Leaving Water before BUH R1T vs AFTER BUH R2T - CoP calculation descrepency

[stephenacrown]

On the main page "how to calculate CoP, it is stated in the formula to use Leaving Water before BUH R1T (as that is a common sensor across Daikin models it would seem)

If we are to use sensors rather than a heat meter, we should all understand constraints and limitations of the sensors and their locations.

However, for those with the backup heater (BUH), and even if not actively used, we are seeing a large temperature drop on the sensor after this stage:

"Leaving Water after BUH R2T"

Below shows schematic of the EDLA 9-16 models with a BUH (Backup Heater) highlighted in red:

Below shows schematic of the EDLA 9-16 models without the BUH:

There is a discussion (link below) over on OpenEnergyMonitor where someone with a MID billing grade heat meter is suggesting the R1T sensor is not appropriate for use in CoP calculations, and that R2T more closely matches their heat meter values, which is extremely useful information for this project and anyone using it.

What does this mean for those without a BUH? Can R1T be reliably used concerning accuracy?

https://community.openenergymonitor.org/t/daikin-9kw-ashp-application-discussion/24848/101?u=stephen_crown

Is anyone else aware of why there is such a large temperature drop (0.8 -14% to 1.8C - 25% drop) have been seen between stages, which feels "off" given the heat loss that would infer.

E.g.

14% of minimum output on the EDLA9-16 units appears to be ~4-4.6kW - and that is then a heat loss outside within the unit of ~640W...

14% of output at design temperature is say 9kW, and the heat loss outside within the unit is a staggering ~1.26kW...

Example of a period where the difference is 0.8C (~14% difference in heating power at the time)

Example of last 2 hours delta between R1T and R2T which as you can see varies a lot:

Photo below captured during Octopus Energy Services installation shows large amount of large diameter pipework exposed without insulation:

[rufik]

Well, really interesting. I've created COP template sensors for R1T and for R2T, so I'm going to follow the both numbers.

[JKGBIT]

👍 I also have seen this, and switched my own code to R2T, disapointed im not actually getting a cop of 8.9 lol. As the old saying goes, "if it seems to good to be true, its probably a duck" - something like that anyway. 4.8 Cop is okay too I suppose. 👍

Interesting about the insulation, wonder if anyone has seen improvement from insulating it?

[rufik]

Using R2T results in about 10% lower COP for me. It's just rough math here, I need to create a template sensor as % of difference I think...

[JKGBIT]

I expect that R1T Is very close to the heat exchanger and will be picking up on the emitted heat from that and is hotter than the water through it. The controller probably uses that to check the heat exchanger is working & outputting heat normally and will kick in the backup heater if required and allowed by settings. And probably not a reliable source of data for cop. Mine is 2.5° off R2T.

[jonputsman]

While I agree that R1T looks too good to be true for use in COP calculations, I tend to see R1T being used to match the leaving water temperature set point:

I’m going to start logging flow and return inside my property, as I think my system is delivering a wider flow-return delta than I’ve requested.

[JKGBIT]

Interesting. I think I saw someone on forum put a request into Daikin, so wonder what they said about it.

[jonputsman]

That could have been my post. I’ve heard nothing significant from Daikin regarding the various temperature probes.

[stephenacrown]

Interesting. I think I saw someone on forum put a request into Daikin, so wonder what they said about it.

@JKGBIT

Here is a screenshot of the Daikin response to Matt's question on this point for his Daikin Altherma 3M EDLA09D3V3 heat pump that includes the BUH:

See here for the Daikin response to Matt:

https://community.openenergymonitor.org/t/daikin-9kw-ashp-application-discussion/24848/193?u=stephen_crown

I will also add, that models without a BUH, the R1T sensor is in theory actually in the same position as with a BUH.

I found a source from Daikin for the EDLA04-08 models that shows this clearly.

Without BUH:

With BUH:

However the service manual for the EDLA 09-16 does somewhat confuse things as it doesn't match the lower end model (4/6/8), nor Daikin engineer response, nor Matt's MID Compliant Heat Meter.

Without BUH:

With BUH:

The above 2 images would indicate that models without BUH do not have a reliable sensor!

[rufik]

I've made a "COP template sensor" using R1T oraz second one using R2T temperatures. Than I've made a sensor with difference between these two COPs. My LT EHVH DA6V (8kW) has a BUH (4+2kW) and the diff between COPs like like that:

As you can see COP R2T is a little but higher than COP R1T :) The average diff is about 0.4°C, maybe 0.5°C. The percentage diff is about 10%:

I use ESPAlthera software from 2021.12.06 with def a little bit tuned, but the essential sensors are:

LabelDef PROGMEM labelDefs[] = {
    {0x21,0,105,2,-1,"INV primary current (A)"},
    {0x21,4,101,2,-1,"Voltage (N-phase) (V)"},
    {0x61,8,105,2,1,"Inlet water temp.(R4T)"},
    {0x61,2,105,2,1,"Leaving water temp. before BUH (R1T)"},
    {0x61,4,105,2,1,"Leaving water temp. after BUH (R2T)"},
    {0x62,9,105,2,-1,"Flow sensor (l/min)"},
}

Is COP diff really that small (~10%) for my HP model or should I update ESPAltherma software because of some bugs fixed? :)

[johngouk]

I have an EDLA09D3V3 with BUH, and I had a significant discrepancy between R1T and R2T, of around 2C, and R2T and R4T of 1.5C. However, after I told Daikin support about this, I had an engineer visit and remove a link on the tape heater power block X7M 5-7. The tape heaters E11H and E12H, on the plate heat exchanger outlet and inlet respectively, which had previously been on whenever the circulation pump was running, stayed off, as they should according to the service manual! The R1T, R2T and R4T readings are now within 0.1C of each other. So it's worth checking this - there is a technical service bulletin on it apparently.