Aeotec Home Energy Meter gen2 ZW095-C

How do one set these values? Do I go to advanced senttings on the 3-phase device and from there to Raw Configuration Parameters? If so How the raw parameters are set? It says id, size, value; etc. ID is 101 or 102 or 103 and value in all cases is 4144911? What is the size?

If the question seems noob, thats because I’m new to homey… :wink:

@KevinLe I installed your version and it worked instantly and correctly. Thanks! You saved my day!

Hi! I want to install your app for the Aeotec Devices. Could you help me how to install? I don’t understand Github :smiley:

To install the source from Github
Please install the 3 steps here
Once installed, download the @KevinLe Github source, extract and access the file directory.
Using the Command Line tool (Window), Terminal (Linux) : Run command

athom app install


the size is 4 (byte)

@KevinLe Any chance that you can update the latest version of the Aeotec app with you fix? There are new devices that are only supported in the new versions.

Interested to know too :smiley:

Any idea of what is this branch for ? Seems it is even recently updated.

@Sharkys The maker of the Homey app for Aeotec devices.

1 Like

Wow, so * Home Energy Meter Gen5 - 3 phase - ZW095-3C is officially supported ? Wow, great. Thank you.

I successfully installed HEM Gen 5 (3 phase) and connected it to Homey. I can see Power and consumption, but Voltage and Current remain empty. Any ideas? Connection is according to manual.

I am using it to measure HeatPump consumption. Power is a little bit high (4 kW), Heat Pump is only 2,85 kW according to manual.

Strange, its working fine for me.

229 x 1.18 is not 132.56 W

Feel free to report it as bug then, I was responding to query about empty Voltage and Current

Actually it’s not the official app. Some

Interesting, I have TP Link Kasa devices and it reports the same… D
So we have two apps doing the math wrong?


Some difference might be due to the fact that all electric power is not active power -> watts.

What I mean is that your current measurement shows the active and reactive current combined. Meaning that with 0,13A*239,03V=31,07VA not W. The measurement shows that your active power is 17,34W. This means that you have reactive power of sqrt(31,07ˆ2-17,34ˆ2)= 25,78VAr

I’m not saying that this completely explaines the difference, but to keep in mind. More info

Other error comes about the update deadband of the measurement device. I don’t know for sure but for example there can be a 0,1A deadband for current update and 1W deadband for active power update. The current deadband of 0,1A amps equals about 23W of active power. So these low wattage and current levels I think this is the cause of error.

1 Like

Thank you. Never thought it’s so complex :slight_smile:

Anyway, for the last comment about the deadband, I was just sharing two devices used in Homey, where one is Aeotec GEN5 3 clamps meters and the second is TPLink HS110 , while HS110 shows same readings with the native app same as Honey TPLink app.

So I would say there is nothing wrong with the values reported/calculated as such rather then our understanding what they really correspond to, or?

Thank you for info on Active/Reactive power!


Actually I came up also one factor that affects the measurement:

The accuracy of the measurement. The Aeotec claims that the HEM has 99% accuracy. (The TP Link doesn’t give the accuracy level at all) This actually doesn’t tell much and can be interpret as anyone pleases. In industry the measurement error should always be informed by amplitude error and phase error, these both affect the total measurement accuracy of the active and reactive power. Let’s not dive any deeper in that end.

But all and all it has to noted that these home automation devices are not up to industry standards and should not be interpret as such. In other words the measurement accuracy of these devices can be quite bad at least with low amperage and wattage values.

What comes to the deadband: The TP Link app and Homey TP Link device should give the same results as you described and this has nothing to do with the deadband. The deadband I meant was for example following:

The report deadband for current is assumed to be 0,1A and for the power 1W.

  1. The consumption is first 1A@230V=230W -> Lets assume the device will report the correct values 1A and 230W
  2. Consumption changes by 20W into 250W -> The device will report the power change to 250W because the deadband for reporting is 1W but not the current change into 1,087A because the change in the previous value is not over 0,1A. So the updated values in the device are 250W and 1A which is incorrect.

This is only an example to show the problem with the reporting. The deadband values can be something else in real device (for example I have 50W deadband in the HEM and the current report deadband is not changeable and is not known) The deadband is still always needed because the measurement update cannot be “realtime”, it would overload the zwave network.


I’ve just purchased a 3ph meter.
Working great - For most of it.

Problem 1:
Clamp 1 - Perfect watts, amps and voltage.
Clamp 2 - Perfect amps and voltage (Watts seems to be about half of what it’s supposed to be?)
Clamp 3 - Same as clamp 2.
HEM - Incorrect sum because of clamp 2 and 3 errors.

For clamp 2 and 3 it seems that the wattage calculation is closer to being calculated with 110v.
Could that be the error?
Though it shows the right voltage (About 230v) on each clamp.

Problem 2:
I have solar panels - And it makes a lot of errors on HEM when solar power is produced.
Is it possible to fix so the clamps can measure negative values?