Energy value Mitsubishi split-airconditioner

[BDVGitHub]

When the air conditioning in the bedroom is operational, it does not show any energy consumption; only the air conditioning in the living room shows consumption. So, in my opinion, the consumption I can see in the living room air conditioning is the total consumption of the split air conditioning system.

Subdivision of electricity consumption

Air conditioning consumption in the bedroom: it was heating a bit before noon.

Air conditioning consumption in the living room: it was heating the same time like bedroom air conditioning

[aremoio]

This happens also to me:
I have a split system (one outside unit serves 3 indoor units).
Only one room shows hi power load, even if it is off, and anohter room is "working"; either in "power" and "energy".

my understanding is that those data refer to the whole system (in the end it should have a single motor/pump that operates), and you can't have a specific value for each inside unit.
Also the "outside temperature" is probably the same value, repeated by each internal entity.

[niobos]

This is the data as exposed by the heat pump as-is.

My system behaves similarly: the first indoor unit on each multi-split reports the sum of both its own (indoor) power and the outdoor unit's power; the other indoor units only report their own power (which is fairly low, just the fan).

This is documented deep in the code, but maybe it should be somewhere more visible as well. Where would you expect to find this information?

[BDVGitHub]

This is the data as exposed by the heat pump as-is.

My system behaves similarly: the first indoor unit on each multi-split reports the sum of both its own (indoor) power and the outdoor unit's power; the other indoor units only report their own power (which is fairly low, just the fan).

Yes, but when I look at the displayed wattage, it always remains at 0W for the unit in the bedroom, so there is no idle consumption or consumption when it is running.

[niobos]

What type of units do you have?

Here's my graph of a "non-first" indoor unit:

[BDVGitHub]

My setup is:

• Outdoor unit: MXZ-4F72VF3
• Indoor First unit: Wall model MSZ-EF50VGK => I see temp, power consumption, ...
• Indoor Second unit: Floor model MFZ-KT25VG => I see temp, ... but no power consumption

[niobos]

My indoor units are MSZ-LN##VG, so it seems this is a feature not present on all indoor units. I'm wondering if/how we should document/expose this...

[BDVGitHub]

Maybe or its on a different adress register

@niobos Is there a tool you can scan the registers to check if thy are the same foor wall and floor model?

[niobos]

I'm not sure I understand your question correctly, but I'll give it a try: The underlying library pymitsubishi can be used directly to show all the data exposed by the WiFi module. It also shows the "unitinfo" output, that should include what type of model it is.

Using the pymitsubishi library directly is not hard, but it assumes some Python development knowledge. Once the pymitsubishi package is installed, python3 -m pymitsubishi -vv <address of pump> will give you all the bits it gets from the module. Do note that this doesn't have "registers" and "addresses". It's just a bunch of bytes that we tried to make sense of. Some bytes we know what they mean (and are exposed to Home Assistant), some bytes we don't know what they mean.

[aremoio]

Considering how these system are hardwired (mains line powers the Outside unit, and inside units are wired from the outside unit), we should question the accuracy and meaning of those power data:
-may the higher power value sum up the whole system consumption, even including those of the other internal units shown separately?
-should we worry about few watts of power (less than 10W in my case), when the active system draws 100-500 times that value?
-how really accurate those data are (even only considering a total system consumption). It should be tested against a known accurate power meter.

I took a simple experiment:
my system (3 inside units) is off, and i read: 16W, 1W, 1W.
I turn on "fan mode" in a "non primary" internal unit: both the primary unit (blue) and this other one (red) show an almost equal rise in power draw. Now i have 26W, 12W, 1W.

Should we consider the primary unit as "total system draw" ??

[niobos]

Should we consider the primary unit as "total system draw" ??

Maybe, but I don't see a way to know what unit is "the first one". So this would be up to documentation and the user configuring the right thing in Home Assistant...

As for accuracy, the power reported by the heat pump is within 5% of the power reported by a Shelly EM energy meter (Shelly reports 2.95kW for the startup peak vs 3.046kW reported by the Heat Pump itself). For energy the resolution of the heat pump is too low to compare, but for today I get 4.8kWh reported by the HP, 4.77kWh by Shelly.

[aremoio]

That seems good accuracy!

Maybe, but I don't see a way to know what unit is "the first one". So this would be up to documentation and the user configuring the right thing in Home Assistant...

yes, that would complicate the setup, also because you should consider many variants of the outside unit, with 1 to 4 or more inside units. But ideally an additional device "external unit", gathering its specific data, would be a nice implementation.

As a fast remedy, I suggest you to give more evidence of this situation in the README, not everyone goes into deep reading of the code.