Three-phase electric power system

Good evening,

I am writing to seek advice regarding the monitoring of a three-phase electric power system with OpenEnergyMonitor devices.
I have seen the code of the firmware you provided on Github (link) and I would kindly ask you some advices:

  1. After the EmonTX with three-phase firmware update, do I need to modify other config files on EmonTX /EmonPi?
  2. A comment in the firmware state that “this methods is an approximation”. I am quite sure that in my test case the line-neutral measurement (CT4) will not be equal to zero, so how I can proceed in this case? do I have to set one EmonTX for each phase and each TX with one CT sensor and one AC Voltage Sensor Adapter?

Futhermore, I would like to connect more than two emonTX to my EmonPI (i.e 4 TX devices, each one with 3 CT).
Currently I have set the first emonTX with the nodeID 8 (default), the second one with the nodeID 7 by manually setting the DIP switch. How and how many devices can I connect to a single emonPi?

Thank you for your attention.
Kind regards,


Yes, you will need to copy the template (“emonhub.conf node decoder settings for this sketch:”) from the sketch and paste it into your emonhub.conf file. If you have two emonTx’s, then you must give those different NodeID’s (11 & 12 ?). Copy two configurations into the file and put the correct NodeID into each configuration and into each sketch. The sketch does not use the DIP switches, so you must have 1 copy of the sketch for each emonTx.
You must have Node 5 for the emonPi. Then, you can choose any number in the range 6-30 to use. Each device must have its own NodeID, and you must have that once only in emonhub.conf. (i.e., if you “steal” a NodeID that is already in the file, you must delete the default one.) You can edit emonhub.conf in a web browser.

The answer to this depends on how accurate you need the measurement to be. The problem is voltage balance.
Using one emonTx, only the voltage of the first phase to neutral is measured. If the voltage of the other two phases is different by a significant amount, then the power you measure for those two phases will be wrong in the same proportion.
If you can measure the voltages several times at different times of the day, you can see what the imbalance might be. Two or three volts will mean a 1% error approximately.
If you think the error will be too great, then you can have 3 emonTx’s, and connect them one to each phase. Then you do not want the 3-phase sketch. You would need 3 x ( emonTx + a.c. adapter + c.t.) and you could add 3 more c.t’s to each emonTx if you want to measure individual circuits.

I think you will not have a problem with 4 emonTx’s sending the data to your emonPi. They connect by radio.

I am not sure what you mean by that. If you have a c.t. on each line (Phase 1 - brown, Phase 2 - grey, Phase 3 - black), then you do not need a c.t. on the blue neutral. Neutral current is normal and is not a problem. You can (in the sketch) allocate C.T.4 to either phase 1, phase 2 or phase 3 and measure the power in another circuit, or if you do not want to do that, leave it unused and do not have "#define CT4LINE … " in your sketch.

1 Like

@Robert.Wall is there a document that specifies the allocation of node numbers?

Yes - in most of the sketches, also here, in Appendix 2. But that list is only a recommendation. You or anyone is free to allocate whichever device to whichever node ID they choose. I said 6-30 only because it is hard to change Node 5 (the emonPi itself), not because it is sacrosanct. But there might be issues when the emonPi is updated.

JeeLib can only use 0 - 31, but emonHub is not restricted to that.

1 Like

Hi Robert,
Thank you for you kind and quick response.