@kev-m 

Your observation re luke warm radiators brings me to a question.   All of my downstairs radiators are fed by drop down pipe work in the corners of the room.  Aesthetically I am proposing to box them in or cover with Talon pipe 'boxing'.  I now wonder if I should suffer the drudge of painting in order to obtain every last bit of heat from unboxed pipework.  Marginal gains and all that.

@morgan box them in and lag them

@derek-m 

that should be interesting!

The MCS spreasdsheet says 6.47kW at -3.2.   It's probably more than this.  About 10% of the house (utility, cupboards) doesn't have heating and wasn't included in gthe measurements.  Also, they didn't include the fact I have a chimney/fire - that makes quite a difference because of the allowance added for air flow.  I'd say add 15% overall. 

All rads are Steldad K2.  House is a bungalow with lots of small rooms.  

No, MMSP still not giving sensible data. 

@kev-m 

Hi Kev,

Thanks for the info, I will add it to the spreadsheet.

I assume that you placed the sensor under the insulation? What temperature readings are you getting?

Is it copper pipework or plastic pipework under the insulation? If it is plastic then I would suggest that you tape the sensor to the brass pipework of the flowmeter, ensure that the tip area of the sensor is in good thermal contact with the pipework and tape in position. Then cover the metal section of the sensor with insulation and tape in position.

I've stuck it under the insulation on plastic pipes.  I'll have a look later.  

Posted by: @batalto

@kev-m how does that look if you plot temperature on the X axis? Is the power use fairly linear?

Like this?  Apart from the glitch in the middle, yes.  Remember the values are for a 24 hr period starting midnight.  The night of the 11/12th was quite cold but daytime on 12th was warm so average was relatively high. 

@kev-m fairly straight line relationship for it. If you plot a trend line you should be able to estimate your usage based on temperature moving forward.

Posted by: @kev-m

Posted by: @batalto

@kev-m how does that look if you plot temperature on the X axis? Is the power use fairly linear?

Like this?  Apart from the glitch in the middle, yes.  Remember the values are for a 24 hr period starting midnight.  The night of the 11/12th was quite cold but daytime on 12th was warm so average was relatively high. 

Over a larger ambient air temperature range I think that you will find that it is more of an exponential curve rather than a straight line. As the ambient air temperature falls, the heat demand increases, but also the ASHP is less efficient, so at lower ambient air temperatures the energy consumption increases more rapidly. Basically, the system needs to produce more heat energy, but does it less efficiently. I know it is not the type of news that you wanted to hear. 😒 

@derek-m should probably make that graph based on heat provided and not power used. That would deal with problem

Posted by: @batalto

@derek-m should probably make that graph based on heat provided and not power used. That would deal with problem

Would be useful to see both I think.  

@kev-m 

Hi Kev,

Whilst going through the manufacturers data I noticed various aspects, which you, and other 14kW Ecodan owners may find of interest.

The first thing that struck me was that the unit could produce its full rated 14kW output, over the full range up to a water flow temperature of 55C at ambient air temperatures down to -10C. So a correctly sized unit, should be able to satisfy the heat demand over the vast majority of weather conditions that may be experienced in the UK.

It would appear that the unit has also been optimised for maximum efficiency in the Mid (11.2kW) to Nominal (14kW) operating range, which of course is more desirable when the unit is working harder during colder weather.

The modelling tool shows that a unit (based on your operating parameters) would go from cycling to continuous operation at an ambient air temperature range of 7C to 8C, which is what you found in practice. It would therefore be useful to see how closely the other parameters compare with the modelling tool.

The modelling tool should also help define the parameters for setting the weather compensation curve. In your case it is suggesting a water flow temperature of 43.7C at an ambient air temperature of -7C, and a water flow temperature of 22.8C at all ambient air temperature of 20C, which should produce an indoor air temperature of 21C.

If you should require a different indoor air temperature then get the settings from the relevant table in the Detailed Data sheet.

OK folks, here are the first 3 weeks or so in November.  I started the new weather compensation regime on the 8th.  The high number is because I overshot and the house was very warm at 23-24C.  Over the next few days I tweaked the curve and got it working quite well diring the relatively mild weather we had.  When it got cold, the warmer radiators heated the house too much and on 21/11, the house was again nudging 24C. I tweaked the curve again and on 22/11, a similar day in terms of outside temps (actually a bit colder), energy used was a lot less and the house was a nice 21-22C.  This is 21/11:

You can see the ASHP isn't cycling; even during the day it's running constantly.  As I said, the house was very warm.  This is 22/11:

House was 21.5C or so.  It's steady going in the evening and night but cycling again during the day. You can see the cycling get faster as the target flow temp decreases and it reaches it ever quicker.  It settles down a bit when the outside temp is a bit more constant. Very interesting. I wonder if there is a way of decreasing the sensitivity? 

I need to tweak the curve a little more as the house is cooler when it's warmer outside than it is when it's colder.  It's not far off though.  

In terms of cost/energy use, 33kWh per day at 4C outside and 21.5C inside is liveable with.  I could reduce this if I knocked the house temp down a degree.  Will be interesting when it gets properly cold; more winter weather on the way apparently. 😮 

Kev