HeatpumpMonitor.org v2 discussion

Hardware Heatpump
41

Hello,

I am the owner of the ‘Brighton & Hove’ ecodan system! Over the past year or so I have made some really significant changes to the original installation to get it working correctly.

The actual COP for the CH is around 5.5. We still have a little air in the system which is causing a drop in the measured efficiency in the DHW cycle as air appears when the flow temp hits 45C.

What have we done to getting the CH working so well?

i) I have upgraded all the pipework from what was originally installed. The primary pipework runs are now in mainly 28mm with a little remaining in 22mm. We have a large Emmetti manifold that distributes water directly to each radiator uisng 16mm MLCP pipe. There are no TRVs.

ii) there is a single large Grundfos MAGNA pump. Our flow rate is typically 1440l/h. That is 24 l/min. There is a section of 22mm pipe remaining and I expect we will see it increase further once that is shorted and replaced in 28mm.

iii) we only ever turn off two radiators in the loft extension (office) at the weekend. We have no other radiator controls. Each radiator gets a measured 4l/min from the manifold.

iv) the heating controls are a combination of loxone/knx. In reality is is very simple config. We have a precise digital thermometer in each room. When any room ‘calls for heat’ the system comes on and heats every room in the house. The system has a minium run time of 1 hour. It currently runs at a manually set 32C.

v) as the temperature drops as it cools the system will remain running all day. At that point we will increase the flow temp. I am hopeful we will not need to push it beyond 40C. This is estimated and hopeful, rather than being scientifically calcuated.

vi) radiators are typically long K3 ones. It seems to be that there is only one room (the kitchen) that is causing the heat to come one. I think that this due to poor insulation/drafts and an undersized aluminium radiator that was originally used with a gas boiler. I am considering replacing the rad with a floor standing 1600x300 K3.

I have found the width of a radiator to be the most critical factor, and the height is not needed. A thermal camera image of our 900x900 radiator shows the bottom half doing very little.

I convinced that TRVs are bad for a ASHP installation. They throttle the flow by partially closing when a room is at temperature. ASHP is much more suitable to a ON/OFF arrangement. And then using the precence of an ON to start up the system; then we give every room heat while the system is running.

Currently the system comes on when a single room wants heat. I could change it so that it requires two rooms demand heat before the system starts up.

Let me know if you need more, or pictures.

Incidentally we use every rooom in the house each day (other than the office being unused at weekends), and the house is always occupied.

George

9 Likes
42

Hi George, thanks for identifying yourself and giving a really good description of your system.
You’ve done a great job of getting the unit to run really well.

I was interested to hear what the custom controls were to see if this was why it was running so well but it sounds like it is mostly ‘just’ a really well set up system and the main difference between our 2 systems is the inadequate output of my radiators and pipe diameters.

I’ve been finding that running on pure weather comp I’ve been getting much worse COP than Auto Adapt which shouldn’t make sense however without enough output the system cycles loads as the flow creeps up. Whereas AA compensates for this by allowing the flow to drop more between cycles meaning heating the water in the system can match the HP output and then it takes the flow temp much higher to maintain a larger DT between the radiators and room so again providing a better sink for the output. All while checking the effect on room temperature isn’t too high.
Basically it is good at compensating a bit for the inadequacies of the system. I was thinking when we get the big hall radiator added I might get as much primary pipework increased to 28mm and I think you have confirmed this would be worthwhile.

43

One metric i’m becoming more interesting in is total system volume.

Selfishly i’m trying to work out why my 5kW Arotherm cycles way more than others on the list.

I’m around 90L total system volume including pipework (10 rads, no buffer/volumiser).

So I think it would be useful to collect that data? :grinning:

2 Likes
44

That sounds like it could be interesting. I guess it might depend how much of that volume is ‘in play’.
Our bedroom radiators have TRVs set to 19.5C to stop them getting too warm, this time of year they are mostly closed as they are better insulated than the downstairs rooms where TRVs are all open.

45

Yes, ‘in play’ volume is something that i’ve just read on another thread. Makes me wonder if my rads are balanced. As Marko says, perhaps i’ve got water just rushing past rads and back to the heat pump.

46

How were you able to determine that the original pipework wasn’t up to the job?

47

A lot of it was narrow 15mm in copper and plastic. With excessively long runs. I could not get it balanced correctly. It is all now in 22mm including the elbows into the radiators. We have no TRVs or balancing valves: just a neat 22mm chrome elbow.

Have a good read about pipework pressure drops. A narrow pipe really impacts flow rates and makes it harder to balance. John Cantor has a brilliant simulator.

Balancing is so much more critical with ASHP as you are trying to run the system with every radiator is fully open.

This website also describes maximum flow rates:

Some of our primary pipework is still in 22mm. The flow rate through this section is 1.5m/s, close to the 2m/s limit mentioned by John Heafield.

I installed a manifold because I reached the end of my teather. They have flow meters in them so you can adjust and see the flow rates. It works brilliantly. Because each circuit is individually controlled you can easily add another radiator with no downtime.

A high flow rate allows you to drop the delta T (difference between flow and return) and the flow temp by a degree or so. This improved efficiency. Though, we do now consume more power on pumping the water than before. BTW our pump is a Grundfox Magna3 32/80 180.

2 Likes
48

Thoughts…

  • Something is causing it to modulate and turn on and off. It should be running continually.

  • Your flow temp is very high. Right now our flow temp is 32C and our house is warm.

  • Try running with manual flow temp and drop that temperature until you get too cold. The weather compensation is a gimmick as it does not consider demand or solar gain.

  • what do you have connect to the FTC to turn the system on or off? There are a couple of contacts that you attach a switch to. Our original installer connected to a digital thermostat that turn on off the ASHP. It causes all manner of problems and lasted one day.

Our contacts are permanently connected together and we issue a modbus command via the Procon A1M modbus unit to turn on/off.

image
image1119×695 146 KB

Have a look when compared to our trace. Ignore the high peaks they are from the DHW and the consumption is for CH only. The only reason it went off between 1630 and 1900 is it shuts down when our Octopus Agile gets more expensive.

image
image967×484 81.5 KB

1 Like
49

The ASHP can only run down to a certain power, then shuts down. The compressor motor has an inverter, but this can only go so far. Then it will turn off.

For example, our ecodan can only go down to 550W. Any lower it will shut down. I have increased the flow temp by a degree which stopped it modulating.

How much will an increased volume do? Somewhat. There is certainly more thermal mass so the cycling will be less frequent. Add more/bigger radiators and fatter pipes. :slight_smile:

1 Like
50

“liar” :sweat_smile:

image
image1117×733 69.5 KB

So it looks like every time is at the minimum and tries to go lower, it finds that it’s too low, and steps up a little bit, if that it’s too much it stops.

1 Like
51

Alright… I stand corrected. It goes down to about 500W.

It can only only go down so far, then shuts down.

52

Probably would go lower if warmer outside and flow bellow 30C.

This is the 6kW ecodan(same unit with 8.5kW) that @johncantor had with UFH.

image
image1231×788 126 KB

53

It would be good to see a photo of this setup.
It isn’t really an option for us apart from the Ground Floor as all the pipework there is easily accessible from the cellar. Mostly just interested to see what it looks like.
Although we should be able to fairly easy to up all the pipework to the first few Tees to 28mm. Even have some that could be reused from dead legs to where our boiler used to be.

54

Have a couple of pictures of the smaller manifold. The other one is downstairs and harder to photograph.

It currently supplies two floor standing K3 radiators (1200x300) in our loft area - one radiator under each desk (for warm feet!). The other three ways will be used when I redo a shower room and put in underfloor heating/radiators.

On the manifold you can see a couple of shut-off lever valves, air vent, pressure gauge and drail off. The box with the green cable is a KNX controller that operates the 240v actuators. The small vials with teh red balls show that the heating is operating and each radiator is currently getting 2l/min.

Ignore the network socket hanging off the wall. This is unused and unrelated.

image
image1513×2017 693 KB

image
image1513×2017 604 KB

3 Likes
55

Excellent, I’ve never come across rads coming from manifolds before but seems like a good way to make sure heat is getting everywhere and just have to worry about the large diameter pipes to the manifolds.

1 Like
56

That’s a nice idea and I like your concept image. One question is how or if we need to show both COP calculation approaches?

  • Method 1

    • System 1 COP 3
    • System 2 COP 4
    • Average = 3.5

  • Method 2

    • System 1, 1000 kWh electric, 3000 kWh heat, COP 3
    • System 2, 100 kWh electric, 400 kWh heat, COP 4
    • Average based on kWh = 3.1

Not sure what the best approach would be to do this. Feels like it could be quite complex to implement?
Perhaps a way to notify of a forum thread that mentions a system?

57

Do we need to show both approaches? The numbers don’t seem to be very different over large enough dataset, showing both could just add confusion.

The next question is: which is more useful to show? Is it best to have the mean of the COPs, or have an average that weighted towards the bigger systems? I’m not sure, I but I feel that maybe method 1 is simplest and easiest to understand.

1 Like
58

Maybe for each system there could be the option to identify your user in the forum then you could be DMed and forum posts could be searched for.
I was thinking it would be good for anyone who wants to to create a new thread for their system outlining anything that might not be obvious from the dashboard information and maybe explaining a bit of the history. Also messages could be posted to the thread when things change like settings, when experimenting with things, observations in the data or when anything has been physically changed in the system.
And any links to forum posts where the system was discussed.
Maybe the dashboard could link to that thread.

5 Likes
59

Now that more people are getting on the separate DHW / heating train.

Have we considering adding separate DHW/heating COP columns on the opening chart where they are available?

It might make be feel better about my total SCOP knowing my heating is doing alright and it’s only DHW dragging me down!! :rofl:

2 Likes
60

Working out Heating vs. DHW from kWh feeds isn’t possible with a boolean flag, so we’ll have to compute it from the raw power readings instead. Quite a bit more computation needed for that.

1 Like