Performance of 9kW Daikin

@tiger_cook I am interested in learning what your 8kW system can do, probably in a Jim Bowen “Look at what you could have won” way…! :joy:

But perhaps a separate “Performance of (4kW to) 8kW Daikin” thread would be useful to keep the discussion distinct from the 9-16kW troubles?

Hi Jonathan,

Thanks for that.

I am looking forward to my visit from Daikin next week.

I hope I can keep it constructive and get them at least appreciate and acknowledge the issues that I am facing and I expect every other owner of this heat pump will be facing.

It just doesn’t run low enough in relation to it’s nominal size.

The last Daikin engineer I spoke to agreed that there is a massive gulf in reality between the 8kW and the 9kW.

And it is because it is not a 9kW heat pump, it’s a de-rated 16kW heat pump.

That is where the physics are involved.

The compressor is already running at 25% capacity at its lowest level in our 9kW heat pumps as it produces 16kW flat out.

I doubt they can turn it down any lower and for that reason they cannot reduce the flow rate any lower than 10lpm

I understand what they have done and why they have done it but it has backfired massively, it doesn’t work as well as a heat pump designed as a 9kW would have done.

I bet the 14kW and 16kW are fantastic in an appropriately sized home.

But we have a 16kW heat pump in houses that need 6 or 7kW of heat at most and usually require 2 to 3kW.

It’s like having a car that has a minimum speed of 50mph and if you want or need to go slower you have to slip the clutch and apply the brakes.

1 Like

I hope they’ll be able to explain why performance is so poor at the start of a cycle. Even better if they could fix it, so that cycling the heat pump becomes a practical way to run your system in mild weather.

1 Like

hi @squarepeg77 Maybe its worth continuing on the original forked thread for the 8kW and below units -
https://community.openenergymonitor.org/t/help-daikin-configuration-wd-curve-and-modulation-with-a-madoka-room-thermostat/25089/7

I feel the pain of others, hopefully without sounding unempathetic, we are really please with our system installed by Octopus (not the government increasing the incentive just after our install, you win some, you lose some) - we had a great lead installer who was a ‘heat geek elite’, however we did have a full re-pipe and almost all the radiators replaced that took 2 weeks to install. I am a telecoms architect and appreciate the detail and intricacies of complex systems, I am just trying to squeeze a little more out of system without obsessing, I love the data and visibility of stats, I DO think Daikin has too many variables and I am sure they could simplify the interface, manual and configuration options to accommodate 80% of use cases.

It would be great if Octopus/Daikin could offer an after care ‘optimisation’ or ‘fine-tuning’ service (happy to pay) in addition to the warranty break fix contract.

3 Likes

Hi Tim,

It is actually much better with the new enormous radiators. I not enjoying the on for two hours and off for three hours but that is how the Madoka is running it based on a set room temperature of 22c.

What I need is more heat loss!!! The essence of the problem, the house doesn’t lose enough heat, it will have been an average of 10c outside today and my heat loss is 2kW with an average indoor temperature of 22.7c

The first 30 minutes is OK, then it goes wrong for 30 to 40 minutes and then it’s really good by which time it will soon be ready to turn off.

I have mentioned my observations about its apparent inability to hold a flow temperature and I see that however I run it.

Here is the latest run, all it is well until the flow temperature drops, the dT closes and the heat output drops, then the flow temperature recovers and the dT increases and therefore so does the heat output.

I have called it the `Daikin dip’ before and I see it on others too, its where the performance goes in the initial phase.

The electricity use is also large at first, in a small way down to the massive circulation pump but I think primarily due to the large compressor from a 16kW heat pump.

This morning was not so bad but I cannot explain the difference, the dip is less pronounced. What there is, is a boost to the flow temperature that maintains the dT and doesn’t let it close like it usually does. The drop in flow rate from 25lpm down to 10lpm took 45 minutes this morning compared to only 30 minutes in the example above.

If I could control the flow temperature and flow rate manually I think it could be eliminated, that means it could be programmed out or programmed better, but not by me, not my skill set unfortunately.

My experiment for tomorrow is a Delta T of 4C.

I figure since we’re hitting the minimum flow of 10.7 litres per minute a lot, the pump has nowhere to go and the control algorithm just has to wait it out.

With a dT of 4C, this should allow the pump to run at around 13.5 litres per minute for the same ~3.7kW heat output.

Maybe that will give the system a bit more room to manoeuvre.

2 Likes

Hi Jonathan,

Give it a go, you have nothing to lose.

Octopus set mine up as fan coils with a dT of 5c and I have tried 4c too, and 7c.

I have always found radiators works best with mine, it is always trying to get to a dT of 8c, I sometimes get there but I always get the highest dT it can give and the best efficiency.

I would try radiators too if you haven’t already.

I know its not same system but: I’ve done this with my samsung and its improved stability a lot. It would keep on dropping the flow rate trying to achieve DT of 5, every time the flow rate would work its way slowly down to 12lpm (the min), just about get there, then the control algorithm would (presumably) say to itself something like - “I’d like to lower flow a bit more but I can’t because its below my min, so instead of keeping it steady where it is, I’ll throw my toys out the pram”… The control loop would go through a full reset, stop the compressor, restart the compressor (at full speed), back to full flow rate, etc.

By setting a DT target of 4, my flow rate now stabilises at around 15-16lpm and I get almost none of these cycles, for a steady ~4kw output.

2 Likes

I don’t have this available.

I can only limit the pump speed when sampling.

Maybe it is only an option with fan coils or only on the 4kW to 8kW range?

Or maybe only if you have more than one heating zone with circulation pumps other than the one in the monobloc?

Where did you find this @tiger_cook

Hi @matt-drummer I have the MMI user guide, hopefully I can attach the pdf its 5MB, let me know if you can download?
D series MMi.pdf (5.2 MB)

Thanks Neil,

I was able to download it.

I have never seen this document before, it is nothing like I was supplied with by Octopus.

I really don’t have the option to to limit the pump speed as shown on page 61 but it does show two zones.

I wonder what would happen if I set up another zone even though I only have one?

Do you have more than one zone enabled?

If not do you have this option available?

Are you in installer mode? ‘5678’

Yes

I have found that I can limit the pump speed to 60% all the time.

I had to scroll up.

I doubt it will do much but we’ll see

Thanks for your help.

1 Like

Well, the data from this morning running with delta T of 4 is here:
https://emoncms.org/app?readkey=0e9ab50e73d8ae59d9e6dcc70e5dd5f7&mode=power&start=1704259100&end=1704287120&cop=1&flow=1#myelectric

My observations are:

  1. The pump speed was up at 13 litres/minute, as expected
  2. the first cycle was longer than yesterday’s, before the first cooling phase
  3. the flow temperature went above target set point more slowly
  4. it took longer to settle down to dT 4C
  5. the COP of 3.3 was lower than yesterday’s first cycle of approx 3.6

Note: I still haven’t calibrated my R2T / R4T values, so maybe my real COP isn’t quite this low.

I had an interesting chat with a Daikin “Product Champion” today. He was very friendly, knowledgable and was interested in the monitoring we are doing here.

It does seem like the 900W minimum power is the floor at which the compressor can operate, so I don’t think we can expect any changes there.

I did learn something about the emitter type selection that I had not read before: radiators and fan-coils are fast response, while underfloor heating is slow response. He suggested I could try changing my emitter type to UFH to see if that helps the unit settle to the delta T better.

So, I have tomorrow’s experiment: UFH with a delta T of 4C.

1 Like

Hi Jonathan,

I tried UFH once.

It didn’t help in my house but then as yesterday, you have nothing to lose by giving it a go.

What do you mean by calibrate your R2T and R4T values?

Surely they are what they are, you can do something with them?

By calibration, I mean this:

I see.

HI @squarepeg77 I also tried a DT of 4 last night and noticed a lower COP in HA with ESPAltherma gone back to DT5, + I have just created a better additional sensor that averages the COP template sensor over 5 mins, I added my workings/config to the other ESPAtherma Home Assistant thread, hope this helps?

great write up of the results, really interested in the UFH results with a DT of 5 as we are on a fancoil configuration…

1 Like

hi @squarepeg77 it would be good to see the power consumption (fluctuation) from HA of your unit with UFH vs fancoil, I have noticed on our 8kW system the Daikin fluctuates the input power in our case from ~250W to ~450W (from a shellyEM) to control the flow temps and DT5, if it could stick with ~380W, I think it would be stable but its appears to be not that intelligent, maybe it’s a compressor speed limitation?

Hi @tiger_cook, even on fancoil our power usage is pretty smooth.
Here’s the data from 2nd January, DT5:

and here’s a smaller time interval:

1 Like