-rick-

I was thinking more along the lines of add more insulation, reduce air leakage, etc. Assumed this question was in relation to the building you are in the process of converting from a cow shed. In which case there should be lots you can do to improve things if your loss is that high. (Though I'm not sure how you'd be meeting building regs on a new construction with a high heat loss). If it's an older building, then likely lots of easy things you can do to bring the loss down. And as usual on this forum, it's worth checking where you have got the heat loss calcs from? The heatpump installers have a terrible record of vastly overestimating it which can then lead to wrongly sized systems and poor operation. Though as a regular I assume you know all this already?

Doesn't really matter too much on modern heatpumps as they are all inverter driven. ie, they convert the incoming power to DC and then generate custom 3 phase (both voltage and frequency) to drive the motor. A traditional heat pump that didn't have the inverter would gain a little efficiency (and reliability from less components) from 3 phase supply but irrelevant today. Generally, 3 phase heat pumps don't start until about 15kw. Is your heat loss really that high? Anything you can do to lower it?

And have to deal with audio cables to each location and potential hums. I like the idea of multi room audio but I've only ever got proof of concepts running and then got bored and never made a finished system. These days it's so easy to just get some smart speakers and cast to them that doing more just seems like a slog.

Not exactly. I've got a Pi 2 still chugging along, though admittedly it doesn't do that much just run some audio passthrough stuff (keep meaning to make it do more but I'm just not that interested in fiddling around with computers ATM). The key is to run an O/S that is optimised to minimise writes to the SD card and get SD cards designed for application hosting (A1 spec or better IIRC, think there are even fancier specs these days). Some of the Pi O/Ss just churn huge log files onto the SD cards and the SD cards are not designed for that. If you want to run something which generates huge amount of writes get the right SD card, or install it on a USB SSD (or PCIe SSD on the Pi 5). If you want to avoid the issue all together I think you can netboot Pis now so you can eliminate the SD cards altogether (with the offset that they depend on a server working which depending on how you do it might be more hassle). Nothing wrong with the Pis if used properly.

I think @Pocster wants to do it for the 'fun' of it. Maybe mistress is away?

It's not just smarthomes. The trend for EVERYTHING to need an app is so annoying. I like the idea of various smart features (never lived in an automated place) but key for me is that it should add things on top, not make things harder. If you can't invite your families oldies and technophobes to stay at your house without them asking how to make something work you are doing something wrong. The basic stuff should work even if the smart stuff is broken and it should all work without the internet.

I agree, I suspect that it might not come across in some of my posts but my thoughts are to try and keep things simple and avoid complexity. - Do do the calculations about how much piping you need to emit enough heat - Build a layout that allows you to output that heat at very low temperature - Within reason try to avoid different length loops and transit areas where pipes group densely (hallways can be heated by the heat from loops traversing the hall way to other areas, doesn't need it's own loops), etc - Minimise crossing movement joints (make sure you follow best practice where crossing happens to allow the pipes to move in the slab) You don't need to design everything to the nth degree, the low temperature nature builds flexibilty so everything doesn't have to be perfect. @Gus Potter I think would suggest to have more shorter loops with some overprovision on the assumption that loops will fail. This is my main disagreement. I think it adds complexity that is detrimental to the chances of a successful install. If you are paying someone to install pipes, rather than paying them to install more than you need and the complexity that comes with that, pay for an independent witness to test the system after it is layed, before slab pouring and after the slab has solidified but not fully hardened. That way you can detect faults and get them fixed at the right time rather than living with a damaged system long term. Edit: Overall, if you are paying people to install the heating system, they should be involved in the UFH design and also be able to be the witness during installation (if they aren't installing the pipes)

I took 50 years as that's what the pipe manufacturers are claiming is the lifetime of their pipes. If you are putting pipes into an insulated raft that is also the foundation you are unlikely to rip that up for the lifetime of the building. Definitely not proposing to spend significant time or effort on it. From my POV, not worrying about precisely where kitchen units go, etc, at UFH design stage just means less dependencies on progress. You can get on with UFH install while still deciding between a couple of kitchen options, etc.

I think that also describes some of my issues with what Gus has said. To me, if you are building in redundancy, you are adding more pipe than needed. If you are adding more pipe than needed you are: 1. Adding more work, materials, cost. 2. Making the installation more complex 3. Reducing the pipe spacing and therefore making bend radii tighter (more chance of kinks, etc) 4. Increasing the number of pipes that have to feed into manifold(s) which increases problems of local heating in the area, and produces installation risk with potential overlapping pipes, pipes closer to the surface than intended, etc. In a low temp system you retain an awful lot of flexibility to increase output just by adjusting the water temp by a degree or two and adjusting flow rates. Adding extra pipes just adds complexity and risk. Though it should be said, in a low temp system with thick slab, flexibility is more limited anyway no matter how many pipes you have due to the response time of the system. In return you get a more efficient system that you can just turn on and leave on and not have to think about.

My reasoning for putting pipes everywhere is that in 50 years rooms get rearranged. Kitchens get moved to entirely different areas. Partitions go up, come down. Assuming you are designing a very low temperature system with a thick slab (one with floor temps targeting 1-2 degress above room temp) then the impact of some units on top of some pipes should be minimal. You can leave small air gaps at the top of the kickstand that won't be visible. I completely understand why you don't want to put pipes under units if you are designing a higher temp/fast response system.

Watching the Restoration Couple on Youtube. I think their ~400m2 floor came to about £37m2 inc dry shake, powerfloated, not polished though. Said the dry shake bags added up to £2k. That is the concern. Keeping it sealed.

It's funny in a way. You see that picture as a warning to avoid and as someone interested in a concrete floor as a finished surface see that as exactly what I'd expect (except if done in a domestic project I'd expect the movement joints to be filled more neatly and would personally have gone for a coloured concrete shake).

@IanR did a poured resin floor that might be an alternative: Microcement exists but isn't cheap. The only other suggestion that comes to mind is exploring commercial concrete contractors rather than 'polished concrete' specialists. I've no real-world experience, just researching for a future build but if you haven't already looked, have a look at https://concria.com/dry-shake-floor/ and their polishing system. Would only work on a large floor area without obstructions, but if thats what you have then might be a lot cheaper than conventional polished concrete.

Concrete can work with the right design. Does need careful thought though and I certainly don't think it's worth a premium price.

Thats the right thing to do. Worked as intended then! Good to know of the problem straight away. Though if it really made an impression on you maybe you had the pressure a bit too high