TerryE

There are lots of ESP32 based relay boards available on AliExpress at all sorts of relay combos. The build quality and electrical design of the better named suppliers is excellent. The advantage of going the ESP route is that you can load these OTA with ESPHome or Tasmota firmware which makes them v, easy to integrate to Home Assistant, etc. You need to be very aware of the contact life, especially without snubber protection (or flyback diodes if switching DC). I chain mine to contactors if I am driving anything about 1A, but this definitely requires snubber/flyback protection for the control relays.

Another happy MBC owner here. We put some extra 11 cm pipe runs down from the plant cupboard under the slab and out beyond the peripheral path . We ran various services through these using various flex sleeving covers. These proved great during initial build, but were flawed for future additions -- basically after 2 years of external ground-works and landscaping the external covered pipe end got lost / buried too deep / covered by paving etc. We had a rule that no trades were allowed to breach the twin-wall, so we did all the through ducts / conduits then once the sparky etc had done using them we foam filled and silicon-sealed them all before the pressure test. I only wish we'd put some extra through pipes "just in case" as well. There are no air leaks and tiny U-value hits doing this to it;s a damn site easier drilling one of these out if you need to add extra services later than trying to add a new airtight through duct once the building has been boarded out and plastered / outer skin complete. I ran power and Gb enet to a small shed close the house and added a PoE switch here so that all of the external cameras etc. could be cabled to this shed.

It doesn't have to be a door of course. Depending on the diameter of your fan, it might be easier to fit one to an opened window light.

You need to be careful that you don't do an apples and oranges comparison. MBC contractually guarantee passive class performance, so in our case (and others on this forum who have used them) all three MBC crews spent their time doing such builds, so they really understood the fine details of making it happen. We got 0.5 ACH on first attempt without any trouble shooting of missed leaks. Tightening some of the loose Internorm window adjustments to spec dropped this by another 30% or so. @SBMS point is very valid if the contractors teams have a similar understanding and attention to build detail. It is more difficult to avoid air leaks and bridging in a B&B build compared to the MBC twin-wall technique which avoids most of these by design.

My last house was a circa 1700s stone farmhouse, concrete and tile ground floors. Extremely picturesque. Lovely to live in -- in the summer; f'ing freezing in the winter unless I had two downstairs multi-fuel stoves on the go and the CH cranked up. The exterior walls were typically cold (say ~10°C) to the touch. My new self-build is passive class. Jan and I were pretty anal about keeping U-Values to standard for class, and tracking down then eliminating / decently mitigating bridging. There were three bridging areas that crept into the build and that we addressed, but even so our house only performs at maybe 30% worse than the as-designed calcs -- but that's still 5× or more better than the typical new build. You have a huge insulation barrier internal to external in this class of house -- assuming that you haven't missed any bridging cock-ups in the build. The internal floor-to-floor barriers are extremely small in comparison so everything everywhere is at or near room temperature to the touch. We run our house as a single zone at around 22½°C though the CH algo only uses the cheapest Octopus half-hour slots so the temp does wobble around throughout the day by a degree or so. We heat the GF with UFH in slab. We also have a couple of free-standing oil-filled rads that we use in winter for top up on the upper floors (also CH controlled), and we have a electric towel rad in the master bedroom en-suite; other than that there is no heating at all on the two top floors, just parasitic heat from the GF. I once went around the house one winter evening with a spot thermometer: the ground floors, ceilings, internal walls were within ½°C of the same temp. The external walls about 1°C cooler and there was a cool hand's-width in the reveals around the external windows where there was some bridging through the timber framing of about 4-5°C but well above condensation thresholds and small in total W terms. Ditto the first floor, but this was about 2°C cooler overall. (This was before we used the top-up rad.) My son's bedsit is on the in-loft floor, but this is boosted by his games PC and two huge TV monitors. 🤣 So you can really run a passive-class house as a zone with GF-only heating. Simple boosts work, and are extremely cost effective. IMO, there is too much focus on getting run-rates to the absolute minimum, and not enough on overall through-life discounted costs. In our case we have an electric-only resistively heated house. This is still cheap to operate on an Agile tariff. We built it ASHP-ready but have never made the investment case to install one as it would take maybe 15 years to get a payback.

I have just been searching the forum for posts on oil-filled rads, because my "big old clunky" has finally given up the ghost after 20 years, and TBH the most informative posts are my own. 🤣 I've been through a number of supplier sites and Amazon and the postings continue to be misleading quoting the electrical heating element power rather than the actual typical thermal output. I note that the CPC technical datasheets on the site that @ProDave linked to do quote an actual thermal power O/P so 👍 for CPC. I also note that many now include some form of embedded micro control with timer and remote options. I want a dumb heater where I can (statically) switch select the heater power that I want, and then control the timing and actual energy use through a Tasmota smartplug and my HA system. CPC lists some nice simple models. Simple is good as far as I am concerned.

If the ext walls, floor, roof do genuinely have a U-value of 0.1 then that is pretty much an order of magnitude better then interior wall so treating floors as a single zone and ignoring interior walls is a good simplification. Don't forget airtightness and go for 0.5 ACH + MVHR. so no weep vents. I circulate my UFH loops for 15 min every 3 hours so I can use the manifold return temp to take the average slab temp and this also spreads any solar gain across the whole slab.

We've lived in a 3 storey new build with a comparable spec for 7 years now. Our ACH is around 0.5. We have an MBC WarmSlab that's 100 mm reinforced concrete on ring-beams encased in (mostly) 300 mm structural EPS and UFH embedded in the concrete raft, so ground floor only. The UFH is run as a single zone and heated by a single 3kW inline resistive heater. We keep the whole ground floor 24 × 7 within a ~ 1°C tramline of 22½°C, the 1st floor is typical 1-2°C cooler in peak midwinder, though I do use an 1kW oil-filled rad (controlled by my CH system) on the 1st floor for the peak midwinter period. My son's bedsit in the warm loft is mostly heated by his Games PC and huge TV screens. 🤣 The house was built as ASHP ready, but we put the inline heater as a stopgap until we had confirmed the heating characteristics and loses of the house. However in practice, the heating costs are relatively small, and we have never made a payback case to install, maintain and depreciate an ASHP. We have never missed upper floor CH. I also use the Green Octopus Agile Tariff so my CH system schedules the CH and HW on-times to buy and use the cheapest half-hourly price slots. I am not necessarily suggesting you do the same, but what I am saying is that for this class of house running the whole GFL as a single zone is entirely workable, and we use less than 30 kWh heat daily even in cold Dec-Feb, so a ~ 5kW ASHP should be fine for this class of house. With a ~ 10m³ slab acting as a thermal store, it is even questionable whether you need a separate buffer tank for the UFH. I suspect that most ASHP installers have no idea how to design and right-size an ASHP for a (near) passive-class house.

So is ~£100 plugin rad in the hall. Yes, it costs more to run, but no installation costs. How many years before you break even, if ever?

IMO, if you have a reasonably low energy build with UFH then you need to think very carefully about the pros and cons of multiple zones, upper floor heating, etc.. There's a lot to be said for KISS (Keep It Simple Stupid) IMO, as this saves in installation costs, complexity and through life maintenance. E.g. upper floor heating is a PITA. You can run UFH at ~30°C so you can run your ASHP at a low temp and have a significant boost in CoP in doing so. Low temp rads take up lots of wall space and are expensive to install. We have no heating on our 1st and 2nd floors, but use a cheap oil-fill electric rad for a few hours on cheap rate electricity during the really cold months just to top-up upstairs.

@JohnBishop, you need to be careful switching 240 VAC stuff especially at high current draw especially if you don't have a grasp of current Building and IEEE regs. If you do this without Certified Electrician sign-off and have a fire then you could find that your insurance company rejects the claim. This is my set up. The CU to right was installed by my sparky. I added the left-hand extension myself because he couldn't find the time when I needed it installed, but I did this in a way that he'd be happy to sign off. He's comfortable with the 24VDC / 240VAC spilt as there isn't really a safety issue with the 24VDC side. Node-RED is a Javascript based environment that is well suited to IoT control and comes as a Home Assistant add-on as well as being able to run standalone. There are lots of YouTube videos on this.

This depends on what you are controlling and what inputs it needs. You can't make a general choice, as it depends on context. A lot of this was discussed in my topic with this post summarising my final setup and we are pretty happy with it. In short I have a bunch of contractors in a CU extension next to the main Consumer Unit and these switch power to the HW immersion, slab heater and pump. These are in turn driven with 24 VDC control signals by an off-the-shelf ESP32 relay board running Tasmota. The Tasmota firmware is switched by my NodeRED app. I do have a lot of Zigbee sensors but I prefer to use TCP layered stacks such as Tasmota for any switched devices relating to Central and HW Heating.

@WisteriaMews, sorry I am in Greece ATM, and so not checking the forum daily. The OSO UVC has an double skin: inner steel pressure vessel and our plastic or powder coated aluminium with vacuum panel insulation between. The two immersions are installed in small access panels with the wiring terminals, thermostatic switches etc. There is an internal duct channel for the 3kW flex cabling running from the bottom to up the two access panels. I used a couple of waterproof ds18b20 digital thermostats to measure temperatures. Look them up on eBay etc. The "waterproof" bit isn't important, but they are encapsulated in a ~8mm dia aluminum can that can be thermal-taped directly to the steel cylinder with a bit of foil+mineral wool insulation on top; you can get a get a range of lead lengths but 1½ or 2m is enough to take it down to the bottom channel of the cylinder, and thin enough to be run alongside the flex in the internal cabling duct. These thermometers are wired into a small IoT board, a Wemos D1 Mini which in a small ABS box about 25×40×50mm IIRC and powered using a USB lead; This ESP8266-based module uses ESPhome firmware to talk back to my home automation system over local Wifi. I use the Node-RED javascript code to track temperatures, do the heating calcs, and send the MQTT commands to power on and off the power to the two immersions. Here is a screen grab from my Home Automation system for the last 24 hours. You can see that my son had a bath or long shower at 5PM, and it scheduled top up at 00:30, 04:30 and 09:30 to bring the UNC to target 50°C because of the cheap spply pricing in those slots. (The price actually went negative in two of these slots so I actually got paid to heat the water.) The ~ 4°C cooling / day is the parasitic heat loss from the UVC which is very small compared to conventional cylinders or thermal stores.

@ProDave, in a nutshell, our decision to replace the SunAmps was because of the poor reliability of their engineering implementation: the circuit boards broke basic layout design guidelines, e.g. lack of 240V isolation and inadequate track sizing for 3kW power routing, so they had an MTBF of maybe 2 years; all of the pump and control plumbing was hidden inside a pretty steel case, which looks sexy but makes maintenance and repair a total dog. Yes, the concept of a PCM thermal store for DHW has great potential, but not this model and company. @Duncan62, Jan is our procurement specialist; she found a supplier online, which had the model we wanted in stock and could deliver it in a week. Seek and ye shall find. @WisteriaMews, We have an electric-only passive-class house. All of our white goods use timer-controllable cold-fill in-device heated water, so our HW use is pretty low: primarily showers for three adults and occasional shared baths for the 2 OAPs (mostly in the winter). Our OSO model is double immersion heated only, which makes sense for our use. I need to do I full blog post on this, but my CH control system uses 4×Finder contactors driving the 2 OSO UVC Immersions, my UFH pump and a 3kW heater that heats the UFH. I use Home Assistant to do the user interface and graphics for my Home Automation, but in terms of CH and HW control I have a custom NodeRED application (about 900 lines of Javascript that I wrote) to control a 4-relay Tasmota module via MQTT, and this in turn outputs 24V control signals to drive the contactors. I buy my electricity using an Octopus Agile Tariff. This has an API which published its day-ahead half-hourly prices at around 5PM starting midnight C.E.T. In essence, at 11PM each night, this app schedules the daily CH and HW heating for the next 24 hrs to use the cheapest half-hour slots. In terms of HW and heating the OSO, I have installed top and bottom digital thermometers to monitor the tank temp. The app uses these 2 tank temps to estimate the total kWh needed to bring the tank up to 50°C and it uses the day-ahead prices to schedules the cheapest time slots to heat the tank. (If the top temp falls below a preset, it also stuffs in a ½hr top-up regardless of price, but this happens so rarely that this isn't a significant cost premium for always having HW.) The OSO has a parasitic loss of just over 1 kWh / day, plus replenishing used water, which varies but is typically equates to another 2-3 kWh at most except during the winter bath season. The kWh price for this last couple of weeks has been negative on 4 days , small pence on another 4 and only peaking at 14½p on the stiller overcast days, so the cost of of our water use runs at under 50p / day on average. We tend to have our big bath days when Octopus will pay us to do this or it only costs pence. I've been coding for over 50 years so the Javascript bit is easy for me, but I understand that those who aren't so IT literate will have a lot fewer options. I believe the like of Octopus and OVO are exploring some Smart Control options, but these aren't yet prime time.

I call our build passive-class or low energy rather than passiveHaus complaint. For me, IanR hit the spot with his above comment. BTW, we use resistive heating only, and on current Octopus Agile tariff pricing our monthly energy bill varies from about £75 or so in the warm season to maybe £200 for the coldest months. It's hard to put a precise value for the bill as we take all of the price variation risk on the Agile tariff. However, looking at the potential monthly savings in going from a CoP of 1 to an ASHP-based 4, say, for the heating element of these costs; These savings are significantly less than amortised cost for the installation, maintenance and depreciation costs of adding an ASHP, so unless something changes radically in our case we won't be installing an ASHP.

I am 182 cm tall, and a couple of our internal doors were about 181 !! I'd just duck slightly automatically, but about twice a year (especially is some spoke to me from behind) I'd forget and clip my bonce. Ouch. 🤣

In our previous ~2-300 year old Cotswold stone farmhouse we put it a silicon DPC back in the mid 80s and completely repointed over the next 10-20 years the infill showly turned from damp loam to bone dry dust and grain husk. The effective U-value slowly improved down to around 1½ W/m2K. The main problem was that the house leaked air terribly and the floor slab was uninsulated. Our current house has a U-value for exterior walls, roof, floor in the range 0.12-0.15 W/m2K and an an ACH of around 0.4, plus MVHR. Just no comparison in terms of winter comfort. When the kids were young, they ran around naked in doors in the winter but then again gas was dirt cheap in the 90s so the boiler was going flat out when we were in the house.

Actually the code and graph were mine; the context was a passive house twinwall (high insulation with decent thermal capacity) with an outer stone skin. This is a world away from the thermal characteristics of you stone cottage (which is of similar construction to my previous house). The U-values you get from a solid stone wall are terrible as stone is an awful insulator compared to a passive profile (the U-values are typically 10-20× greater). Yes there is little temperature variation inside as the walls are just always cold in the winter. You can really only mitigate this by adding something like an inner aerogel backed plasterboard liner to the exterior walls. As heat just wicks away through every external surface, you just need to input a lot of heat to get a comfortable room temperature.

That being said, I believe most ASHPs can be programmed to have a few time slots each day when the ASHP outputs at the higher temperature setting and a Y-plan valve switches between the UCV and the manifold loop, so the CoP is 4, say, when doing the heating, but 2½ say when heating the water. This assumes of course that you using that you are using low temp rads or UFH. But this all adds complexity. @JohnMo or @ProDave will tell us if I am talking rubbish. 🙂

The LCoE for new solar and wind is about £30 / MWh, that's 3× less. The revenue starts to come into on about a year after the mean investment point, that 10-15× less. It is just really hard to make new nuclear numbers economically competitive, even SMR technology. We can get up to ~90% renewable with investment in grid-scale battery and maybe pumped hydro, whilst running our existing gas peaker stock until EoL. By then other options will be available.

Octopus are quite transparent about how they set their Agile price: 2W+P where W is the half-hourly EU wholesale price which is fixed for the day-ahead each Midnight CET. The 2 is a markup of 100%; P is another uplift of 14p/kWh from 4-7pm GMT -- a time limited markup to discourage peak use. The two markups combine to cover their distribution and service costs. This 2+ markup may seem a bit steep but it is less than the others charge. This total price also capped to 100p / kWh inc VAT as per current UK legislation, but other than this, consumers on the Agile tariff don't have to pay a premium for buying future prices, so whilst they take the pricing risk, the expected price is less overall. If I we doing this, I'd have had the markup 2 for W>0 and 1 otherwise, but I am not going to make this suggestion. 🤣

Yes. there's an excess of energy being generated -- such is the variable nature of renewables.

The 20A Power Relay and the 25A Contactor are pretty much identical in form factor. The contact is far more robust and only about 20% more expensive. I'll just buy 10 1N4007s and make up my own flyback straps.

Mike, it's nice to give back on this one. ESPhome does now support the Pico W. See ESPhome - Raspberry Pico W. The ESPhome environment is layered over RTOS, so I assumes this is just a port of their RTOS build. The standard ESPHome dev environment typically runs in a docker container -- either via the Home Assistant add-on or a standard docker container from the Docker Library running under Docker. This is just a builder WebUI using a load of Python scripts to wrap around the make environment. I find it easy to use. I like it. You configure the particular device using a YAML file which can include C encoded actions using the "lamba" nodes. It's all pretty straightforward once you are past the initial learning curve. Update is normally via OTA, so if you need to tweak a config, then you just edit the YAML file and hit deploy. The build and update takes less than a minute. If you want to discuss further then why not spin off a specific ESPhome (on Pico) topic? PS. The Pi Hut 8-relay board that you linked to has a similar form factor to the ESP one that I suggested, but you would need to mod the case slightly to take the external Wifi antenna. Pi Hut also sells a 4 relay version. It's a pity that I can't find a proper 24VDC relay driver board / hat for the Pico or ESP32. I see that some designs use PWM relay drivers with RC smoothing and flyback protection. The idea here is that you set the op to 100% for 50 mSec for force contact close then drop the PWM to 50% dropping the smoothed voltage to ~ 12V which is enough to keep the contactor closed but only a ¼ of the power loss in the relay and if you don't need sharp one, then drop the PWM ratio to 0 over 100 mSec, say, which pretty much dissipates the coil energy before open. Anyone know of / recommend such a off-the-shelf board? PPS. ESPhome can be configured to act as I/O subsystem for HA or work independent of HA using MQTT as its command interface.

Precisely. And the answer is no, not at the moment 😱 but I should have a freewheeling diode across the control contacts of each. Do the math and Finder relay coil stores about 0.6mJ of energy when closed and as soon as the Sonder relay opens this coil field collapses creating a flyback voltage in the kV range at its contacts which will rapidly degrade the relay. Probably don't need an RC snubber with a freewheeling diode (something like the 1N4007 which are orderable in 5 off quantities from Farnell etc. for pence). What I hate about this was that I really wanted to avoid customise tweaks. I really wanted a compute off-the -shelf module. I have found a for ~£200 from an AliExpress supplier I've used before but it has 32 channels and is really overkill. No the 22.20s are power relays really designed for straight AC1 resistive loads like the immersion heaters. The 22.32s have sensible AC7a rating but the profile is slightly different so I need to check that they fit in my enclosure. If anyone have setter advice then I am more than willing to listen.