Nelliekins

The spreadsheet from @JSHarris said a floor temp of just over 21C, so we want a flow well under 30C I would think...

Either go with system as designed in OP, or find some alternative as stop-gap until we can afford a Sunamp unit (which I think should be the long term objective, given our 5kW PV system will be massively under utilised most of the time). OP solution has the benefit of mucho cheapness, but unless it'll work there isn't much point. @Nickfromwales seems convinced it'll work even if not very good from heat loss perspective... Do you concur with that assessment? If so, maybe a slightly bigger cylinder is warranted since it'll mean a longer gap between recharges... As it happens, that TMV is already purchased, although it was originally for the DHW ? Returning to the original design I was going to use the coil to limit the heat coming out of the cylinder, but TMVs are cheap enough to justify. Hmm, we have 28 circuits... That's 400 quid... Or 20% of a 6kW Sunamp! Eek... There are 3 manifolds: - 12 port upstairs, pipes in EPS under chipboard - 12 port downstairs (10 used), pipes in concrete - 6 port in basement, pipes in concrete Actually was looking to follow @JSHarris and have the whole house as a single zone, keeping the house at a constant 20C all the time.

OK. So the buffer tank affords us the luxury of no mixing valves, provided we control the cylinder temp accurately enough. Then we tee off the coil flow and return for each of the 3 manifolds, and have a pump on each circuit? Presumably we put a bypass and PRedV on the boiler flow/return, and a pump from the boiler? I will have to mount a couple of stats on the cylinder I guess, to give us tighter control over the temperature of the flow to the manifolds? Then there is the UFH temp... We have temp gauges on every manifold, but since most of the house will be automated, I'd like to improve on that... Can we just put DS18B20s into brass tee fittings, and connect them over 1wire to the automation system? I guess a simple way to know if the floor is up to temp is to track the differential between flow and return across a manifold? Would that work? Kitchen is installed, half of the house is plasterboarded and we need to start warming and drying the place. MVHR will be commissioned next weekend all being well... More importantly SWMBO won't "camp out" until we have something more than a rocket heater to warm it up... But we need to camp out there so we can make progress more rapidly (the mortgage payments on both houses are now more than we earn every month ?) The blog will catch up eventually, I promise! ?

OK, SWMBO has decreed that I need to get the UFH working as the priority, so it looks like the 140L cylinder will be used for that. If the system gets replaced in 6 months (once we get the VAT reclaim) it will either have worked well enough for us and will stay, or will be replaced. Presumably I just plumb the cylinder in reverse (ie boiler connected direct with a PRedV and bypass valve) and the UFH circuit across the coil? Do I use 1 pump per UFH manifold, or will 1 pump suffice? (I am thinking that 2.2km of pipe is quite a volume of water for one pump to shift)

Yes at the moment. It'll end up in a cupboard with either EPS70 or PIR lining though, so not overly concerned about heat losses, especially if it's being used for the UFH side of things where the temp won't get much above 35C...

Oddly I can't find it on the manufacturer's website... Here's a link to it from one of the many vendors though... RM Aquastel 140L stainless Steel indirect cylinder

Now gents, there's plenty of me and indeed my lack of understanding to go around... Can't you share? ? @Nickfromwales will the coil in this cylinder be up to the production of DHW @ 24L/min (2 decent showers)? I had assumed I would need to use the 100kW PHE across the cylinder to get the DHW. @PeterW if I use this cylinder with a PHE for DHW, and a similar cylinder for UFH (a 10-15 minute burn on a roughly 3 hour duty cycle as the worst case scenario in winter doesn't sound that bad to me?), that's the cheapest (and least elegant, granted) solution I can see to solve my needs. Is that right? Both cylinders connected to the boiler flow using a 3 way diverter, both with immersion heaters (priority to DHW since the required cylinder temp is higher, and thus more energy stored as heat). DHW cylinder goes to 65/70C, which should still allow 50C out of the PHE to the hot water manifold, and delivered to taps at around 45C (the hot water runs are all lagged and longest is about 7m). UFH cylinder goes to 35C?, with the intention of a floor temp of 21.5C, based on @JSHarris wonderful spreadsheet. There's an argument to be had that the coil in the DHW cylinder could replace or supplement the immersion in the UFH cylinder, and maybe that'd be enough to heat the UFH on its own... But I figured that might drain the heat out of the DHW cylinder too fast, and you both seem to confirm that. FWIW I have confirmed that the floor upstairs will cope with 2 cylinders, although one could hang off the ICF wall if needed anyway... (weight has become a consideration given the posi joists the cylinders are sat on!) Longer term and as budget allows, I could look to replace one or both with a Sunamp 9/12kW unit, but the cost differential at the moment (between the 2 cylinder approach and the Sunamp) is about £2k - more than the budget can take this side of the VAT reclaim unless there's RHI or similar benefit from the Sunamp?

So from what @JSHarris has said re running the UFH from the cylinder, and what you have said re the DHW, it'd be better to use the 140L cylinder for DHW and a larger cylinder for UFH? Or is a 2h24 cycle OK for a boiler providing heat for UFH as the worst case scenario? Assume that we get 24kW from the boiler, it'll take maybe 15 minutes to recharge the cylinder (assuming 6kW capacity). We could add another cylinder (170/210L?) for either task, and heat from the boiler could go to both cylinders as required through a 3 position diverter valve. Or I could bite the bullet and just get a small Sunamp for the DHW... Depends on the cost, of course - I would only pay 300 quid for another cylinder and pump, whereas a Sunamp is probably 2k? I guess the key is to avoid short-cycling the boiler, and let it have a single continuous burn, correct?

Sorry, I didn't explain myself very well... My thinking was that at least some of the extra heat energy output from the cylinder would be taken via an extract vent through the MVHR, which would then transfer the heat back to the fresh incoming air. So in winter some of the excess therms from the TS go back into the fresh air which is circulated around the whole house. In summer, its wasted energy because it would be lost via the summer bypass.

OK, seeing 21.8W/m2 required power for UFH, and 22.3C floor surface temp. Of course that's the worst case, because we have UFH on upstairs floors as well as above slab... We have around 280m2 of heated floor area, which as a whole needs to generate (3651-400-750)=2501W (deducting heat of 4 people and 3x250W towel rails) for equilibrium (in the worst case). Does that make sense? So if we can generate a (moderate) excess above that amount, we have a system that'll work (we need the excess to heat it up from cold). So will the cylinder deliver that much heat (I guess that's a lot less than the DHW demand because its spread over a longer period) ? Is there a way to calculate the losses from the vented side of things? If the cylinder is a non starter, I will look at a Sunamp solution because I prefer that to a UVC... If I have to spend big bucks it may as well be on something that is largely maintenance free...

So are the losses from the cylinder (which we are fine with) or the header tank (which only helps us if the tank is also inside the MVHR area as opposed to just the thermal envelope)? Looking online the Sunamp units start from about £2k, is that right? Yes, it's about 3m from the extract manifold, which has 7 spare ports IIRC. That is genius, love it!

Hmm. Doesn't that rather depend on if you have ready made uses for the heat loss? For example, our cylinder will sit in the laundry room, which will: 1. Be used to dry clothes, and 2. Act as it's own heat source for the MVHR (which has an extract above the cylinder) Contrived I know, but it's the best defence I can muster!! ?

Thanks for this. Run through the heat loss spreadsheet and it says for 25C delta, I have a heat loss of 3651W, so the fag packet (3.6kW) is borne out! ? Will try the floor heat loss next.

Classic regular. The first post had a PDF showing me original idea for the system, but at the time I had no idea on the heat load for the DHW so couldn't size the cylinder. Got 140L cylinder since I was 100% sure it'd work for UFH if it wasn't big enough for the DHW. ?

No need just a suspicion at this time... 6m wide bifolds facing south could lead to overheating of the kitchen, but I don't know this to be the case at present... The WB Greenstar Classic 30i boiler has a telescopic 100mm flue vented out through the side wall, already fitted. Re cooling, got it. The cost of the boiler was low enough, and the PV was never intended to provide all the heat energy when I started this project. If I had known more about the likes of TS, and Sunamp units when we started, it would have been a different matter... We have the 30kw open vent boiler, a 140l indirect cylinder with 3kw top immersion, and a 100kw PHE. We have a 140l RM indirect cylinder, which I was planning to use as a vented direct cylinder (by just ignoring the coil) with a tank above it. The cylinder would be either for primary heating (UFH buffer) or for use with the PHE to produce DHW. Yep that's how I feel right now too! ?

The heat loss figures were actually offhand remarks from our EPC guy. He did a "back of the fag packet" job with the U values and areas for each element (roof, floor, walls, glazing) for the 3.6kW value and IIRC it was for 20C inside with -5C outside. It actually seemed a really simple calculation, so I might try to replicate it... The 2kW value was the ventilation loss based on volume of air in the house and 1ACH. That figure will hopefully decrease, because we are aiming for 0.5ACH or less, and the MVHR will further reduce the heat lost. Plus it has 1kW of post heating built in. I was assuming that we would store/use excess PV in 4 separate places: - in the DHW cylinder using an immersion - in the UFH cylinder using an immersion (summer only) - in the MVHR unit (for the postheater) - in a solar battery (to be added later) So how do I size the DHW and UFH cylinders? If the 140l isn't adequate for the DHW, would it be adequate for the UFH buffer? Would a Sunamp unit be a sensible solution for the DHW, with the 140l cylinder for the UFH buffer? Or should I stick to getting another (larger) cylinder?

Hi Peter. As mentioned in the OP we have a fairly large Brink Re450+ MVHR unit, with a 1kW postheater built in. The 2kW figure is based on 1ACH, which we are hoping to improve on quite a bit. Any recommendations on ASHP? I understand that monoblocs can be used for cooling as well as heating, so that might work very well for the UFH. I presume we would never get close to the dew point on the floor? I am not a fan of UVCs (don't know why either - just a personal thing), which is why I was heading down this path of DIY heat Bank in the first place... What's the actual heat energy requirement for the DHW system? I had thought that a 140l cylinder, heated by a 30kW gas boiler (and potentially a 3kW immersion as well) would replenish heat as fast as it was cooled by the loop through the PHE... Am I wrong to think that? Sorry - I don't know enough of the math to work it out yet!

I figured that if we used a 2nd tank as a buffer, and bypass the coil (which is the heat source) from the 1st tank, then that would at least distribute the solar gain heat throughout the entire UFH system. The system is closed, so there wouldn't be "new" water being introduced. I don't suppose there is an equivalent to an immersion heater that cools the cylinder is there? Boiler is a WB 30i regular boiler. The heat loss from fabric is around 3.6kW and 2kW from ventilation losses (assuming 1ACH - we are hoping to improve on this though) Yeah, the budget wouldn't cope with the SAs I suspect. I wasn't thinking of cooling with mains. I was thinking of a second tank as a UFH buffer, and bypassing the heat source - which would as you suggest hydraulically separate the UFH from the DHW. At worst, we can use the UFH buffer to distribute solar gain a bit, and at best we could actively cool the UFH buffer (must be possible somehow?) Neil

Oh, and a feature request for the eventual solution... we would like to be able to bypass the heated buffer tank and circulate cold water through the UFH to achieve even a tiny bit of cooling in the summer... the 6m wide bifolds are unshaded at present and face South! I presume this would simply require a bypass valve between the heat source and the buffer tank? Here's a sketch of what I was thinking of... DHW Design - draft 1.pdf

Ok, so we have already made a number of mistakes on this build of ours... 1. Not enough insulation for the floors (because it's an uninsulated slab) 2. Open vented gas boiler already purchased and hung on the wall We have a largish house (just under 300m2 over 3 floors) with UFH everywhere @ 150mm centres. Concrete floors in basement and ground floor, and chipboard upstairs (on top of 50mm EPS70 into which we have cut channels for the UFH pipes). The UFH is 16mm Pex-al-pex with a manifold per floor, connected in parallel using 22mm plastic pipe. Total pipe length is approx 2.2km. Originally, I planned to have a TS provide both DHW and CH via the UFH. However after reading dozens of posts on here, I figure I can probably do better for less money (which is kind of important now that our self build mortgage is the size of GDP for a small third world country). My plan is to use the gas boiler and a 3kw immersion heater (powered by our 5kw PV system using an iBoost diverter). They would heat a 140l indirect cylinder (that i happen to have) directly to e.g. 75C. I would use a 100kw PHE to extract the heat from the cylinder (also directly, not via the coil) and provide mains pressure DHW via a manifold (to provide centralised isolation). So the first question is this: is my 140 litre cylinder big enough for our DHW requirements? Realistically, we need either 2 showers or 1 bath and 1 shower at a time. If it takes 10 minutes to fill a bath, that's fine by me. For the UFH side of things, I have no clue. The plan is to keep the house at a constant 20C all the time. I thought of using the coil in my cylinder to provide the heat, but I doubt it'd keep up with the heat load... From memory, our space heating requirements are roughly 3.6kw and our ventilation losses are 2kw. The latter will be further mitigated by the use of our Renovent Excellent 450 Plus MVHR unit with its own 1kw heater built in, of course. Help!!! ? How do I work out if the coil can produce enough heat for the UFH? Or if the cylinder needs to be larger to support the additional load of the UFH? If needed, I can always add another cylinder, can't I? @Nickfromwales you said "A problem shared"... ? Neil.

Having nearly finished a Logix build under the occasional supervision of someone widely accepted as an ICF expert, I have a layperson's take on these... I am staggered an SE would mention any of this TBH! 1. Typically you would never inspect the poured ICF wall. Your SE should be engineering the steel in the walls to the point that crack widths become largely irrelevant except in relation to water penetration (which itself should be a non-issue). A good SE would realise that the concrete will be a LOT stronger than the stuff they make blocks out of (our mix came out at around C68 for the basement walls - 10× the compressive strength of dense concrete blocks). As for waterproofing of ICF, tell them you will either go with a waterproof concrete (like we did), or simply render the outside of the house... 2. A 25mm diameter steel bar can take a LOT of load, and you can get 6+ into a 158mm ICF wall with little effort so long as you assemble the blocks around them... the biggest problem then is making sure that the concrete pour properly fills any voids between bars. Beyond that diameter you are getting out of the realms of rebar and into dedicated members (as they rightly point out), but then you benefit from not worrying about the pour filling voids between members, so it's swings and roundabouts. Don't forget that (provided you have lifting gear on site) you can thread ICF blocks onto bigger horizontal steel members with the steel in midair. 3. Forget about fixing kitchen cupboards to the ICF walls. Line the walls with either a double layer of Fermacell or a single layer of 18mm or thicker OSB/ply (screwed to the ICF webs) behind 15mm moisture resistant plasterboard. Either approach should be more than sufficient to hold up a few wall cupboards using e.g. 5x50mm wood screws. Why would an SE care how you hang cupboards - that's nothing to do with them?!?! 4. Uhm, surely that's the job of the architect and the SE? Is the SE seriously trying to dodge that? 5. I would expect the SE to produce drawings to show where to tie the rebar, and doing the tying really isn't a very skilled job. Clearly this SE hasn't been on any ICF build sites, or they would know how easy it is to tie rebar... I did nearly all of the lintels, including one over a set of 6m wide bifolds, on my own (and that included 4no. 25mm rebar, 7.2m long, with 35 H12 links around them, all set into 2 courses of Logix 6" blocks). It's helped by knowing that typically links are placed either singly or in pairs per web spacing in the ICF blocks. If you want some photos and exemplar SE drawings to see how they marry up, I can provide them from my build. As for pouring of concrete, I don't think it is massively skilled either. It's more about understanding the process. The bigger problem is obtaining enough bodies to help, because really you want 2 people moving the pump hose around, 1 person or more vibrating the concrete behind the pump hose (carefully so you don't damage the ICF blocks), one person indicating when each circuit of the pour has reached a suitable depth, and one person looking for blowouts and bulging. I had helpers on each pour, because it's a 4-6 man job, not a 1-2 man job. FWIW my first SE was worse than useless when it came to ICF construction... he was specifying steel that could never work within the limitations of the blocks. He wouldn't listen to me, so I took my ICF expert to talk to him. Wouldn't listen to the expert, so I had the ICF company phone and email as well. When he still refused to listen, we parted ways, and I found an SE with ICF experience. The difference was night and day. 8 weeks of grief with the first SE, and under 1 week turnaround with the second SE, plus plenty of phone and email assistance afterward. Happy to recommend our SE though, via PM, if you can't find one you are happy with. Also, speak to the ICF firms - they will have a list of preferred SEs that they have a working relationship with. HTH. Neil.

Just a heads up to everyone, the blog has now started... Clancutt Lodge 3 entries so far, you will find a bit more detail than in the posts above... At the moment I am working on my phone, so hopefully the grammar isn't too appalling! When I get home in a couple of days I will start pulling together more photos etc, and writing with a bit more thought. Enjoy!

Kim and I were/are avid watchers (and readers) of Grand Designs, and so we took the plunge - we would build our own house. Well, sort of. Our plan was to "armchair self-build". I would design the house (big box with a pointy lid and some windows and doors), and we would get a builder to construct it. Seriously, at the point we were buying the plot, that's as much as we knew had to be done. Of course there was the detail, but we expected the builder to handle most of that - after all, that's what builders do, right? Wrong. But by the time we sussed that out, much time had passed and we owned the plot. And by "much time", we are talking many many months. Here's why... First, because the house and plot were on the same deed, we had to arrange for them to be split. That led to negotiations between 3 sets of solicitors. Then my wife and I decided we would have a basement. That led to Party wall negotiations, because we were going below the zone of influence for the neighbouring foundations (just). During all this, we had visited a number of self build trade shows (Harrogate, Swindon, Manchester, etc) and had settled on ICF as the way to go. So we found ourselves a highly recommended ICF specialist builder who agreed to build our house for us - splendid. But by the time we had completed the purchase of the plot, nearly 6 months had elapsed, and our builder had moved on to other work. To his infinite credit, he offered to consult on our build for free, but we would have to build it ourselves... ... Now I am a software engineer and project manager by trade, so I had already decided I would manage the build... and how hard could it be, building an ICF house? After all, I am good with Lego - here's proof!

I am Neil, and this is the story of my self build - warts and all. First, some background. My wife and 2 girls and I live in Lancashire, in a 5 bed detached modern build house on a smallish estate. For a few years, we have looked around for somewhere a bit bigger, but in the same village. We had no luck for ages - everything was either too expensive, or too small, or too near a busy main road (and frequently all 3). Then in December 2016, a house with a large garden came on the market only 4 streets away. Perfect location, and it needed a load of work doing to it, so it was going cheap. We went to have a look. House was perfect for what we needed. Ok, so it needed extending - it was 100m2 of floor and we were thinking of making it more like 250m2... but the garden was easily big enough to make that viable, and we had the budget to do it. Better yet, there was a plot of land for sale (with outline planning permission) which used to be the side garden for the house. We put in offers for both of them at the asking prices... So did several dozen other people. It went to sealed bids, so we offered 17k over asking for the house and 15k over asking for the plot. We got the plot but NOT the house. Hmmm, what to do? Here's the plot and the side door of the house - it doesn't look much, but both sides go back another 25m behind those trees... total of approx 600m2, and a snip at £100k plus fees for the plot...

Here's the original outline planning decision notice: ----- 11. The approved dwelling shall achieve a minimum Dwelling Emission Rate of 38% above 2013 Building Regulations (equivalent to 19% above Code for Sustainable Homes Level 4) and shall be constructed in accordance with the details provided within the submitted Regulations Compliance Report, Predicted Energy Assessment, DER WorkSheet and TER WorkSheet. Reason: Policy 27 of the Adopted Central Lancashire Core Strategy requires new dwellings to be built to Code for Sustainable Homes Level 4 however following the Deregulation Bill 2015 receiving Royal Ascent it is no longer possible to set conditions with requirements above a Code Level 4 equivalent. However as Policy 27 is an adopted Policy it is still possible to secure energy efficiency reduction as part of new residential schemes in the interests of minimising the environmental impact of the development. ----- The original outline planning application was dated January 2015, which is probably why it was included. Interestingly, we applied for a revision to the reserved matters a few months ago, because we had increased the footprint of the building by approx 1m2 through the use of thicker ICF walls... I just looked and the planners have omitted any reference to efficiency whatsoever in the -latest- approval - does that mean we can ignore the matter entirely???