Part Thirty Eight – Heating And Cooling Controls

This thirty eighth entry was published originally by JSHarris on the 1st September 2015 and received 1,496 views on the closed forum

As with all things associated with this house, it took me several goes to get the heating and cooling control system to work properly. Despite having done masses of calculations that showed the house didn’t really need much heating, I couldn’t, in my heart, accept this. We’re conditioned to houses in the UK needing heating, usually a fair bit of it and often through large parts of the year, when we get sudden cooler spells in normally warm weather. I started off thinking I couldn’t buy an off the shelf system, so built a home brew DIY controller, and and then spent hours writing and rewriting control code. Finally it dawned on me that there may come a time when I wasn’t able to write code, or that someone else might have to work on the system, so I had better switch to using off-the-shelf parts.

The main problem was finding a way to control devices that weren’t easily compatible with each other. For example, the ASHP has “dry contact” controls to turn it on or set it to heating or cooling mode (and a few other options I’ve not bothered with). The heating valve needed 240V switched to a thermal actuator to allow water from the heat pump (or buffer tank) to flow around the UFH heating circuit. The buffer tank needed a valve to stop the ASHP cooling it down in cooling mode, so a 12V ball valve was used to obtain a robust and complete seal between the ASHP flow connection and the buffer tank coil. The thermostats I chose, for heating, cooling and hot water, all had zero volt changeover contacts, which made life a lot easier.

I still needed to implement some very simple logic to make sure that the right signals reached the right devices at the right signal/voltage levels. Clearly I needed a few relays to switch things, some low or zero voltage switches, one 240V switched supply to a thermal actuator on a valve. Running these relays at 12V seemed both safer, and had the advantage that I could use relays with LEDs fitted, to indicate whether they were on or off.

One problem was how to house all these parts and allow easy access for maintenance, without ending up with something looking industrial. There is a standard used for mounting industrial relays, power supplies, connection blocks etc, called the DIN rail. By good fortune, this is also the standard that is used in all new consumer unit boxes. This meant that I could buy an empty small consumer unit, with a clear cover, and fit relays, connector blocks and the power supply as if they were switches or MCBs.

I’d worked out that I could control everything with four single pole changeover relays. One to turn the ASHP on or off, one to turn the ASHP to heating or cooling mode, one to open or close the 12V valve to the buffer tank and another to switch 240V to turn the UFH valve on or off. For safety reasons I fitted the 240V relay at one end of the rail, separated from all the low voltage stuff.

The thermostats are fairly simple, although when I bought the room stats I could only find one manufacturer, Computherm, who made low hysteresis (+/- 0.1 deg C) units. Now others make similar low switching hysteresis units, I believe. The room stats I used were capable of being pre-set as heating thermostats (operate when the temperature drops) or cooling thermostats (operate when the temperature rises). They are remote radio linked units, which made life simpler in terms of wiring, plus the receivers just had a set of changeover relay contacts, with no mains voltages around. The buffer tank thermostat also has a changeover contact, but is a non-powered, remote sensor type, with the sensor embedded into a hole in the buffer tank insulation, then foamed in place.

The next problem was how to translate all these thermostat contact signals into the right sequence to operate the ASHP and valves in the correct sequence for each combination of thermostat operation, with no chance of setting an incompatible sequence.

In output terms, there were several possible operating states:

A – ASHP off, all valves closed, nothing operating

B – ASHP, in heating mode, but only supplying hot water to the buffer tank

C – ASHP on, in heating mode, but supplying hot water to both the buffer tank and the UFH

D – ASHP on, in cooling mode, only supplying water to the UFH, not the buffer tank

E – ASHP off, but UFH circulating pump running to distribute passive gain around the slab, with the buffer tank isolated.

To do this, I reverted to some fairly simple diode logic, combined with some relay and contact logic, to convert the basic commands, off, heating, cooling, hot water, to the signals needed by the various devices. Here is a wiring diagram showing the +12V part of the feeds to the four relays (each of the four relays has the other side of its coil connected to 0V):

This lot is housed in a small CU box:

Cables from this box run to the ASHP to control whether it should be on or off and whether it should run in heating or cooling mode. Cables also run to the two control valves, one that turns the supply to the buffer tank on or off and one that turns the supply to the UFH heating manifold on or off. To prevent the ASHP giving a fault when running when both valves are closed (which can happen, as the thermal actuator takes round 6 to 8 minutes to open the valve) there is a pressure operated bypass valve that “shorts” the flow and return if the pressure rise is too high.

There are also input cables from the three thermostats, that not only allow correct control, but also provide a “lock out” capability so that, for example, heating and cooling cannot be commanded at the same time.

The input side of the control box is three thermostats, the two room stat receivers (the white boxes to the left of my plug-in power import/export display receiver):

The tank thermostat is a remote sensor mechanical one, and is fitted close to the rest of the UFH manifold and control system here:

The room stats are fitted to the wall in the hall, clearly marked as to which controls the floor heating and which the floor cooling:

The UFH manifold is a standard Wunda three port manifold, fitted with a low energy Grundfos pump that is set to it’s lowest power setting (around 25W). The pump is turned on and off by a simple time switch and runs all the time the time switch is on, as its main task is evening out the slab temperature by recirculating water around the whole slab, moving it from areas heated by the sun to areas that receive no solar gain. The time switch also controls power to the ASHP controls, and currently the switch is set to run from 7 am to 10pm. We may change that when we are living in the house. Here is a photo of the manifold and complete system under the worksurface in the utility room:

The CU box is to the left, the tank stat is at the upper right. The manifold has a thermally actuated valve on the return line, just visible as the white unit in the cut out at the bottom right. The two port TMV is in the flow side of the manifold (the unit with the large temperature setting knob in the centre at the bottom. The sensor for this is inserted inside the upper, flow, manifold.

The box with the displays on it at the top left is a data logger, recording and displaying temperatures from a lot of sensors around the house. The display alternates between two modes, and logs all the temperatures every 6 minutes to an SD card:

 

[EDIT – THIS DATA LOGGER HAS BEEN EXPANDED TO ALSO LOG CO2 AND  RELATIVE HUMIDITY AND RELOCATED TO THE SERVICES ROOM, WITH A NEW LCD DISPLAY IN THE HALL]

So far this system seems to work very, very well. I have it set so that the floor cooling kicks in at 22.5 deg C and the ASHP delivers water at around 12 deg C into the manifold:

The heating thermostat is set to turn on at 21 deg C, and delivers water at around 22 deg C to 22.5 deg C to the UFH manifold.

If the tank thermostat calls for heat (this is really only a winter and shoulder season requirement) then it will usually be set for 35 deg C and will turn the ASHP to heating mode, with the buffer tank valve open, to allow the tank to heat. Heat will flow up from the buffer tank to heat the DHW pre-heat plate heat exchanger to the same temperature. This then allows DHW when there is no PV available to heat the thermal store, as the 9.6kW instant water heater can easily increase the supplied hot water temperature from 35 deg C to the required 42 deg C.

The Genvex MVHR has an inbuilt air to air heat pump, so can heat and cool the air delivered through the supply air system. It can heat to around 1.6kW, and cool to around 1.2kW or so. This is a lot, more than the house needs, but air heating, in particular, tends to deliver dry air, so it makes more sense to use the Genvex mainly for cooling and the UFH for heating. One snag with the Genvex is that it has a fixed 3 deg C hysteresis between the set temperature and cooling cutting in, as an energy saving measure. You can’t programme this out, and frankly it’s annoying. In our case it means that I have the Genvex temperature set to 19.5 deg C, which means that it starts cooling at 22.5 deg and starts heating at 19.5 deg C. This is OK, as the heating requirement is very, very small, far less than the cooling requirement.

I know the above sounds complex, but underneath it all it is really a pretty simple system. It has the major advantage of using non-proprietary control components, so has a lot more flexibility, but it has needed a lot of work to get to the stage where I’m happy to make the system public.

It can be adapted to run other types of systems fairly easily, it’s just a matter of working out the output commands needed in order to convert the input signals from the thermostats to the control signals needed for the particular heating/cooling system.

Edited to add:

The schematic of the heating/cooling system has changed slightly since the last version I posted, in part 33 of this blog, so to keep things up to date this is how the ground floor part of the system looks now:

 

________________________________________________
 

recoveringacademic 02 Sep 2015 06:34 AM :

 I’ve read this four times now; still more work needed on my part to understand it, and I’m clinging on to the latter bit of this sentence.

 ‘…I know the above sounds complex, but underneath it all it is really a pretty simple system…’

 Reading ‘… writing code […] I had better switch to using off-the-shelf parts…’ was a huge relief too. I’m keen on code but only when someone else is paying for it. Perish the thought that I’d actually have to live with the code I write!

________________________________________________

 jsharris 02 Sep 2015 07:06 AM :

 It’s simple in terms of function, it only gets a bit complex when you have to find a way to switch all the things needed for that function, and make sure you can’t accidentally switch other things.

 Take under floor cooling as an example. The input side is simple, just a thermostat where a contact moves from one position to another when the house gets too warm, that tells the system to cool the floor. The complex bit is that for the floor cooling to work, several things have to be made to happen:

 The UFH thermally actuated valve has to be opened, to allow water from the ASHP to flow through the UFH manifold, so a relay has to switch 240V to that valve to open it (they are normally closed and powered open).

 The ASHP needs to be switched to cooling mode, so the relay that controls whether the ASHP is switched to heating or cooling needs to be in the right position.

 The ASHP needs to be turned on, so the relay that closes the contact that turns it on needs to operate.

 Then there are some things that have to be prevented from happening when the floor is cooling. You don’t want the heating buffer tank to be cooled down, so a relay has to be in the right position to make sure that the valve to the buffer tank coil is shut.

 You don’t want the ASHP to be able to go into heating mode, so using the same relay to switch either heat or cool make sense, as that ensures that this selection is either or.

 Finally, you (or I) didn’t want a hot water command from the buffer tank thermostat to over-ride floor cooling, so I made sure that the 12V power to the tank stat is only available when the cooling stat is not calling for floor cooling. I reasoned that if it was warm enough for floor cooling then there would be enough excess PV to heat the water and pre-heat from the ASHP wouldn’t be needed.

  Each of the operating states has a sort of crude set of logic statements like those above. I designed the pattern of thermostat switch connections, diodes and relays to fit a set of four input states (off, floor heating on, tank heating on, floor cooling on), with some being mutually exclusive (the example of floor cooling disabling tank heating) and some being abe to operate at the same time (for example if the room thermostat cals for floor heating and the buffer tank stat calls for tank heating, then both will operate at the same time. If the tank gets up to temperature but the house still needs heating, only the tank valve closes, the ASHP remains running in heating mode to heat the floor).

 The key was to scribble down all the things on the input side, define those that were mutually exclusive, then decode that to the right set of contacts closing and opening from the four relays to make the system do what you want. Much of it relies on the diode logic array to operate multiple relays exclusively from one set of input switches, so, for example, in floor cooling mode one thermostat (the floor cooling stat) switches two relays, the ASHP on relay and the UFH valve open relay. The relay that determines whether the ASHP should be in heating or cooling mode defaults to cooling mode when not on, so doesn’t need to be switched.

 The diodes stop power being fed back to the other inputs, effectively isolating them from what’s going on. There is one odd state that can arise, in that the heating thermostat always over rides the cooling thermostat, so if the cooling thermostat has been set at, say, 20 deg C and the heating thermostat has been set at 20 deg C, then both will operate together at 20 deg. The heating thermostat will “win”, as it has the extra diode that turns on the ASHP heat relay, so despite the cooling thermostat trying to cool the house the heating will over-ride and heat the house.

 This is a nonsensical condition, as in practice you always have the cooling thermostat set to a higher room temperature than the heating thermostat, so they shouldn’t ever both operate together.

________________________________________________

 

Alphonsox 02 Sep 2015 08:38 AM :

 Many thanks for sharing this.

Just to be clear on the system you controlling here – are the schematics you published in Part-33 still accurate or have there been some further tweaks in the last year ?

________________________________________________

 Markblox 02 Sep 2015 09:01 AM :

 Very interesting, quite an achievement, and well beyond most of us. Loxone or similar for most I think, including me.

________________________________________________

 jsharris 02 Sep 2015 10:06 AM :

 The schematic in Part 33 is almost the same. Swap the manual valve on the manifold input with the thermostatic valve and replace the thermostatic valve with a valve controlled with a electrothermal actuator (no plumbing changes, just changes to valve heads).

 The slab thermostat has been replaced with the current system, as I found it just too hard to implement the slab control method. It needed a lot of complex matching of heat loss to slab temperature and only custom software would do the job, and I decided that I didn’t want to rely on a non-off the shelf system.

 The very big problem with systems like Loxone etc is that they won’t actually do what I’m doing here without a great deal of programming and expenditure on extension units. What you find is that the only way to control a Glowworm ASHP is to buy the very expensive Glowworm bus controller, and then that won’t allow cooling, only heating. The same goes for pretty much every make, they all have their own complex (and expensive) control systems and very few are capable of providing the fine degree of control needed for a low energy house. There are few thermostats with a 0.1 deg C hysteresis, for example, and that’s essential. If I wait until the typical half or one degree change has occurred with a conventional control system then there will be a large overshoot or undershoot in house temperature, just because the house has a very long thermal time constant, so takes a long time to lose or gain heat without assistance.

 The ASHP manufacturers are often very reluctant or unwilling to give access to the raw controls of the system. I had to reverse engineer mine to work out that as well as the proprietary bus control system, there was also a set of dry contact controls accessible outside under the cover. There are more than I’m using, I discovered, as well as switching the unit on and off and switching it from heating mode to cooling mode, there are also contacts available for switching the unit to a low power quiet mode for night time operation and an option for running it in hot water mode (where the flow temperature is much higher, at the expense of COP).

 Without knowing about this stuff, no fancy control system is going to be able to gain access to the ASHP controls needed to do what I’m doing. You couldn’t, for example, get the Loxone to do it without doing what I’ve done to reverse engineer the raw dry contact controls in the ASHP, as how are you going to interface its server with the ASHP when you don’t know the bus protocol? Several people have had a go at reverse engineering the control bus that the Glowworm/Kingspan etc ASHPs use to communicate between their controller and the outside unit, but I don’t think anyone has cracked it properly yet. The same goes for every other ASHP (or GSHP probably) they are designed to be tied to a proprietary control system, that you will struggle to break into.

 What I’ve done is bypass this need for proprietary bus control, and I suspect you may be able to bypass other makes of heat pump the same way. This still doesn’t allow a home automation system like Loxone to provide direct control though, as you’re going to need some form of interface (like my box of relays and diodes) to decode the ASHP and valve control signals into something a Loxone, or similar, control system could drive. Loxone do a relay extension unit that you may be able to configure their (£399 relay extension) to do the job, IF you can decode what your heat pump needs and if you can also programme the diode and relay logic sequences I’ve got above into the Loxone miniserver. It would be technically challenging and pretty expensive to get a home automation system to do this at the moment, but could be done if you have the programming skill and are prepared to spend a lot of money.

________________________________________________

 DeeJunFan 02 Sep 2015 10:38 AM :

 I don’t think cooling is going to be much of an issue for me so i think that is going to remove a lot of complexity.

 My wife would happily sleep on top of a stove so hopefully i will only need to control turning the heat on and off.

 Really interesting work though.

 I have looked into a number of home automation systems and they are very like houses you can have all the bells and whistles but it will either cost you a lot of money, a lot of time or both.

________________________________________________

 Alphonsox 02 Sep 2015 11:57 AM :

 It’s the lack of the slab thermostat that surprised me. I would be concerned with the lag introduced by reacting to the air temperature cooling rather than by monitoring and reacting to the actual slab temperature. However I do agree that an off-the-shelf system is highly desirable.

 Wouldn’t a simple thermostat in the slab in place of your room thermostat give a faster response ?

________________________________________________

 ProDave 02 Sep 2015 01:38 PM:

 There is just one thing that seems to be missing from this that I had intended to implement in my own heating setup:

 Namely ASHP temperature set point.

 I envisaged running the ASHP EITHER to heat the UFH tank at a low set point temperature (to get maximum COP) OR running the ASHP to heat the hot water tank at a higher set point and accepting the COP would be worse.

 I’m interested to know what sort of set point you have, and why you don’t appear to ever vary it?

 I’m of course assuming / hoping it can be adjusted between two pre set temperatures using one of the volt free contact inputs.

________________________________________________

 jsharris 02 Sep 2015 03:48 PM:

 Answering the slab temperature question first – the answer seems to be that the slab stays very stable when run from a room stat. I don’t know why, but it is more stable run like this than it was when I was running the slab thermostat!

 The ASHP is programmable to death via the command unit box, with weather compensation curves for both heating and cooling, including the ability to write your own curves with three set points plus an offset. It uses it’s own outside air temperature sensor to do this, although it isn’t exactly easy getting your head around how the settings work in practice!

 Interestingly, me going in and controlling the ASHP directly via dry contacts, rather than using the bus, doesn’t affect these settings, and the programmable timer functions on the command unit (that allow things like night set back, going to silent mode at certain times, all work the same. The dry contact controls do over ride the programmable on/off time and day settings, though.

 At the moment, I’ve found that heating above 35 deg C makes the COP drop off, as the thing starts defrosting in humid weather and that REALLY hits performance (because the heat pump reverses to defrost, so can take around 1/3rd to 1/2 of the heat in your buffer tank out in one go just to melt some ice).

 I’m running at a fixed 35 deg C for now, as that temp happens to be fine for DHW with the inline booster heater. Even if we get no sun for days, running the ASHP at 35 deg C is plenty to give large volumes of water at 42 deg C from the booster (bearing in mind that the ASHP can recharge the buffer at around 5 kW whilst DHW is being drawn off, plus I have the thermal store capacity plus the buffer capacity for DHW.

 For cooling I’m running at a fixed 12 deg C, again just because it gives a reasonably good COP and anyway the ASHP is only ever running at a low power setting to deliver that.

 Yes there is a hot water dry contact, that runs the ASHP at DHW temperature (around 50 deg C I think). I’m not using it, because it really slugs down the COP and it’s better to keep the tank at 35 deg C and use the booster heater in running cost terms. The impact on COP of asking for high flow temperature water (anything over 40 deg C) has to be seen to be believed. I had it set for 40 deg once and wandered by and saw that the buffer tank, that had been around 35 deg C ten minutes or so before. was now down at 23 deg, as the defrost cycle had reversed the heat pump and lowered the buffer temp by around 12 deg C in ten minutes or so!

________________________________________________

 jsharris 02 Sep 2015 06:50 PM:

 I’ve just edited this entry to include an up to date version of the ground floor heating/cooling system schematic, which may make the description of the system make a bit more sense!

________________________________________________

 Alphonsox 02 Sep 2015 07:50 PM:

 Thanks – That makes things clear.

________________________________________________

 notnickclegg 03 Sep 2015 06:05 PM:

 Thanks for this Jeremy, it’s very helpful. I, too, was surprised and very interested by your move to air-based rather than slab-based temperature sensing. I’ve got recesses in the slab for 1-wire temperature sensors, but I’ll be sure to try out both options to see whether your experience is replicated in our house.

jsharris, on 02 September 2015 – 10:06 AM, said:

What I’ve done is bypass this need for proprietary bus control, and I suspect you may be able to bypass other makes of heat pump the same way. This still doesn’t allow a home automation system like Loxone to provide direct control though, as you’re going to need some form of interface (like my box of relays and diodes) to decode the ASHP and valve control signals into something a Loxone, or similar, control system could drive. Loxone do a relay extension unit that you may be able to configure their (£399 relay extension) to do the job, IF you can decode what your heat pump needs and if you can also programme the diode and relay logic sequences I’ve got above into the Loxone miniserver.

 Loxone’s ordinary digital outputs (eg, on their Miniserver or ordinary extension) are in fact 250V 5A relays, which you can use to switch whatever voltage you supply to them. Accordingly, there’s no need to use a separate relay board for switching. From memory you get 8 of them on the Miniserver.

 Inputs are slightly trickier, as you’d need some sort of contact to convert any switched 240V signal to a 12-24V DC input.

 The programming module itself is easily flexible enough to program in all the diode logic and relay control. It has a full complement of logic gates, counters, flip flops, fuzzy logic and PID controllers, multiplexers, comparators, programmable heating curves and a lot more, all of which are programmed via a simple but powerful visual block language.

 Of course, this all assumes you can decode the dry contact language of your ASHP or other device. As it happens, I have info about these contacts for both the Panasonic ASHP we’re looking at, and the Brink MVHR unit we’ve acquired. That said, I’m not yet sure whether I’ll integrate them via Loxone or not – in some ways, it may be better (and certainly simpler) to leave independent systems to operate independently.

 As you say, Loxone would be an expensive option if all you planned to do was control an ASHP or the like, but if you’re having Loxone for other stuff anyway (as we plan to), it’s probably just as easy and not that much more expensive to keep it all integrated in the one system if you wanted to go that way.

 Thanks again.

 Jack

________________________________________________

 ProDave 03 Sep 2015 08:18 PM:

 Thanks for the info on the defrost cycle. I hadn’t even thought for one moment it would take heat OUT of your house to defrost itself. I just thought it would have a low power heater in the unit to gently defrost the heat exchanger. I have a bad feeling that for a large part of the winter up here an ASHP may end up pretty useless.

________________________________________________

 jsharris 03 Sep 2015 09:05 PM:

 Yes, the majority of ASHPs just run in reverse to defrost, so effectively switch to cooling mode (it’s a good clue that they can be used for cooling as well as heating). Some have heaters, but these end up using more power.

 The staggering discovery for me was how long an ASHP spends defrosting under specific conditions. Cold is fine, its cool, wet, conditions that are the problem, but only when combined with a high flow temperature demand. The spec COP temperatures are very carefully chosen to show off the best performance of an ASHP. The worst case is when the ouside air temperature is around 3 or 4 deg C, the RH is up over 60 to 70% and the ASHP is being asked to deliver 40 deg C plus flow temperature.

 It’s interesting that this isn’t a performance point that’s quoted by the manufacturers, yet is a condition that isn’t that uncommon for a badly set up system.

 I’ve found that there is a massive performance improvement by keeping the flow temperature to less than 40 deg C, so massive as to practically double the performance of the system when it’s cool and wet.

 It’s this that makes my backup system of heating to 35 deg C, then boosting that to 42 deg C on demand, a viable option when there’s not much sun around.

________________________________________________

 Alphonsox 04 Sep 2015 07:01 AM :

 Was there any particular reason to pick a 240V actuator over a 12V version ? I would have thought that keeping the whole DIN rail assembly to 12-24V might have been desirable.

 Any details on the data logger ? That looks particularly useful.

________________________________________________

 jsharris 04 Sep 2015 06:18 PM:

 You’re right, running everything at 12V or 24V would have been better, but I already had a 240V NC thermal actuator, so just used it. If I can find a faster acting one that runs on 12V then I’ll look at changing to it, as the standard 240V one I have is very slow, it takes around 6 to 8 minutes to open.

 The data logger is a home brew thing, the first version is described here: http://www.picaxeforum.co.uk/showthread.php?20551-Environmental-uSD-card-data-logger  The version I’m using uses the same circuit board, but has a 20 x 4 line display to show more data and also uses the SD card version of the µlogger µSD interface that I used on the original (just a cable to extend the leads out to the bigger SDlogger module, mounted in a slot at the bottom of the case).

 The code’s been chopped to read just temperature sensors, as I found that more useful for this job. The unit happens to fit inside a deep double socket surface mount box, with a blanking plate modified to accept the display cut out.

 I also made a portable logger that has a limited set of sensors, aimed at measuring and logging air quality in a room: http://www.picaxeforum.co.uk/showthread.php?23893-Air-quality-monitor-and-logger  This is handy, as it can tell you pretty much how well your ventilation system is working at maintaining not just a comfortable temperature but also a comfortable humidity and CO2 level

________________________________________________

 Alphonsox 04 Sep 2015 07:30 PM:

 Uponor are the only 12V versions I’ve found. There seems to be a larger selection of 24V manufacturers.

 Thanks for the data logger links – more to read !

________________________________________________

 notnickclegg 16 Sep 2015 09:27 PM:

 “jsharris, on 02 September 2015 – 03:48 PM:, said:

The ASHP is programmable to death via the command unit box, with weather compensation curves for both heating and cooling, including the ability to write your own curves with three set points plus an offset. It uses it’s own outside air temperature sensor to do this, although it isn’t exactly easy getting your head around how the settings work in practice!

Interestingly, me going in and controlling the ASHP directly via dry contacts, rather than using the bus, doesn’t affect these settings, and the programmable timer functions on the command unit (that allow things like night set back, going to silent mode at certain times, all work the same. The dry contact controls do over ride the programmable on/off time and day settings, though.”

 Supplementary question if I may Jeremy:

 If I understand your setup correctly, you’re leaving the ASHP controller to tick over doing its own thing (other than to the extent that it’s overridden by the dry contacts). Does the use of a thermostatic mixing valve mean that the weather compensation effectively has no impact (ie, because the temperature of the water fed to the slab is controlled/limited by the TMV)?

 Thanks

Jack

________________________________________________

 jsharris 17 Sep 2015 06:36 AM :

 Yes, the weather compensation isn’t doing anything, in fact I’ve programmed it to be a constant temperature using the custom curve option. So far I can’t see that weather compensation that uses outside air temperature makes much sense for a passive house, as it isn’t really the outside air temperature that governs the house heat loss/gain at all, but sunshine, or the lack of it. Even on very cold days the solar gain can be more than high enough for the house to not need any heat at all, it may even need cooling on these days.

 At the moment I’m running the ASHP with a custom heating curve that’s a straight line at 35 deg C flow temp, and a custom cooling curve that’s a straight line at 12 deg C, irrespective of outside air temperature. This seems to work fine although the heating characteristic of the ASHP is such that the flow temp does overshoot 35 deg C to around 38 deg C at times, probably due to the way the ASHP modulates. I found the same when I had it set for 45 deg C, in attempt to get usable hot water from it; it would overshoot to around 48 deg C at times, though it would ice up and go into defrost mode frequently when set that high, which is why I have it turned down to 35 deg C.

________________________________________________

 notnickclegg 17 Sep 2015 07:54 AM :

 “jsharris, on 17 September 2015 – 06:36 AM, said:

Yes, the weather compensation isn’t doing anything, in fact I’ve programmed it to be a constant temperature using the custom curve option. So far I can’t see that weather compensation that uses outside air temperature makes much sense for a passive house, as it isn’t really the outside air temperature that governs the house heat loss/gain at all, but sunshine, or the lack of it. “

 Great stuff, thanks Jeremy. My gut feeling was that external weather probably wouldn’t have that much of an impact, but good to hear your confirmation.

 Thanks

Jack

________________________________________________

 

Leave a Reply

Your email address will not be published. Required fields are marked *