contacts and isolation, so could be used with an MVHR, which would save any tricky wiring runs. The Smartwares range (which used to be branded Homeeasy) isn't too costly. You can programme one receiver (like this one: https://www.uk-automation.co.uk/smartwares-built-in-power-switch-1000w-sh5-rbs-10a/ ) to operate from several wall switches like this: https://www.uk-automation.co.uk/smartwares-wireless-single-gang-wall-switch-sh5-tsw-a/ All you'd need to add would be a suitable 230 VAC relay in an enclosure to operate the boost function on the MVHR.

Should work at least as well as an ASHP, as the ground temperature below about 1m down is fairly constant and typically around 8 deg C. I've found that pumping water around our UFH at 12 deg C is very effective, so a suitably sized ground loop should be fine. Whether it's cheaper than an ASHP I don't know, as the cost of trenching a GSHP collector (which is essentially what would be needed for the ground cooling loop) can be quite high. The cheapest way to cool a house (apart from designing it with appropriate levels of shade and using a high decrement delay structure) is probably an air-to-air heat pump. These are pretty cheap, under £1k, and will very effectively cool a house if the internal unit is mounted somewhere high, where cool air can freely circulate. If the house has PV, then the chances are that the running cost for the air cooling will be zero, as it will only really be needed when the sun is out.

I like the idea of using rate of change of humidity as the boost control parameter, rather than just a set RH level to trigger boost. I fitted a humidistat to ours, with the remote sensor in the extract plenum, and it's not great, as the threshold has to be changed in the spring and autumn to account for the general shift upwards of RH in winter and down in summer. I'd be inclined to just use something simple, like a boost timer that's triggered by the RH increasing by 2 or 3% in a short time period, say 2 to 3 minutes. I have our humidistat set to keep boost on for 20 minutes once triggered and that seems to be about long enough. I suspect you'll need to do some monitoring for a time to get some data on how RH changes with time in order to set the rate of change figures. As far as I can tell, our summer bypass (which is continuously variable) just works on the basis of intake air temperature versus the room temperature set point (the Genvex can heat and cool so has a programmable thermostat built in to the room controller). I'm not convinced it really makes that much difference, as by the time the thing is at 100% bypass the cooling system almost always comes on, and when I experimented with turning air cooling off, to see what effect bypass alone would have, it didn't seem to do much, if anything. If you can control bypass so that you can set it to do as many in hotter areas, like around the Med, do, and go to 100% bypass at night, so you can get the MVHR to do a traditional "night purge" then that might be useful. I find a cooler night time temperature more comfortable, and using the cooler night air to take the previous day's heat out of the internal structure may well help to keep the house cooler during the day. That's the traditional practice in places where there is a big day/night temperature swing, leave the windows open with just ventilated shutters closed at night, then close the windows during the day to reduce the amount of hot air that can get in. Combined with a house that has a high decrement delay structure it's a technique that seems to work well.

I took the precaution of fitting a fresh air inlet immediately above the entrance to the kitchen/dining area. The utility and downstairs WC come off the kitchen/dining area, so overall we have a lot of extract moving air away from the rest of the house, and adding a fresh air inlet tends to force airflow away from the stairwell and down and out through the extracts. I did this because I was concerned about cooking smells going up the stairwell, and it seems to work. I think putting an extract in the stairwell would encourage air from downstairs to rise even more than it would tend to do anyway, so may make things worse, whereas feeding fresh air in at the top of the stairwell may well encourage air to flow back down to the kitchen extract.

Might be fun, but winding up people like this just ends up with more grief, I think, especially if you give them a genuine issue to complain about, rather than their very irritating, but essentially harmless, imaginary issues.

That's very useful to know, had I known that I think I might have looked at the Combi rather than the Premium 1L, as it does seem to be a very neat package.

No need for pipe stats with the re-badged Carrier units, they have a sensor built in to the flow pipe from the ASHP that is used to set the flow temperature, depending on selected mode. For example, I have ours set to deliver a flow temperature of 40 deg C in heating mode, 55 deg C in hot water mode (which I don't have connected) and 12 deg C in cooling mode. You can make the heating mode temperature follow a weather compensation characteristic if you wish, but I found it worked better for us with it set for a constant 40 deg C. By default the unit will be pre-programmed to a weather compensation characteristic, probably one that doesn't work that well for the UK climate, if my experience is anything to go by. In terms of controls, all I have is two room stats (one for heating, one for cooling), a standard single channel programmer to turn the whole system on and off, and a tank stat on the buffer tank that has priority over the room stats. Giving the tank stat priority means the buffer is always charged to between 35 and 40 deg C when the programmer is on, irrespective as to whether the heating or cooling is on. If cooling is on with no call from the tank stat for heat, then the buffer tank valve (in the flow to the buffer from the ASHP) will be closed, the UFH valve (on the return from the UFH manifold) will be open. If heating is on with no call from the tank stat for heat then the buffer tank valve is left open (so the buffer works as a buffer for the UFH) and the UFH valve is open. If the heating is on and there is no call from the room stat or the tank stat then the buffer tank valve stays open and the UFH valve stays open but the ASHP and UFH pump is turned off. If the cooling is on and there is a call for heat from the tank stat the ASHP is turned off, then set to heating mode, the UFH valve is closed and the buffer valve opens then the ASHP restarts in heating mode. This way the floor stays cool but warm water can recharge the buffer.

Could be a sneaky tactic to be able to price the next generation Powerwall at a lower price - they must be close to switching pretty much everything (except the Model S) from 18650 cells to 2170 cells, as that's now the primary product coming out of their cell Gigafactory. My guess is that in the short term the new 2170 cells are more costly to produce, even if they will be cheaper in the medium to long term, so now that Tesla has a significant share of the home battery market they may feel they can up the price without hitting that hard, whilst ensuring that the next generation is perhaps more competitively priced.

On the topic of batteries, this article about how disruptive battery technology can be when used in specific cases is interesting: https://cleantechnica.com/2018/05/15/elon-musk-harpooned-baseload-power/ In essence, Elon Musk took a gamble by making South Australia an offer they couldn't refuse, following major power outages and fluctuating demand that was making wholesale spot market prices vary by thousands of %. He offered to build a large battery storage facility to enhance grid stability for $50M by a set date, if he failed to meet the date the plant would be free. In the first four months of operation it saved $35M, a massive return on a $50M investment. It's an extreme example, created in large part by the chaotic way the power generation system was working (or rather failing) in South Australia, but we're getting to the stage here in the UK where we have enough excess renewable generation to make the wholesale spot market prices go negative very occasionally now. In a couple of years I suspect we'll see negative pricing on a more regular basis, which then makes big battery installations like this look increasingly attractive. I will admit to admiring the way Elon Musk does things; pretty much all he has done has created significant change, from Paypal, through Tesla to SpaceX.

Non-contact IR thermometers are calibrated for a specific surface emissivity, usually around 0.9, as that's about the emissivity of painted surfaces, concrete, etc. Something with a very low emissivity, like a shiny pipe, or especially a chrome plated surface, with cause a vary large under-reading. A chrome-plated surface, for example, will probably have an emissivity of around 0.1, maybe less, so the error will be very large indeed. The fix is to put something like a bit of masking tape on the surface. That will have an emissivity that's close to 0.9 and will conduct heat well enough to get a more accurate surface temperature reading. Worth noting that something like a matt black surface will cause the opposite effect, and make the IR thermometer over-read slightly, but not by much, as a perfect black body has an emissivity of 1, and a mat black surface is not going to get to this level, so the error is probably only going to be small.

86.5 deg C is enough to burn, whereas 86.5 deg F is just a bit warmer than most swimming pools (I seem to remember they are around 28 deg C usually). I found that a towel rail at anything over about 55 deg C (131 deg F) is a bit too hot, so that's what ours are set to.

Not at all uncommon, I think. If you want a reasonably well-balanced view of medical research data, then Cochrane is pretty much the gold standard, I think: http://www.cochrane.org/ If you have any spare time, and an interest in it, you can help as a volunteer. If you have a background that has involved peer reviewing scientific papers, then you can be even more useful. Cochrane is staffed largely by volunteers, who give their time freely, solely to try and better sift out hard evidence on a wide range of medical treatments.

The impact on my health was much the same, except my wife didn't really spot how bad I had got until things were really bad. I got in a bit better shape physically, apart from knackering my knees, a wrist and a shoulder, but mentally it was a very different story. I've never (as far as I'm aware) suffered from any mental health problems before, but there was a point in the build where I got very depressed indeed, to the point where I pretty much stopped functioning. That episode slowed the build down a great deal, as there is no quick and easy cure for depression; it takes time to recognise you are ill, that you need treatment and to adjust to the changes you have to make in order to cope with life. It changed me a lot, sometimes in unexpected ways. I've never liked working outside, for example, but now get a lot of pleasure from doing jobs like preparing the ground, planting and looking after the hedges and trees I've recently put in. A neighbour commented on this last week as he was passing, saying that our trees must be the best looked after in the village, from the time I spent working out there around them. For me it's therapeutic, not something I'd have believed possible a few years ago.

Yes and no, as I understand it a lot depends on the reason for filming or taking photos. I believe that there are also different rules for CCTV and hand-held filming, and that you are pretty much OK just filming something with a hand-held device most of the time, even if it is recording activity on someone else's land. I think the only restriction is if the recording is being made covertly of an area where there would be an expectation of privacy, the sort of thing that covers peeping tom like actions. I doubt that anything could be done about someone openly recording activity on a building site, from their own property, with a hand held device. I do know that for CCTV the rules are different, in that you have to position cameras to avoid invading someone else's privacy. It's OK to have cameras that include a bit of the public highway, or your side of a boundary fence, but not OK to have fixed cameras that look into a neighbours garden or at their windows. I ran into this problem when our neighbour opposite cut a very tall Leylandii hedge down, which meant that the background of our front door CCTV camera now included a view into their teenage daughter's bedroom window (since fixed by planting a line of tall trees alongside our drive, as screening).

I agree, and wish we were able to have some facing East, in particular. The combination of the cooler temperature early in the morning, the tendency for the air to be clearer and the fact that the sun rises a fair bit North of East in summer, means that East facing panels here would start generating well early in the morning and continue through for several hours. West wouldn't be so useful for us, as the air seems to always be less clear in the late afternoon and evenings here, probably a consequence of a hot day and the stream and ponds to the West. I'd did some playing around with a small solar panel when I fitted our solar powered outdoor lights, and was surprised at just how well an East facing panel did. It slightly out performed the same panel mounted facing South, which I suspect is mainly due to the temperature difference between the two, due to the different times of day that they were working. One long term plan I have is to add some panels to the East facing side of the garage roof, as I'm convinced that a bit of early morning boost would be useful pretty much all year around. I doubt I'd bother with an MCS install, as we already have consent from the DNO for up to 10 kWp, and currently only have 6.25 kWp. and I doubt that I could get more than a couple of kWp on the garage roof anyway. The most useful thing would be to have a bit more early morning generation to top up the Sunamp first thing.

I think you can probably just replace the element with a thermostatic one, like this: https://www.nwtdirect.co.uk/86-thermostatic-electric-heating-elements (I'm not recommending them, they were just the first hit on a web search). Our towel rails are both ordinary wet ones, intended for use with a central heating system, but converted to electric power by plugging the unwanted hole, fitting thermostatic elements and filling them with antifreeze and inhibitor (car antifreeze is ideal, as it has inhibitor already). I'm pretty sure the elements I bought were a fair bit less than the prices on that link above, but I can't quickly find where I got them from (I have a sneaking suspicion it was ebay).

FWIW, my money is on Tesla when it comes to the "best" battery technology. Their battery packs are proving to be very reliable, and outperform those made by just about anyone else. Other companies, like Renault, and even Nissan, are struggling to come up with reliable battery packs that are durable. Both Nissan and Renault accept that their packs will lose more capacity with time that it seems that Tesla packs do, and I strongly suspect that a large part of that has to do with the way Tesla have integrated cell production into their business, rather than just buy cells in from someone like LG. Nothing at all wrong with LG cells, but Tesla just seem to have the edge over everyone else, and their latest 2170 cells seem likely to be cheaper, have a higher capacity and longer life than any cell they've produced before. AFAIK, these new 2170 cells have just refined the chemistry of the LiNiMnCoO2 cells (also referred to as NMC) that they were using in their previous 18650 form factor cells, but have been aimed specifically at reducing cost for the Tesla Model 3. I'll lay money that the next generation Powerwall will have these 2170 cells, and there seems a good chance that it may well be cheaper than the current versions which use the older 18650 cells.

LFP = LiFePO4 normally, or can refer to LiFeMnPO4 cells, which are broadly very similar.

Yes, they are LiFePO4, a decision made largely because of the slightly better safety margin LiFePO4 cells used to have. The downside is that the energy density of LiFePO4 is lower than other chemistries, because of the lower cell voltage. My experience with early LiFePO4 cells wasn't positive, either, I bought a lot of 10 Ah cells for my first electric motorcycle battery pack and the failure rate was high and the capacity dropped with age pretty rapidly. It's notable that LiFePO4 cells are not the chemistry of choice for companies that have invested heavily in lithium chemistry cells - they are almost uniquely a China-only product, it seems.

AFAIK, @SteamyTea just has Economy 7, with no other energy supply (no PV, no batteries). The high night time use will be at the E7 rate heating up the hot water tank. For those with PV, then the sensible thing to do is use an excess energy diverter to heater the water (or other heat storage system) during the day, when the PV is generating. There's merit in looking at E7 before battery storage as a way of reducing energy cost, as that significantly reduces the night time rate, albeit at the cost of a higher standing charge. For us E7 is just slightly more expensive overall, as we don't save enough on the lower night time rate to cover the increase in standing charge.

Predicting real world life for any cell of any chemistry is pretty damned difficult, and cycle life is very far from being the only, or even the primary, consideration. As mentioned before, the massive variability in cycle life with cell SoC variability from cycle to cycle (which is not the same as the apparent DoD available from the BMS, that is often quoted) has a major impact on cycle life, but calendar life is also massively variable too, being highly dependent on things like temperature and the mean cell voltage. If a battery pack has a high mean cell voltage, higher than the storage open circuit cell voltage, then calendar life is reduced pretty dramatically (this is the primary reason laptop batteries fail early on machines that spend a lot of time plugged in to the charger, for example). Also, if the mean cell voltage in a pack is lower than the optimum long term storage voltage then calendar life is also adversely affected. In practice, house systems may be slightly better than electric vehicle systems, in terms of the mean cell voltage through life, as a lot of electric vehicles will spend most of their life with a fully charged pack (mine does, for example, it sits fully charged for at least 3/4s of every day). Taking any 3.7V nominal lithium chemistry cells as an example, most will be at around 100% SoC with an open circuit cell voltage of around 4.15 to 4.2 V, and will be close to 0% SoC with a terminal voltage of around 3.2 to 3.4 V per cell. Best storage voltage is around 3.9 V per cell, but this is only around 50 to 60% SoC, plus open cell voltage off charge is a very unreliable indicator of SoC, the BMS has to actually measure the energy that goes into and out of the pack to control SoC accurately. Most battery management systems for lithium chemistry cells tend to use "top balancing", where cells are periodically charged to the 100% SoC voltage, where the terminal voltage of every cell in the pack is the same. They are then cycled between the defined limits (typically this might be 10% to 90%) for a defined number of charge/discharge cycles, before the balancing system kicks in again to bring all the cells up to 100% SoC. The reason for this is that there will be gradual drift over time between cells in the pack, due to the tolerance on cell capacity from one cell to the next, slight variations in internal resistance, temperature variations across the cells in the pack etc. Needless to say, the manufacturers don't give much away about how their systems actually work (except Tesla, who make pretty much all the details available as a matter of policy). Anyone who drives a plug-in hybrid, or just a normal hybrid, car will notice (if they are observant) that the behaviour of the car changes periodically, with the engine running when you might otherwise expect it not too. Prius owners started noticing this right from the time the first cars went on sale, and some owners managed to reverse engineer things to discover what was going on. In my case I bought a CANBUS reader in order to be able to set some of the non-user changeable functions (like the audible seat belt alarm, the reversing beeper etc) and as a side effect this will show a lot of information from the battery management system. It's clear that once every few weeks the battery management system goes into cell balancing mode, where it takes the cells up to maximum SoC. My guess is that the frequency it does this is dependent on several factors relating to the way the car has been driven and the environmental conditions, as it's not a fixed time period as far as I can tell.

Our 6.25 kWp system behaves much the same; plenty of winter days with virtually no generation at all. I wholeheartedly agree with the difficulty of trying to model how a storage system might perform, too, even with a lot of data on consumption. The only way I could try and make some sort of reasonable case for battery storage was to ensure there were no large loads on during non-generation periods, and use a battery capacity that would meet the house baseline demand overnight and for part of the following day. The combination of load limiting (just by not using things like the dishwasher or washing machine during the evenings or overnight) plus a relatively low capacity (and hence cheaper) battery system came closest to breaking even for us, but even then I think we'd have been losing money overall, compared to just using grid energy.

Yes, my electric bike BMS does the same as the system in my car, as I designed it to run the cells between 30% and 85% SoC, but the issue with lithium cells was never really cycle life, but the relatively rapid rate of degradation with age, the calendar life. Lithium cells used to have a pretty poor calendar life, around 4 to 5 years even if you rarely ever cycled them, but this has gradually being improving, and newer chemistries seem to have a much longer calendar life, as do older chemistries with newer manufacturing methods. I did a stack of measurements a few years ago on some test cells of different types (but all lithium ion exchange cells), holding them at the best life long term storage open circuit voltage and cycling them once every few months to measure the capacity loss, and it was then between 5% and 10% per year, so for most applications calendar life was more critical than cycle life. The worst cell type in terms of calendar life degradation was the first generation LiFePO4 cells, they lost capacity more quickly than the older (and potentially more dangerous) LiCoO2 cells. The newer LiCoO2 cells I bought around a year ago seem to be significantly better in terms of much reduced capacity loss with age, so I'm guessing the manufacturing process has improved. Cycle life is highly non-linear with SoC range for pretty much every cell chemistry, massively so in the case of some lithium cell chemistries, like LiFePO4, where 10% to 95% may only give 1000 cycles, but restricting operation to 30% to 90% SoC may well give well over 20,000 cycles.