Improve efficiency: Was it worth it?

Hardware Heatpump
1

I’ve spammed this category with comments about each of the tweaks I’ve done to the heat pump system. I used the prefix “Improve efficiency” like this topic.

Of course lots of failed experiments have gone by the wayside.

I class the default behaviour as “original”. In November I decided take computer control and created the behaviour I called “improved”. I tweaked that in January to make it more “gentle”. Each of those experimental periods is represented by a solid line on this summary graph.

image
image1897×1781 248 KB

The gold dashed line “dev_nominal” is how the vendor expects their system to perform with flow set to 40Celsius. The other dashed lines are some of their measurements with other power levels.

Since I had it installed in October space heating has been my primary energy consumer so I’ve focussed mostly on improving that. My hot water is a bit more efficient after my changes, but not markedly.

image
image1900×808 108 KB

As my heat pump is rated for something like -2Celsius it’s over-powered for most of the work it’s doing. Luckily it modulates pretty well so I’ve been able to take advantage of that. By implementing the on / off strategy I’ve been able to drastically reduce the cycling.

Yes, you’re reading that right, it was cycling 80 times a day using the original setup.

image
image1891×815 76.3 KB

You might be wondering if I’ve saved any money by doing this. I should obviously mention that’s not my motivation. Even if I paid myself minimum wage for the time spent on this I wouldn’t even be close to break-even.

What I’m trying to do is to get the system running more efficiently with the hope of improving how heat pumps are deployed in the field. Clearly my experience is limited to just one heat pump in one house for five months.

I’m sharing what I’ve learned to help improve things.

Perhaps one way of doing this will be to extend Emoncms to provide an open platform for people to experiment with.

Anyway, here’s the impact on costs. I don’t have an opinion on how useful those savings are.

image
image1892×815 87.3 KB

1 Like
2

I think you shoulda also added [Ecodan advertisement], too. Elaborate guerilla marketing.

3

Sad face.

It’s certainly not my intent. Apologies if it came across that way.

It just happens to be the system I’ve got. I’ll go remove the system name from the posts.

[edit] I’ve now removed reference to specific system name in all those posts.

4

I think it was a joke :grin:

1 Like
5

I would be interested to know how you got your cop improvements.

@johncantor This topic is my summary.

You can see each type of tweak in it’s own topic. I’d love to hear your thoughts on those topics.

6

when I got my gshp heat pump it was cycling like that as well when i first installed it it was to sensitive to temp changes … but for me all I did was shift the temperature differential by 1 degree and added a 10 minute delay before it can cycling again… at your mild temps my cycling is roughly 1 1/2 hours between cycles on 10 minute run times . and that heating my home and making my DHW as it all the same tank. the reason for the 10 minute delay it improves the cycling pattern as some times a heat pump can get stuck in a aggressive cycling pattern on for 5 off for 5 at certain temps … the delay just insures it can not get stuck in such a pattern

7

@stephen

Good to know you’ve got your GSHP working well too.

The initial problem with my setup was just as you describe, an aggressive cycling of 5 minutes on then 5 minutes off. Here’s 2019-10-19:

image
image1367×1477 290 KB

I don’t see how to use the vendor’s controls to put hysteresis into the heat pump. I did widen the band in the room thermostat to 5 Celsius but it is also missing a “delay” setting to slow it down.

My control algorithm has a weather-dependent delay algorithm which seems to be working OK.

I’m holding down the flow temp for efficiency, so my “on” time is longer than your ten minutes, but I am seeing 1 1/2 hours between cycles like you. Just as an example, today some of my cycles were on for 50 mins and some were 15 mins. The long run was the first one of the day and the outside temps were obviously colder early in the morning.

Here’s how things looked this morning (2020-02-21). I’m using the return temperature to tell me when get going again - it’s a proxy for the radiator temps.

image
image1372×1469 234 KB

8

well I guess for my gshp it is a water to water ( I am not sure if yours is air to water or water to water heat pump…) I did not change the room temp differential… I changed the the water tank temp differential when I first got it was 2c differential and just the simple short inrush of cold water into the the tank was enough to trigger another run cycle … changing it to 3c was enough of a change to stop that… do you have full access to your heat pump controls or only basic user setting … my heat pump is fully open I can change any and all run parameters… usually they will not give you full access. and here in north America they rarely give even base user access… as they want “servicemen”. just to change the simplest of settings ie temp differential or even restart your heatpump- which can be triggered by a power outage- and i see people sitting months waiting for someone to reset the heat pump and running on backup booster elements for months – maybe search the web you can figure out the service entry button combo and password then you have all the option I mentioned and more

9

Interesting @stephen - I am tempted to tinker with the system settings. Our installer left them open for me because he gauged that I knew what I was doing. I’ve actually left them alone because I don’t want anyone to question my warranty - I’m on a seven year contract with the installer and the UK government for this system and it would be very expensive for me to fall off that contract. They normally leave the systems with the base user settings open in the UK.

I don’t have a buffer tank like you. My air source heat pump is connected directly to the radiators. As a result, it’s load is very much correlated with the room air temp.

I like the sound of what you’ve done with your system to get it behaving nicely. It’s always nice when a small tweak makes things work properly.

10

Great to see the performance improvements @MyForest really quite substantial!

It makes a lot of sense to me that turning your heatpump off and on to create longer cycle lengths is providing better performance, given that the heat pump compressor cant reduce it’s speed enough to run continuously without the excessive cycling you saw.

I have been thinking to do similar for night time set back temperatures. My system is stable at 30C flow temperature but that keeps the house often overly warm at night. If I drop the flow to 28C it starts frequent cycling as the radiators are not able to dissipate enough heat to match the minimum input power of the heatpump, so I could do with this approach at those times. I generally turn it off at night when its a bit warmer as a result rather than cause a lot of cycling.

I undersized my heatpump, partly to better match the minimum compressor speed to the typical heat load of the house and radiator design and its given me COP’s that Im very happy with generally. Average is 3.9. The other reason for initial undersizing is that I plan to do efficiency work on the house.

I do a bit of your ‘holding down the flow temp’ for DHW heatup. The standard control scheme just runs the heat pump flat out to heat the tank and can be a bit noisy especially when still and cold outside. To address this I send a flow temperature ramp control scheme, starting from the tank starting temperature through to ~46C (always heating at least the bottom half of the tank up from cold). I dont react to the heatpump flow temperature or heat output during the heat up period as I’m weary of causing issues for the defrost cycle - and the interaction between the heatpumps own control algorithm and mine just becomes unstable if I try and react too quickly… giving it a stable pre-set temperature ramp seems to avoid that.

It would be interesting to hear more about the heat loss factor of your house W/K and kWh/m2.year and heat output rating of your radiators, what’s your total radiator surface area and heat output at their rated values? I see in your other post they are Stelrad K2’s. Is your heatpump a 14kW ecodan?

1 Like
11

Hi all,
I’m quite surprised to read that your heat pump was like cycling 80 times a day, that’s a LOT.
I’m heating my house (living surface of 250m2) too with a heat pump. When it’s very cold, around 0°C or less outside, my HP is running about 20mn every 1.5 hours. It’s heating a buffer tank of 400 liters till 35°C. This water circulates in the floor heating… The COP is even higher than the one given by the manufacturer, I’m around 5.5!!
Fred

1 Like
12

@FredM67 Yep, I was surprised too, but when it was in the “shoulder months” the over-spec of my heat pump becomes a problem with the default control algorithm.

I also wanted a buffer tank but was talked out of it. I accepted that on complexity grounds. I don’t want to create a system that only two engineers who are on-call can fix and then find out they have 10 other customers to fix first.

I have heard a number of stories of heating engineers turning up on-site and finding an unusual system and not knowing how to fix it or set it up properly. I was hoping that after more than 10 years of heat pumps being installed this would be a thing of the past but it’s still happening now.

However, given what I’m hearing about your system and what I’m seeing with mine I could consider a buffer tank. I have space.

You can see in this experiment that I don’t actually need to circulate all the time. My radiators are serving as a buffer tank because they are over-sized. Admittedly I get more fluctuation in my room heating than you would. Being able to turn all the pumps off is good, it makes our house quieter (yes, I know pumps shouldn’t disturb us, but it’s stupidly quiet here)

13

@FredM67 You can see my ASHP is cycling about the same as yours now

image
image1371×300 66.9 KB

Sadly my CoP is only 3.64 though :expressionless:

14

sorry – i get a bit of chuckle when I read that. I would consider <-30c very cold and <-40c extremely cold…

not to distract from the topic. But it makes me wonder on your general housing standards over there. as at that temps and other temperatures mentioned I would not even have a heating demand. as I get enough from other heat gains… for me I would be trying to improve the your shell you live in then trying to squeeze more out of the heat pump… as so far all the temps mention you should be at or near zero heating requirements if following similar to Canadian R2000 high efficiency house standards.

15

Well that with the buffer tank isn’t a tweak from me or any other one. It’s highly recommended by the manufacturer, exactly to avoid that the HP runs to often. The floor heating is making the system very stable… If I want to reduce the temperature in a room, it take like 2-3 days!!
The pump for circulation is so quite, I can only know it’s running by putting my hand on it to sense the very subtle vibration.
By mild weather like today around 10°C the HP is running every 4-5 hours, or even less.

Screenshot_20200223-171842
Screenshot_20200223-1718422560×1440 119 KB
The high quick peaks are when the HP is running.

16

Yep @stephen you’re quite right.

In the UK it’s bad. Most of the old housing stock has poor insulation.

Our house is 200 years old and only one-third of it has wall insulation. I work with a lot of Canadians and they tell me it’s common for houses to only last 30 to 50 years there so when you re-build you bring them to code at the time which means the whole stock is getting better all the time.

We jumped to a heat pump to allow us to get rid of our oil boiler because Greta told us to. The UK government is basically funding the switch to heat pumps at the moment.

We will have to get external wall insulation. The pay-back is about 30 years. Our house is attached to another one and our neighbour isn’t interested in doing it. As a result we’d be doing it to our half of the house which will look weird. It’s not always easy to do the right thing.

17

@FredM67 That looks great.

I like how your system has been designed and how it’s working.

Underfloor heating would be great, but in our 200 year old house it’s a big disruption and it was vetoed.

Most of the systems in the UK don’t have the buffer tank. The installers don’t seem keen on those models. It’s the same with plumbing in the existing hot water system behind my log fire…no-one wants to do it.

I’m willing to live with the limitations as long as things are working OK. I was seeing CoPs of 2.5 at 5 Celsius so was motivated to at least see if I could improve that.

BTW our pump is less than 30dB at one meter - I’m just fussy.

18

if a 200 year old house - then it a mostly a high mass house… and your biggest issue would be draft… for me I would insulate from the interior and put in a vapour barrier if you can not do the outside wall… and leave the exterior alone to breath … with modern insulation that might mean loosing about 2 in around the perimeter on the inside to get your self up r20 wall as currently from your estimated heatloss it seams you are around R10 wall. and then blow in R50 in to attic space. and try to insulate to 20R in the floor space… Canadian house today have a lifespan of about 25 years due to the use of engineered joist and the over use of OSB… after that major overhaul will be required at some point soon after. if the house is not well maintained … if you want a to see where you can get the best bang for you buck - download and install remrate it a 14 day trial but that enough time to figure out what you could do to your house to improve it

19

Thanks @stephen I’ll look into remrate.

We’re relatively draft free for such an old house because we blocked up a lot of leaks ten years ago to stop the mice getting in. We had a pressure test done back then and the chap doing it was aghast that a house like this was so leak free. This thermograph shows a lot of heat leaking out of the older part of the house through the walls to the left of the picture (although it did help us find a few more leaks under the windows).

IR_1671
IR_1671800×600 54.9 KB

I’d like to keep the walls inside the thermal envelope to keep the thermal mass bigger. This is partly due to my concern about damp between a cold outside wall and a layer of internal insulation.

The Canadian use of timber makes some sense, you have quite a lot more of it than we do.

20

that’s why you have a vapour barrier on the inside and you allow to breath outwards… you have more of an issue from the warm damp air migrating through the cool wall. if you prevent it from migrating then there is no issue… currently my outside temps are quite warm ( about 0c ) tonight I will try to take a thermal image

1 Like