@sunandair I agree, I remembered this yesterday then forgot again, thanks for posting

Posted by: @sunandair

↑

Posted by: @gary

↑

@jaket yes leave at 5 19 litres per min is fine mine runs at a similar rate

apologies. I have to make a comment on flow rate. Jake’s HP is a 6kw it is not an 8.5 model and it’s flow rate recommendation is around 11 LPM. I also note that the first post from Jake showed the HP trying to maintain a DT5c however on the recent hourly temperature chart it was down to 3 LPM. In order to accommodate all the modulation settings that the HP can provide I have found it’s important to keep Farley close to the recommended flow rate for your model of HP. Mitsubishi have changed the data book recently which now shows the recommended flow rate for each model. This is new because earlier copies of the data book only presented a range of flow rates. See below. 

-- Attachment is not available --

I hope this helps rule out one possible cause. It clearly is cycling which suggests it’s not able to dump the heat fast enough. Do you know what the emitter value of the whole system. One side issue - is your flushing valve fully closed?

 

 

Thanks that's super helpful - I will adjust the flow rate to target 11L/min. 

How do I check the emitter value & that the flushing value is fully closed (what does this look like / where is it positioned?) 

🤣

Posted by: @gary

↑

@jaket as the unit is cycling so much you wont get a stable dT from the brief periods its stable it looks like it would be 35C flow 32 return so a dT of 3C that is exactly what I would expect at those flow temps, you won't see a dT of 5 until you get into the 40's.

The heat pump is trying to supply a flow temp of around 35C, the rads and UFH get rid of 3C of that heat leaving the return temp to the HP of 32C.

Then heat pump transfers in more heat to account for the loss but its working at its lowest possible power. 

Over a period of 15 minutes this causes the flow temp to increase beyond its set point so the heat pump has to shut down.  Once the flow cools its starts back up again and the cycle repeats.

The only way to counteract the cycling is to increase the flow temp to such a level that the heat pump can operate at its lowest power without causing the flow temp to increase beyond its set point.

 

 

Its not at all likely that cranking up the flow temperature to reduce the output of the heat pump is an efficient way to operate (unless of course you are feeling cold).  Basically you are deliberately degrading the thermodynamic performance in an attempt to overcome a system limitation, which the heat pump itself knows how to overcome.  if Mitsubishi thought this was a good strategy I would guess that they would have implemented it, certainly in their auto adapt mode.

Most likely the heat pump is a bit oversized.  Possibly quite a bit - 6kW for a well insulated 110 sq m new build house is probably much more than is required.  You are potentially running into the problem which many comment on, namely that once houses are sufficiently well insulated they become a pain to heat efficiently whilst also having sufficient capacity for DHW.  Thats frankly a good problem to have, efficiency matters less if the absolute energy consumption is low!

Given we are currently in a mild patch, it just cannot modulate down far enough.  You have hopefully removed the main sources of poor COP (ie the external controls), whats the COP like now? 

Basically Im asking - is there evidence that cycling is materially degrading performance.  If not then I would probably let it cycle.  

Another question - are all the emitters 'engaged' (ie on) or are some turned (or turning) off?  

@jamespa

You raise very valid points all round. 

It takes a few days to see if the COP has improved. The Ecodan only provides energy consumed / delivered daily, and with all the tinkering I have done in the last 48 hours - the last two days of data probably aren't meaningful - but here they are:

Friday Jan 19th - Consumed = 28kwh , delivered = 61kwh, COP = 2.17 (pre any changes)

Sat Jan 20th - Consumed = 29kwh , delivered = 70kwh, COP = 2.4 (started implementing changes in the afternoon / evening) 

Sun Jan 21st - Consumed = 23kwh , delivered = 72kwh, COP = 3.1 (this is very surprising, given it appeared to still be cycling, but it was much warmer on Sunday...)

Tomorrow will be interesting... 

All emitters are on. 

@sunandair could you post a link to the databook please

Another interesting piece of data. 

I set my HW tank to warm up between 4am and 6am. 

Take a look at the readings from 3am to 4 am, and then from 4am to 5am (when the tank is heating), then 5am to 6am (tank reaches set temp at 5:20).

While the tank is heating - the flow temp goes up to over 50c? This doesn't seem right to me? Is this just because the temp sensors are at the source of the tank and there may not actually be any water of this temp being pushed through the rads + UHF? 

@jaket The heat pump can only either heat hot water or supply the central heating, it cannot do both at the same time.

It looks like your hot water is set to 51/52C so the heat pump has to get its water temp above that to transfer that heat to the water in the tank.

It all looks perfectly normal.

At 5.20 you can see when the hot water cycle is satisfied it returns to space heating and the flow temp drops to your WC setting.

Posted by: @jaket

↑

@jamespa

 

You raise very valid points all round. 

It takes a few days to see if the COP has improved. The Ecodan only provides energy consumed / delivered daily, and with all the tinkering I have done in the last 48 hours - the last two days of data probably aren't meaningful - but here they are:

Friday Jan 19th - Consumed = 28kwh , delivered = 61kwh, COP = 2.17 (pre any changes)

Sat Jan 20th - Consumed = 29kwh , delivered = 70kwh, COP = 2.4 (started implementing changes in the afternoon / evening) 

Sun Jan 21st - Consumed = 23kwh , delivered = 72kwh, COP = 3.1 (this is very surprising, given it appeared to still be cycling, but it was much warmer on Sunday...)

Tomorrow will be interesting... 

All emitters are on. 

Given that my advice is be patient. If you haven't already, get your system set up so that it should work well as the designers intended it to.  That means emitters on more or less constantly, external controls adjusted so they have little or no effect and the wc curve as low as it can be so your house is at the right temperature (this will take some time and obviously requires some periods where the oat is approaching its extremes).  Once you have thus working well you can tweak.

Many people report that the mitsubishi auto adapt mode works well, this tweaks the wc settings based on measurements taken by the unit.  But the best advice seems to be to get the basic wc tuned first.  Plenty have reported good results with mitsubishi heat pumps but, like any, they can be destroyed by external controls, buffer tanks, low loss headers and other generally unnecessary paraphernalia.  

I wouldn't be too concerned about cycling at this stage until you can show it's materially damaging efficiency in a system that is working as above.  If it is there are several things you can do to extend cycle time, increasing the flow temp is a way down the list.

Posted by: @gary

↑

@sunandair could you post a link to the databook please

Hi @gary I’ve located a copy of it I think it was published last autumn.

LATEST ECODAN DATA BOOK VOL6

LIBRARY.MITSUBISHIELECTRIC.CO.UK "https://library.mitsubishielectric.co.uk/pdf/book/Ecodan_ATW_Databook_Vol.6_.0_.pdf?fbclid=IwAR1D2kGseyisC6SpEs_m-gAlP-gvEClVBoxusxXu5q2JK3MqhwDWpSJKaY4_aem_Acr_ad45Iv0EDEqxXWdaNTMZ-5gyUkcC6O5rdkLHwnRLHNYQm0P2PxMT2lx8J2O97-w&fs=e&s=cl#page-132"

Posted by: @jaket

↑

How do I check the emitter value & that the flushing value is fully closed (what does this look like / where is it positioned?)

I had wondered if you have a list of your radiator sizes and how large your underfloor heating area is since a small emitter capacity can cause cycling. The flushing valve refers to your schematic diagram.... I noticed on the drawing there was a flushing bypass valve which isn’t always fitted. And mentioned it as an aside issue. Just to check that the valve is fully closed since this could cause hot water to bypass the heating system. 

Having said all that, your latest hourly graph looks better for all the changes you have made so things appear to have improved. 

Posted by: @sunandair

↑

A loss

Posted by: @jaket

↑

How do I check the emitter value & that the flushing value is fully closed (what does this look like / where is it positioned?)

I had wondered if you have a list of your radiator sizes and how large your underfloor heating area is since a small emitter capacity can cause cycling. The flushing valve refers to your schematic diagram.... I noticed on the drawing there was a flushing bypass valve which isn’t always fitted. And mentioned it as an aside issue. Just to check that the valve is fully closed since this could cause hot water to bypass the heating system. 

Having said all that, your latest hourly graph looks better for all the changes you have made so things appear to have improved. 

-- Attachment is not available --

Unfortunately I think the issue has not been resolved. See below hourly temps from 13h00.

I’m going to run it for a few days in the current setting to see what the overall consumption / COP figures are like - if these are still terrible I may try out auto adaptive. 

That will be because it was warmer outside more heat in the outside air means the heat pump has to work less and therefore more cycling.