Achieving the highest efficiency from your heat pump requires attention to your radiators. When set correctly, the pump will run more smoothly, with less noise and reduced ‘short cycling,’ leading to greater comfort due to more even temperature control in your rooms.
I would also recommend setting your air source heat pump to ‘Weather Compensation‘ and adjusting your ‘Leaving Water Temperature’ (LWT) to a moderate setting, starting with 35 degrees Celsius for the central area of the ‘slope.’ More guidance on setting ‘LWT’ can be found elsewhere; this article focuses exclusively on the ‘Lockshield Valve’ setting.
The lockshield valve acts as a ‘flow regulator,’ controlling the amount of heat supplied from the flow pipe into the radiator. It should not be confused with the Thermostatic Radiator Valve (TRV) fitted at the other end of the radiator. The TRV serves as a temperature limiter, reducing the flow of heat only in extreme circumstances, such as solar gain heating up a room. Otherwise, the TRV should not restrict the flow.
To begin setting up the valves, ensure all valves (both lockshield and TRVs) are fully open. Ideally, this task should be performed in colder weather to simplify the procedure. At this point, the heat pump can be turned on. Then, feel each radiator to determine which is warming up the most—make a note if necessary.
The radiators, especially if newly installed, should have been sized according to room size and calculated heat losses, along with the desired room temperature. Starting with the radiator that warmed up the most, gradually close the lockshield valves from room to room. Initially, you might close the valve by two full turns, with final adjustments usually being half a turn or less from the fully closed position. Continue adjusting each lockshield on the next warmest radiator.
Allow time for the radiator temperatures to settle after each round of adjustment. If some radiators remain warmer than others, they should be the first to be closed further. This process, requiring multiple steps of adjustment, aims to restrict flow minimally until each radiator sufficiently heats the room, allowing the rest of the heat to flow to other radiators. For instance, we prefer a warm bathroom and keep the lockshield valve fully open on this radiator, which is the first fed from the secondary pump circuit, keeping the room very cosy.
Once the radiators provide proportionate amounts of heat for the rooms, allow the system to settle for about a day. Then, you might find some lockshield valves need a bit more tweaking. Proper adjustment takes time; although installers may make initial adjustments, fine-tuning to your comfort level is crucial.
Consider using thermometers in rooms and a temperature probe on radiators to gauge progress. An infrared handheld thermometer can also assist in assessing radiator temperatures and the efficacy of your adjustments.
If using a room thermostat, ensure it’s correctly set so the radiator in that area warms the room to, but not above, the set temperature. Setting the thermostat slightly higher than desired can initially help avoid frustration. Once radiators are balanced, return to the heat pump controls for any final adjustments.
TRVs should now be set to start shutting at approximately two degrees above the required room temperature. Under normal circumstances, TRVs will not disrupt the heating system but will activate if the room experiences solar gain or another heat source.
An efficient heat pump system prefers a constant load and resistance to water flow; while a lockshield valve provides consistency, a frequently adjusting TRV can reduce the system’s efficiency and increase operational costs.
*Note: Installers might set a low temperature heat pump to a leaving water temperature of 45 or even 50 degrees Celsius by default, even though a lower temperature is more efficient and still provides adequate heat.
Please note: While I am not an expert, plumber, or heat pump installer, I had an ASHP installed in 2023 and have carefully adjusted it after the initial commissioning. This required waiting for colder weather in December 2023 to fine-tune the lockshield valves and pump speed, achieving a COP well above 4 most of the time.
At installation, the smart controller was not available, so a basic thermostat was installed. Setting this thermostat a few degrees above the required temperature prevented frequent on/off cycling. After installing the smart controller, the temperature is maintained within 0.6 degrees of the desired level. I also repurposed a programmable TRV from a previous gas boiler system for the bedroom radiator, allowing for temperature adjustments at night, while the other nine radiators remain with TRVs fully open.
@Potatoman this article from Toodles should be very helpful in light of your recent request.
It is probably better to balance each radiator to achieve the desired room temperature, since the heating capacity of each radiator is unlikely to be directly matched to the heat loss of that particular room.
In a room with a physically smaller radiator the opening of the lockshield valve, and hence the flowrate through the radiator, will need to be greater to achieve the desired room temperature.
@Derek M Perhaps I should have worded my guide a little more precisely; when I said
‘Once the radiators provide proportionate amounts of heat for the rooms, allow the system to settle for about a day.’
What I was trying to indicate is that the heat from each radiator should reflect the heating needs of the rooms they are in. I suspect I am also guilty of thinking that the installers have done their survey thoroughly and accurately and that each radiator is very close to optimum capacity for the room it is fitted in. This of course may be fanciful – but isn’t this why we have the MCS so they can ensure the the installers do their job well? 😉
Tongue-in-Cheek, Toodles.
Unfortunately the type and size of heat emitters is determined by the proposed LWT, the available space and the room heat loss calculation, with the later being very subjective.
@Derek M I was probably very fortunate inasmuch that the eight radiators we needed to replace with greater capacity ones could all be fitted within the wall space we had available; by using K2’s, in most cases, they were no higher or longer than the ones we removed. I’m a bit of a ‘don’t mind what it looks like, will it do the job properly?’ type of person and so is my wife – we find pragmatism has great advantages! Regards, Carefree Toodles.😉
I have 12 loops that seemed not balanced at all when we bought our place. I bought 12 cheap LCD thermometers (20 Euros or so) with sensor on a wire and pasted these to the return flows. From there on, I started to adjust the flow in the loops in order to get them all to the same return temperature (with the bathroom loop a bit higher). My thinking was and is, that if the return temperatures are the same, the individual loops get enough energy supplied at the front end to get the same return water temperature. Maybe this also works for your radiators.
@Toodles I used a similar method to this last October and it works a treat. One thing I did find very helpful is a thermometer with a logging capability so you can see how each room behaves over a 24/7 timeframe, so fine tweaks can be made if necessary :-).
Logging thermometers are also great for getting to know your systems heating characteristics. For example, our house takes approx 4 – 5 hours to heat from 19.5 to 21 degrees so I know to set my smart thermostat to 13-30hrs to start heating to 21 degrees, so when the Wife sits down to watch the tele at 7pm she is not cold and I have not had to have the heatpump running all day to maintain 21 degrees (19.5 is our set back temp which is ok for the morning but not evening)
@TechnoGeek Yes, I invested in 8 Govee logging thermometers (logging in the sense that they have Bluetooth capability and can send the log to my iPad for scrutiny.) Regards, Toodles.
I have just carried out your advice and I am within .5c of whatever the Rads I want at maximum in relation to room temoerature. How do determine WL ON THE FLOW C of Heat pump. presently 44c and has been 50c for some years.
Is it just a case of whether they reach the thermostat setting of what ever I am asking ?
@Alfapat Well done, you are well underway but keep the LSV spanner/key handy for any tweaking you might wish to carry out later! Yes, see how the comfort level is for a day or two and if all rooms are still warm enough to your liking, then you might wish to lower the (what I know as leaving water temperature and I think you know as water law) another degree then wait and see. If you find that most rooms are still at your comfort level but perhaps one has become just a tad cool, you can tweak that LSV open a ‘nadgers’ [very technical term there!]. You can rinse and repeat until you find that the whole house is starting to drop slightly below your comfort level for all rooms. You then just take the LWT/WL up a degree again! Keep us informed how things go. The final stage may be to adjust any rooms that you wish cooler when there is solar gain or for ‘cooler at nightime’ such as a bedroom using a TRV. Regards, Toodles.
Thanks ! Head banging , haha !
Moving on a bit , I had all the rads where I wanted them at WL 44C. Rooms near the Movable room stat 22 . Smaller ones near 20c (Bedrooms) mainly.
In order to get the most economical flow temp I dropped the temp to 38c . The larger rooms were hardly reaching the spot by 1 degree. I also noticed the Kwh had risen on the smart meter .(I assume taking longer to get the room stat to agree and not reaching it) Overnight the two smaller rooms which were set to 2 degrees lower that Sta requirment were cold .
Clearly Wrong !
So I have put the flow temp back to 44c .
Incedently The room thermostat is always placed in the hall overnight where the open rad is stuated , I m not sure that this is the best place to place this , because if it last in line of flow returns then its going to take more Kwh to maintain the rest of the house temperatures.
Another trial to carry out and monitor.
There are several steps to getting the most efficient settings and you need to have faith to see them all through
1. Disable any external controls (TRVs and Thermostats) by turning them up to max
2. Ensure that any room influence function on your heat pump controller is disabled
3. Balance the rads for equal room temperature, or the required differential between room temperatures, dont worry about absolute temperature
4. Slowly turn down the WC curve until the house is just warm enough
You can do 2 and 3 together and you will need to iterate them over several days or up to a month to get it right. When you change (3) you need to allow time to stabilise, initially you can change by a degree or two every 6-12 hours but once you get close you need to change by a degree every 24 hours. When you change (2) you likelwise need to allow time to stabilise.
Before you do this you should read the section on WC in your manual, and make sure you understand which parameters to adjust and the principle of operation. If you dont or are not confident (they can be quite confusing) please ask.
Whilst you are going through this process, comparing kWh from one day to another is futile, as the conditions (outdoor temperature) change, which affect the consumption more than any tweaks you do (the exception is if we get a period of several days with very steady temperatures, which does happen occasionally, but never when you are doing this type of adjustment!). This means that you have to trust, maybe for several weeks, that lower flow temperature equals lower cost, until you have data over a range of conditions both before and after that enables you to make a genuine like for like comparison (which is still tricky). If you aren’t prepared to do this (which I can understand), then you cant possibly optimise your system. However if your flow temperature is 44C you will almost certainly pay about 12-18% more for your heating like for like than if its set at 38C.
Ah Ok , so when I think I have balanced rads at say to 22c then set room stat above by 2 degrees and if it makes the room temp too warm turn down lockshields to suit required 22c?
Do I leave flow at 44c or continue to try 38 and open cold rads a little more?
Our posts crossed so please read/re-read mine above also
In answer to the question, No, once rads are balanced leave them, and turn down the flow temperature/WC curve to get the right room temperature, with the room stat turned well up (I would say at least 5C above the desired temperature while you are doing the tweaking.
Your end point is
rads balanced but with the LSVs as open as possible consistent with being balanced (which means that at least one should be fully open)
flow temperature/WC curve just sufficient to get thee rooms to right temperature with no influence from external controls/trvs/thermostats
OK that complicates matters. What make is your heat pump? Can I presume that the TRVs are time-controlled, or are they basic TRVs?
I ask because some heat pumps have a time control that is ideal for Time of Use tarrifs like Cosy, others are more crude so you have to be a bit inventive (which is where your TRVs may yet come in).
if the room goes above desired 22 and the room stat programme is at 24 c Then adjust the WC Curve down
Water law.
No need to apologise, heating is complicated!
What exactly is the Danfoss programmed to do? Is it switching the heat pump on and off on a timer or if not that then what exactly.
Is there a reason you are not using the time programmer on the Samsung controller itself? Have you got the controller manual/can you post the exact controller model number?
Can you describe the timing you have set up for heating and DHW, is the heating (for example) on only during the Cosy cheap rates, or something else.
Sorry for the questions, Im just trying to understand what you are doing/attempting to do so I dont mislead you!
OK makes sense
How do you achieve the above, is it by switching off immediately before the boost periods or something else.
@JamesPa Its never off , its either on a degree behind or at the same temp moving to next period , so I don’t really think its and advantage Aye those Heat Pump Tariffs. .
Sorry but I dont understand how you achieve this with the timer and manual TRVs. How do you tell it to be a degree behind, or is this just what it does because of random variations.
Unless you have a controller designed for the purpose (ie Homely at present) or one of the heat pumps with a more sophisticated inbuilt controller I think you are very likely right.
@JamesPa s only what the room thermostat says when it moves from one of four periods , generally it doesnt always achieve this . TRV’s completely open now according to new advice. Wow thats been 8 yrs set up wrongly!
Now the room thermostat is set for 24 and finally the Kitchen is at 22c. So I will look at other rooms now.
OK I get you now, you are using the scheduled programmable temperature feature on the Drayton. This will switch thee heat pump on or off according to the programmed temperature and the measured temperature.
I will give some thought as to whether there is a neat way to make things work with Cosy, but a crude on/off function (which is what the Drayton does) is never a good start when it comes to heat pumps!
TBH if I had a Samusng heat pump and wanted to do this sort of thing I would probably get a Homely. However I dont, I have a Vaillant heat pump and a ‘simple’ 7hrs cheap tarrif at night, so just programming a nighttime ‘set forward’ (which the Vaillant controller can do) is the way forward. I haven’t bothered to do this as it happens but probably will when we get to the shoulder season.
I think your way forward is to keep TRVs open with balanced radiators, the programmer set to higher than you need 24×7 and the Water Law adjusted to get the temperature you want. In other words get it working stably with the Water Law being the only control and turned down as low as possible.
Once you have this working stably you could possibly to overlay a timed switch off in the couple of hours before the cosy cheap rate (and during the Cosy expensive rate). This will require you to jack the WL up a little, but hopefully not too much.
Can I just check that you are happy you know how to adjust the Water Law. I think its FSVs 2011, 2012, and either 2031, 2032 or 2021, 2022 depending on the value of FSV 2041. Is that what you have. Your controller is older than the current one (which is the one for which I have the manual)
Thanks James , hopefully getting there. By the way its a Danfoss not a Drayton , probably similar though.
Yes its 2021 , so I will put down the WL 1 degree and see how the room temps fare. Its kind of what I have been doing but Room thermostat has been tight to the mark . I wll see how we go.
Thanks for coming back on this. As you have probably gathered Im not a Samsung expert, just an amateur who has been studying this stuff for a couple of years in prep for my own heat pump, which I now have. Anything I say is offered in good faith, but in the end its your decision.
Please don’t do anything you aren’t comfortable with and please make a note of settings you change just in case something needs to be set back. The Samsung WL interface is, like many others, not a model of user friendliness, but you are on the right track adjusting 2021 while its cold. It should be possible to get it so that the house gets to temp at more or less any OAT, with the TRVs off and thermostat well above desired temp, without having to continually adjust the WL, but that will take some tweaking over a period of time.
Another thing worth mentioning about TRVs is that, if they happen to be shut when the unit does a defrost, they can deprive the unit of the hot water it needs to do the defrost. This is another good reason for setting them above the desired temperature. Of course you may have a buffer or volumiser that provides water for defrost also, although the former have their problems to say the least!
I saw somewhere else on this forum that Samsung basically doesnt allow you to schedule setbacks etc, but there is a ‘PV boost function’ that is adjustable and can be triggered by an external switch. This raises the FT by an adjustable amount while the switch is closed. You could potentially use the Danfoss timer to operate this during cheap rate periods, which is after all what you basically want (to force it to work harder during the cheap periods, so it can slow up during the expensive periods. I will try to find the thread and maybe make a connection in case someone else has already tried this.
Whatever you do Im pretty certain that starting from a position where the water law is basically optimised and the radiators balanced should be better. Really installers should do this but it takes time, time is money, so they opt for something that is quick but not as good. To overcome this the industry basically needs automatic radiator balancing and automatic water law optimisation both of which are beginning to emerge but not yet ready for prime time.
Please keep posting!
Yes there is a buffer tank , I ve been reading about them as not being essential , but will leave that for the moment . What do you mean by when you say that Samsung doesn’t allow scheduled setbacks please. PV’s being a Pressure Valve boost function. Flow setting is 43c at the moment and large rooms still very warm . I will leave till tomorrow. My conservatory is next to lounge and very cold with minimum opening on both rads in there , lounge a bit colder as a result, 1.3 degrees less than next room with door open . easy sorted with wood burner.
Its becoming clear (and installers are beginning to come clean on the matter) that buffer tanks are unnecessary in 95%+ of domestic situations. They come with a performance penalty of up to 15% or thereabouts, and make fault diagnosis difficult. Historically they have been fitted to prevent problems which otherwise cause call outs, but really they are masking them not preventing them. The more enlightened installers now admit this, but many are still clinging to the old ways. When @Mars did a video on buffer tanks (on this site) he managed to find three installers who would join to advocate against buffer tanks but none of those he invited who advocate for would take part. That says it all!
2By buffer tank I mean a 3 or more usually 4 port tank (flow and return from heat pump, flow and return to heating system). 2 port volumisers are useful and have no performance penalty. These are fitted in series in either the flow or the return (arguments for both) not between flow and return. Turning a 4 port buffer tank into a 2 port volumiser is something worth considering.
Re “What do you mean by when you say that Samsung doesn’t allow scheduled setbacks please. PV’s being a Pressure Valve boost function. “
The Samsung controller does have a built in scheduling function, but it only triggers on/off operation of the heat pump using its internal temperature sensor (basically it does the same as your Danfoss but using its inbuilt sensor and timer function). What you really want is a function which, rather than switching the heat pump on and off, turns the flow temperature up or down according to the a schedule (to follow Cosy or for night time set back). Some heat pumps (Vaillant, Daikin, strangely not Mitsubishi) support this, others don’t. Samsung doesn’t.
PV in this case is photovoltaic – solar cells. The Samsung function boosts the flow temperature by an adjustable amount if a contact on the control board is closed. They call it a PV function, I suppose their idea is that if you have free electricity from PV you bump up the FT to take advantage. Yours is a relatively old model so might not have it, but if it does its a way that what you really want to take advantage of Cosy could potentially be achieved.
Re your lounge and conservatory obviously I dont know what you want to achieve but you know how to adjust the lockshields so you know what can be done. There is a limit to how much temperature difference you can get between two adjacent rooms and, counterintuitively, it is usually increases costs to leave rooms unheated (explanation and worked example on Heat geek site).
Keep going until you have the Water Law curve as low as it will go consistent with your house being warm enough. Thats the optimum efficiency point and the best starting point by far for any tweaks.
Hope that helps, I will post my thoughts on your ‘Cosy’ problem in a day or two.
@Alfapat
I hope you are getting closer to the optimum adjustments.
Ive given some thought to how you can operate the optimised system with Cosy. I think there are 3 options, all starting from the situation where you have stable operation at your desired temperature on water law only as low as it can be (I will refer to this as ‘optimum’), with no external controls operative
1. Adjust the Water Law to a degree or so above the ‘optimum’. Then use the Danfoss more or less as you were previously, setting it to a bit higher than you want during the cheap periods and the temp you want during the expensive periods. Leave LSVs open
2. use the Danfoss to turn the heating off during all or part of the very expensive period, and the periods just after the cheap period. Adjust the Water Law up as necessary to get an acceptable temp (should need only a degree or so above the ‘optimum’ – a bit like the previous)
3. If your heat pump supports it, use the Danfoss or an external timer to set the PV Boost function during the cheap periods and just before the very expensive period. Turn off the heat pump during all or part of the the very expensive period and for an hour or so after the cheap period (you could use a second channel on the Danfoss or the heat pumps own scheduler to do this).
I think (3) is probably the best but depends on your heat pump supporting the PV boost function and its also definitely the most complex although at the same time in a way the ‘cleanest’. If you can and do decide to choose 3 I have some thoughts on how you might decide how many degrees to boost by and how long you should turn off by, basically using the ‘degree minute’ concept that at least two models of heat pump use as part of their control loop.
Im happy to contribute further on this if you want, its interesting to me to think about this problem as I have yet to program my heat pump to take advantage of the tarrif I have (EON Next Drive, essentially a variant on economy 7, giving me 7 hrs at 7p/unit overnight in return for ~10% uplift on the daytime rate- Great for charging an EV, averages out at less than the standard rate for a ‘constant load’ like a HP so overall not bad if you have an EV)
Morning James , Iam at 42c and just changed to 41c . The overall temperatures in the house are managing so far with the thermostat 2 degrees at every stage above the the normal 21/22c . The power usage is more levelled on the smart meter but has risen in the last 3 days of this adjustment . . So will see , it is -4c this morning but not seeing that graph till tomorrow.
I will think about your options but very unsre if they will work , especially the PV ONE Aas I dont have photovoltaic . I m sure that was your explanation of that abbreviation.
Ill work on getting WL down first.
You dont need actually need to have solar panels (photovoltaic) to use this option, you are just using a function of the Samsung (if yours has it) called photovoltaic, which in fact is just a contact that you can close to boost the flow temperature. I think this option, if it available to you, has the best chance of working, but is undoubtedly the most complex.
The bottom line here is that the heat pump controller, like most, is not designed for ‘tariff optimisation’, so whatever you do is a bit of a kludge. The only controller I know that is designed for tariff optimisation is Homely
There is actually another option for tariff optimisation – Havenwise. They now apparently support Samsung – take a look at their website. You need an internet interface to your heat pump but then its a service in the cloud (as is Homley at least in part).
Update.
So far on Wl settings for the last 24 hrs is at 41c.
I get it that looking at the smart meter recordings shouldnt be taken as read. The average temps in my area in the last 14 days have been around 5c duribg the day and 2c at night last three a little lower with frost . This could be an indication why my energy usage is still a little high despite dropping down WL from 45C to 41c . I can only relate this to HP adjusting things in order to give the same house temperatures .
The most recent night (morning of 23rd Jan) it was down to -5c outside so the total for 23rd was 79kw the 22nd which had WL SET at 42c the energy used was 82kw. I cant work out what is happening yet so I will leave at 41c
Foot note for the 17th Jan I only used 38kw and looking at temperatures for that day they were much the same 5c and 2c at night . WL was 44 or 45c around these dates ?
Head Banging!
Which is why I ignore individual daily readings completely and plot daily consumption Vs average oat over a period of weeks at least. The scatter on this is sufficient (at least in my case) to prove that looking at individual days is fruitless, and the only real approach that can be taken requires monitoring over several weeks to months.
The underlying problem, I think, is that houses have memory of 24 hours or more, so two identical days are in fact very different if the preceeding day or two was different. Also wind and solar gain matter.
How you deal with this is up to you if course.
Thanks James , just trying to get assurance thats all ,and trying my best within my capabilities.
I will stay static for a whille now.
Cheers!
Understood. Unfortunately, because we can’t control the environment, we all pretty much have to trust the specs and the physics. Fortunately all heat pumps have similar characteristics because they all depend on the same physics.
I have personally found that operating on weather compensation alone gives me a much better level of comfort than I ever had with my previous gas boiler and complicated controls, and I look to be on track to save perhaps 10% in operating cost. So far as I can tell, the physics is working for me as it should. Of course that’s just me and my system.
Hi @Alfapat
I hope you are getting close to optimising your system.
There is a discussion on this thread https://renewableheatinghub.co.uk/forums/thermostats-trvs-pumps-valves/how-to-balance-radiators-the-role-of-the-lockshield-valve/paged/4#post-40825 about how to use a Samsung controller to effect a setback. Also some information about times setbacks using the Samsung controller alone, which is almost certainly the best way to do it if yours has the capability (it may be available only on more recent controllers, but perhaps firmware upgrade is a possibility)
@JamesPa Thanks for your interest again . I thought I was , were I was down to 41c flow but ultimately although the temp in more consistant the energy needed doesnt seem to drop . Currently I have reverted to 44c where I got better results.
Quite interested in your link but it takes me back to the same page we are on, Ill look again
Incedently I am trying to find out if Homely is compatible with my 8 yr old pump but no answers yet
Here is the corrected url to the other thread https://renewableheatinghub.co.uk/forums/postid/41097
Without vastly more information about your system and the measurements that I care to ask (and likely than you care to give) I wont attempt to speculate on the explanation for your energy measurements, but the most important thing is that you are happy and warm.
I will have a look , thank you . Hey , warm , but not happy yet, don’t worry , keep the thoughts ongoing! 😏
Yeah , interesting , looks like Honely is the nearest to solving my system, if it’s compatible with Gen 5 .
Homely gets good reviews on this forum. The Samsung controller historically is a bit lacking in features (maybe the later ones are better) and homely fills the gap, albeit for some of its more advanced features requires a subscription. You have to fit the modbus interface to the Samsung I believe as well.
Help required here , I have had two calls from Homely and I still cannot find out whether Mod Bus is compatible with Gen 5 Samsung Heat pump.
Your Link to Grahams Heat pump Weather compensation video mentioned a Hand over book /manual for Samsung Gen5 . It seems impossible to find on Samsung’s site. Can anyone help.
Something not right yet , In the middlle days of Jan the average temperatures were 6c and 4c they are not that different now . In my quest to get the flow temp down I have been down from 45c to 41c . It did look like that the system struggled to keep things the same at that.
Seeing that I had smaller usage of energy , up to half that ofthe last 4 days I put the flow back up to 44c . Figures I am looking at are in the middle of January between 40 and 60 kwh and last 4 days 70kwh plus .
Something not right.
By ‘flow’ I presume you mean FSV parameter 2021. I am assuming you have left 2011 constant (at what value)
As I mentioned earlier, consumption over anything less than several days is pretty meaningless because the house has a memory of 24-48 hours (sometimes more), and consumption over several days is also meaningless because the conditions vary too much. Pretty much the only way to compare consumption is by plotting vs OAT over several weeks or more.
As I recall you put the FT back up because the consumption had gone up when you put it down, now you are finding that it goes up when you put it up! I fear the problem is that you are looking at the scatter on the figures rather than anything particularly meaningful. Obviously I cant be certain about this because there are features of your system, most notably the buffer, which might behave in an unexpected way, but in the absence of anything odd scatter on the figures is the most likely explanation.
Genuinely the only way I know to do this is to trust the science. Open up all the TRVs, set the thermostat as high as it will go so its constantly calling for heat, set 2012/2022 to a suitable pair of values for the warm end then adjust 2021 down until the house just keeps warm. Monitor over several weeks and tweak slightly up/down as necessary to get the right value. Check the daily but more importantly weekly COP to see if its sensible and address any obvious problems if any. If you are happy with the consumption enjoy!
If you arent prepared to trust the science I think your optimisation process is going to need at least a season, possibly two, so you can do a meaningful plot over a period of weeks for each change.
To give you and idea Here is a scatter diagram for my system over a period of 3 weeks in January. Its plotted vs degree-days, a slightly modified average OAT. Changing FT by 3C will result in 6-10% change in efficiency. The scatter, particularly in the interesting region where most of the points are, is probably double that indicating that more data would be needed to dig out of the noise the effect of changing the FT. No changes to the system were made during the period over which this data was collected.
Sorry I don’t understand the graph, and I don’t know how work out COP. What I can tell you are my Settings 2011 is -5c for Scotland
2012 is 15c ambient temp for wc
2021 is 44c Flow temp
2022 is 32c Flow temp in mild weather
2091 set at No 1 for run signal at terminal B22
Trv’s all open and lock shields appropriately set .
House definitely warm sitting around 21-22c
Attached is pipe arrangement on Heating tank
Sorry for not explaining the graph. Each point represents one day. The horizontal axis is the degree-days figure for the day (degree-days is a slightly modified average outdoor temperature, easily downloaded from the internet for a weather station near you). The vertical axis is power (in Wh) consumed during the day (24hrs). As you can see the power varies a lot from day to day even for the same number of degree days (average OAT). This variation will swamp any change due to flow temperature changes until you have more data.
The plumbing of your buffer tank is correct however a 10C drop is a lot, and makes me wonder if there is mixing going on in it. Not much you can do really, buffer tanks are a pain which are rarely actually needed, but all to frequently installed particularly 8 years ago.
COP is energy supplied to house /energy supplied to heat pump, should be in the region of 3-4 most days. Its basically a measure of the ‘free energy’ you get with a heat pump which is what makes them so efficient. I dont know if the Samsung controller tells you the figures (manual is unclear), some do some don’t, if it does it would be under ‘information’ or ‘energy’. If it doesnt tell you the figures you cant calculate it unless you measure the power externally, so you may be out of luck here.
Your FSVs are reasonable, you may want to drop 2022 a bit if you are too warm when it gets milder, otherwise concentrate on getting 2021 to the point where the house is the right temperature, which you appear to be close to.
Thank you for your patience and explaining my questions. There is -so much I understand , but may be over thinking , it’s just that my first month with Octopus on Cosy seems to be high(yes it’s cosy, make no mistake ) ! It has revealed a cost from 1st Jan to 27th is £509 , which If devide cost by KWh comes to .24pence so moving on I hope I can improve this . Onwards and downwards, cost wise!
Total kWh 2210 all on Cosy bar 1 day.
That sounds quite high, but of course I dont know much about your house or how much DHW you use.
My total January electricity consumption (minus 1 day) is 1076kWh and that includes charging my car (albeit not a lot for this). I have a 200sq m 1930s house solid wall (no cavity!) house with a couple of modest modern extensions, 300mm loft insulation, partial internal wall insulation (so some walls are still solid brick). The measured heat loss @-2 is about 7kW. By no means a passivhaus! Mind you we don’t use much hot water. I keep more or less the whole house at 20C 24*7.
@JamesPa Yes just a bit !
Only hot DHW of significance is a small bath once a day. There are only two of us . Thanks
currently seeking info on Mod bus compatibility with Gen 5 . Fingers crossed!
Good luck on the modbus. I think you are in with a good chance, these units are based on aircon units used in commercial settings where modbus will likely be how the building management communicates.
Do you have the ability to measure the temperatures on the emitter side of the buffer tank, it would be useful to get a handle on these and in particular the difference between flow temp in from the HP and flow temp out to the emitters.
Yes I do, I have another set, I will place them shortly.
Incedently I notice from the sAMSUNG CONTROLLER that the oat side temp outside is 0.6c and two weather stations are saying -1.2c , so at least its not far out.
Attached earlier photos with HP and rads in circulation
Black probes being HP
Later without , when I came back to house all systems quiet
OK, the most interesting thing is the temperature drop across the buffer measured between flow from HP and flow to radiators. I dont think I can see that in the photo.
So , on the first two photos the Hp is delivering 43-5c and through to Rads its 37.4
2nd two photos its 36.6c to rads its 32 .8c remembering system is idle , yes it seems to be quite a difference on the first active pics.
I can take more when I see that the circulation is on ?
To the final question yes.
If the HP is delivering 43 and through to rads its 37, then you are losing about 12-20% in efficiency as a result of the buffer tank, which is going straight onto your electricity bill..
Repost when you have more info with the system running.
37.1C and 32.7 Emitters
All pumps , two on HP AND ONE CH pump are on setting three.
Attached is flow signal , not sure how you read that whether it’s the rh side of flow indicator or LH
@Alfapat Your aim (when the system is running and has settled down after a few minutes) should be to match the temperature feeding out of the LLH to the emitters with that of the flow temperature into the LLH as closely as you are able, likewise the temperature of the return flow entering the LLH from the emitters should be as close to that of the return flow out of the LLH to the heat pump.
Too fast a flow from the secondary pump return to the LLH is likely to cause unwanted disturbance to the water in the LLH and cause mixing – this would cause the returned (cooler water from the emitters) to recirculate within the LLH rather than meekly following the direction of flow back to the heat pump. It is a balancing trick with compromises and a need for a little dose of pragmatism – but it will pay dividends! Regards, Toodles.
OK, so 41.9 in from heat pump, 37.1 out to rads, drop of 5C = 10-15% efficiency penalty. Not sure how the return to the heat pump is higher than the return from the rads though, this can be the case if the flow rate from the heat pump is higher than the flow rate to the emitters but in that case the expectation would be that flow temp to emitters is close to flo from HP.
In summary no idea what is going on (one of the reasons to dislike buffers!).
@JamesPa Sounds to me as though the secondary pump may be running too fast and causing mixing in the LLH. You might care to try lowering the speed of that pump in stages and see if matters improve. Regards, Toodles.
Is it possible that , before I went up to the loft at that time , I noticed the HW tank symbol was on the controller. I did try to determine that by feeling where the valve lever was positioned in relation to each other , as I cant see them I wasnt sure . I can never remember which is closed or open . Perhaps one more visit is required.
Similar really, rads 5C lower than output of ASHP = 10-15% penalty. Difference of 2C on the return possibly a measurement error. I think @Toodles might be right secondary may be running too fast. He suggests reducing the speed in that pump see if things improve. You are really looking for feed to rads close in temperature to feed from heat pump.
I wouldn’t change it too much though, deltaT across rads already 5, I wouldn’t go above say 8. Do you have any control over the speed of the pump on the HP side?
I think there is control, there is 1, 11,and 111, written on them, but which pump do I slow if its possible CH pump or HP OUTPUT OR return . Toodles you talk about secondary pump is that HP OUT PUT or CH.
To clarify has the HP classed as primary and inline secondary .
Being dumb again !
@Alfapat Not at all dumb, much better to ask before jumping to conclusions. The primary pump will be the one in or attached directly to the output of the heat pump flow pipe. The pump that sits in line with the output from you LLH is the secondary pump and it is this one that you should adjust. Were it running too fast for your set up, it would possibly cause the water flow in the LLH to mix due to turbulence. What you are aiming to do is ensure that the flow through the LLH is as smooth and even as possible so that the stream of water is from the heat pump flow to the emitters and then from the return to the LLH and then to the heat pump return with as little mixing as possible. (Almost as though it was just a straight pipe from flow to emitters and another one from return from emitters to the heat pump.) So, in summary, try slowing the secondary pump down and observe if the difference in temperatures narrows; take it in small stages and pause to see the result. Regards, Toodles.
Thank you , I will give it a go , so I am looking at the pump returning the flow to the HP .
Just to be clear , I would have thought that turbulance is caused by to much flow into an area like the the tank (which I believe you shorten as LLH) and not getting out as quickly as it can , but by slowing the process down would make it worse.
On the other hand I suppose slowing down the output and not return would take the advantage of building heat from the HP would be lost and everything being held up in the tank and not circulate . Hmm
I will give it a go anyway
@Alfapat LLH (Low Loss Header) or you may have a similar device (a buffer tank) basically has the same ports and is in line with the flow from your heat pump. The heat pump primary pump sends the energy in the form of hot water around the circuit so that it flows to the emitters. The secondary pump is there to ensure that the energy is carried around the system once it leaves the LLH or buffer.
The water does not normally need to run a grand prix race but just circulates (relatively) gently to do its’ job. When the water is flowing through the LLH, it may just flow as required or if pushed too hard, will look for other ways to find freedom and escape! This will cause the turbulence (or as it is sometimes referred to as, distortion.
The flow from the heat pump ideally should go straight on to the emitters without being mixed with the return flow at all – but in this less than ideal circuit, some of the water will be diluted with the colder return flow. The balance you are aiming to create allows for the flow to get to the emitters but still have sufficient oomph to keep the return flow to the heat pump after passing through the LLH or buffer on the return journey. More speed or pressure will cause more mixing and lower overall efficiency. Regards, Toodles.
So I lowered the return flow and will monitor temperature . In the first 1mn I noticed the the flow return up .2c from 31.1 o .31.3!!
12 hours later and after slowing return pump down I get 36.9C and 28.6c HP and for CH 31.5C and 26.5c
@Alfapat Sorry, I am a little confused as to which is what here. Might you possibly express the temperatures as flow from HP, flow out of LLH, return to LLH and return from LLH to HP please? We can gain a clearer idea of which is what then I think. Regards, Toodles.
@Alfapat It looks to me as though there is either a lot of mixing of flow and return going on in the LLH or the sensors are all over the place. Have you ‘calibrated’ the sensors? By that I mean bundling all the sensors together and noted any differences in their readings. My four were different but a pair were a whole degree different to the other two – the two ‘pairs’ were noted and as the other differences were within 0.1 degree, I have just made mental allowance for the degree difference in readings. It doesn’t matter which is the nearer to correct temperature – we are only looking at the relative differences here. Regards, `toodles.
So stick them all to same heat source and see what they say?
@Alfapat Yes, bind all four probes in a bunch, perhaps with a rubber band around them all so they are in close contact to ensure they are all seeing the same conditions. If the probes are waterproof, they could then be suspended in a glass or bowl of water and then observe the differences once they have settled down. I had the additional complication that 3 of the 4 had less then fresh button cells – in fact 2 died during this testing phase so I had to delay until fresh supplies were purchased, (I note that Brendon Uys {@Brendon Uys} is discussing distortion as it pertains to LLH and Buffers in another thread,[‘4 Port Buffer Tank Connections’ that you may care to read too] in your ‘spare moments’ 😉) Regards, Toodles.
Yes Ill do that.
Let me pose the question , forgive me if Im wrong but if a pump is delivering into a tank and the outlet is at the same level as in position , then say it is at 20l a minute and the outlet is recieving this and is at a figure of 15litres then that would cause turbulence , but if the exit is at 22 /lmn then it comes to be less turbulence. It seems to me that I should swap flows vice versa , like output from HP lower than than feed to LLH then return higher to HP . Am I wrong?
@Alfapat The general consensus is that the secondary circuit should run at a slightly lower rate thean the primary. This is a matter that @Brendon Uys is far better to advise you on than I might. Regards, Toodles.
@Toodles Ok Thanks , meantime I will go to the loft and extract the probesand test them.
This is just a pickle!
@Toodles to my very basic understanding, and again @Brendon Uys is the best person to confirm this, with a buffer tank in the middle, the primary and secondary circuits should have matched up flow rates to avoid distortion in the buffer.
Buffer tanks can work if you are lucky, I am yet to make one work successfully 100% of the time and I do not gamble with other peoples money, it is unethical.
@Brendon Uys I recall a considerable amount of discussion (relating to LLH’s mainly rather than buffer tanks as such though) some while back and one of the threads is:
https://renewableheatinghub.co.uk/forums/renewable-heating-air-source-heap-pumps-ashps/low-loss-header-is-losing-1-5c-from-flow-temperature/paged/2
There were other threads along similar lines and this is where I felt a ‘consensus’ was evident. If I am wrong, my apologies to all for leading you astray!
In my own case, I have a Daikin heat pump and I THINK that the primary pump adjusts automatically to maintain a set Delta T. Having an LLH in circuit, I have attempted to adjust the speed of the Wilo Pico pump (secondary pump) to minimise temperature difference between flow to LLH and flow out of LLH to the Wilo pump. Similarly, I have attempted a similar arrangement for the return to the LLH in and the out returning to the heat pump. I find that at times, the balance is within 0.3 degrees C but at other times, this may be as much as a degree of difference!! I think both pumps have a degree of dynamic control and that this is the reason for the apparent variation. Regards, Toodles.
@Toodles I am totally against any form of system separation in a domestic environment and most commercial environments. I have installations with over 400 Kw in capacity and 9 heat pumps, all connected to a single flow and return, the separation/buffer arguments just do not stack up. Any form of separation makes things complicated and can cause significant performance losses. In my opinion, simplicity is always best.
If only others took the same approach we would see far, far fewer posts on this forum (and others) complaining about poor performance, and indeed far fewer posts complaining about other problems, the origin of which it is difficult to diagnose due to the presence of system separation.
We would probably see an increase in posts about low flow errors which otherwise are masked by the buffer. But the underlying causes in many cases need to be fixed, so in many ways that’s a good thing.
There is probably a business to be made from removing buffer tanks!
Sending for more probes , one is dodgy! Its sending me a higher temp on return to LLH than the feed to rads.
@Alfapat I think there is a line from the Engineer on Star Trek that to cover that! 😦 Toodles.
Maybe mine as well !As in Buffer remove.
New sensor in place and so to correct the flow indications between from HP to buffer /Tank there is a difference of 2.6 c. That is to say HP to tank and tank to emitters . Emitters return is 33.7 and Buffer/Tank to HP 36c
Too much or ?
Not great, 5C = 10-15% efficiency loss (much more according to @Brendon Uys). TBH I cant understand whats going on in the buffer, its like there is no stratification at all and you have essentially a vessel sitting there at ~36C, with both outputs therefore at this temperature. Im beginning to wonder if its correctly plumbed but cant remember if we have already discussed this. Is it worth posting (or reminding us of) some pictures indicating flow/return to/from HP/rads. Also please can you remind me how the pumps either side are controlled and in particular can you get to a situation where one is on and the other off for a reasonable period of time (thus, inevitably, ‘stirring’ the tank)?
Here is the layout as much as I can.
Above left Ch pipes
Above zone valves.
Above right flow from HP.Above showing Ch pump on white pipes. If there is anything more specific I will take more.
no diagram!
Ha should I have , sadly not, Only the one I have drawn.
What would you like to see , and I can give you some information
Sorry I didn’t spot the pictures above.
Buffer tank looks very squat (which would explain lots). How high is it and what is the vertical distance between the ports.
Also what capacity is your heat pump and what diameter the pipes to the buffer tank.
@JamesPa Will check James
Tank Capacity is 200l Pipes are 28mm Copper , not sure if you want the size of CH pipes I can check but definitely smaller of course either 24 or 21mm . Tank vertical distance is 65cm for HP pipes and 67 cm roughly for CH.
Thanks
Sadly with a 60cm high tank I think the chances of maintaining stratification (IE a temperature gradient from top to bottom, which is what you want) is negligible, which accounts for the almost complete mixing you observe with the temp sensors. Whilst the tank remains you are stuck with the 5C drop across it. I suspect that adjusting flow rates will make a marginal difference at best to this.
@JamesPa Thanks , looks like I should remove it then . Question I have , if the HP pipes are 28mm and they are directly connected to the smaller CH pipes , does that not cause a problem , except if you had control of the pump speeds ?
If you remove the buffer you will also need to remove the secondary pump.
Re your concern, it depends. Some further detail will help determine.
OK James,
Ill see if I can answer your questions. the heat pump is 16kh , would the loss be on the commissioning ? The original commission had another HP alongside 9kw., but this was removed when another company accepted to take over the whole thing when the original installers were liquidated. So I only have there proposal.
System is placed in a large loft and CH pipes travel down to one floor , they start plastic but down to copper .
DT?
OK.
16kW is not compatible with 22mm pipework, and barely compatible with 28mm at a typical design DT across the radiators of 5C
That said 16kW may well be an overestimate (or even a gross overestimate) of your requirement. Its quite plausible that this to be the case – over sizing is common even now and back when yours was installed probably endemic. Two relatively recent ‘professional’ 3hr surveys for my house came out at 16kW, its actually 7kW or possibly a bit less.
Essentially, to omit the buffer tank and second pump, you need to satisfy the conditions that
with your existing pipework (or pipework you can reasonably replace)
I think the summary is, this needs a study that is more in depth than that which I can realistically do on this forum. Essentially it needs a partial re-design of your system but wherever possible using the same components. A randomly picked heat pump installer is unlikely to be capable of that (because they would be too reliant on ‘rules’), maybe a ‘heat geek’ qualified one might. Its also possible that a commercial heating designer might because they should be used to actually doing calculations (some aren’t, as I discovered to my cost in a work-related project).. Alternatively you need someone who can do the analysis ‘from first principles’. Its not fundamentally difficult, but it needs a visit and some thinking which is not entirely ‘in the box’.
Unless you are prepared either to do this yourself or to pay someone to do this, you are probably stuck with the buffer.
There is one more option, which is that you plumb switchable bypasses (ideally such that the buffer becomes a 2 port volumiser) and ‘suck it and see’. If you fancied a couple of days of plumbing in the summer season that would be good prep for next year!
Otherwise its a case of living with a 10-15% penalty doe to the buffer, which may or may not be necessary unless you are prepared to replumb your central heating system.
Im sorry to be the bringer of bad news, maybe others have an alternative suggestion.
This is depressing , I have spent now a couple of months now trying to sort out a solution or some sort of headway on this forum to readdress the biggest investment (apart from buying the house) and sorting this Air source running cost. I feel absolutely useless as what to do now apart from trying something like Homely. I have now spent £25000 on a heating system that simply is pants! To put very lightly.
If you were willing to supply more info I could probably give you a good idea if just scrapping the buffer will work.
Some of this is in the certificate you posted above. and where it is I have pasted in the values for you to check I understand correctly. If you could advise the rest, or at least most of it, I can give you some more pointers
With 13kW demand (If thats real – which it may not be) I really cant see how a single 22mm primary feed to/from rads will work at the typical 5C deltaT that heat pumps operate at. So if that is what the demand is, and if there is really only a single 22mm feed, then the buffer is a necessary element of the design to allow the secondary to operate at a larger delta T flow -> return to the primary. But if this is the case then you should measure this on your thermometers which so far I haven’t seen.
More detail may help point you in the right direction or, if you are lucky, confirm that 13.5kW is a gross overestimate and you can in fact get away with 22mm.
Insulation increased before installation to Loft at 300mm. Insulation below floor 2022 170mm
Installation 2015
one pump 16kw , the other sold after recommisioning. Sub section 4.1 (v1) on THE MCS Certificate shown above is Heat Loss 13.48kw
Distance from Buffer to eaves in loft approx. 3m then a drop to outside ground 2.70m.
Timber buit House on Dwarf wall . Insulaation on outer interrnal room wallsapprox 1inch polystyrene silver backed stuff!
Total area 199sq m
28mm hp piping to split joint between DHW and Buffer joints 30mm
Oil burner before with Wood stove on zone valves (amature set up) Unknown running cost as only moved in 2015
Pics to follow easier on phone
transfer.System temps shown passive no circ required.
and layout to eaves. I am told that TRV ‘s would be installed along with buffer and controlled by Danfoss thermostat shown earlier in threads. As the TRV’s are fully open now and the system changed using lock shields
With your insulation upgrades its quite possible that you may be able to dispense with the buffer without changing the pipework. Can you clarify the following (which I cant work out from your photos).
The picture attached shows the Ch pipes top and bottom, (never mind the expansion vessel that is scrap ) the pies come out and in about 15” in copper 22mm then take a right turn in 22mm plastic for 9 feet another turn for a couple of feet into copper 28mm and drop down from the loft and under the floor. I think all rads are on 15mm . I’ll get temp at evening and record flow and return and Oat At present late afternoon 11c , flow temps 27c and 39 out CH Controller says house 21.4 , 9.6 outside temp, circ 39.7, DHW 48.5.