From 4c907ac5932e0c98ae3432faaaa216eddc113aa0 Mon Sep 17 00:00:00 2001 From: Jon Little Date: Mon, 15 Aug 2022 23:08:09 -0500 Subject: [PATCH] Adding notes about mulit-phase applications. (#2151) --- components/sensor/atm90e32.rst | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) diff --git a/components/sensor/atm90e32.rst b/components/sensor/atm90e32.rst index 84ac5ede2..5d5f3524e 100644 --- a/components/sensor/atm90e32.rst +++ b/components/sensor/atm90e32.rst @@ -138,7 +138,7 @@ Here are common current calibrations for the **Expandable 6 Channel Energy Meter - 20A/25mA SCT-006: 11131 - 30A/1V SCT-013-030: 8650 - 50A/1V SCT-013-050: 15420 - - 80A/26.6mA SCT-010: 41996 + - 80A/26.6mA SCT-010: 41996 (note this will saturate at 2^16/10^3 amps) - 100A/50ma SCT-013-000: 27961 - 120A/40mA: SCT-016: 41880 @@ -172,6 +172,28 @@ a time-series-database, e.g. InfluxDB. id: ct1RAWattHours state_topic: ${disp_name}/ct1/reverse_active_energy +If the power, power_factor, reactive_power, forward_active_energy, or reverse_active_energy configuraion variables +are used, care must be taken to ensure that the line ATM90E32's voltage is from is the same phase as the current +transformer is installed on. This is significant in split-phase or multi phase installations. On a house with 240 +split-phase wiring (very common in the US), one simple test is to reverse the orentation of the current transformer +on a line. If the power factor doesn't change sign, it is likely that the voltage fed to the ATM90E32 is from the other +phase. + +The CircuitSetup Expandable 6 channel board can easilly handle this situation by cutting the jumpers JP12/13 to +allow a seperate VA2 to be input on the J3 pads. Make sure that current taps connected to CT 1-3 are on the phase +from which VA is fed (the barrel jack) and the taps connected to CT3-6 are on the phase from which VA2 is fed. See +the CicuitSetup repo for more details on this. + +If a mulit board stack is being used, remember to cut JP12/13 on all boards and to feed VA2 to each board. VA is +fed to all boards through the stacking headers. Another detail is that each voltage transformer needs to have the +same polarity; getting this backwards will be just like having it on the wrong phase. + +Note that the current measurement is the RMS value so is always positive. They only way to determine directon is to +look at the power factor. If there are only largly resistive loads and no power sources, (PF almost 1), it is simpler +to just create a template sensor that computes power from Irms*Vrms and ignore all these details. On the other +hand, one might be surprised how reactive some loads are and the CirciuitSetup designs are able to +handle these situations well. + Additional Examples -------------------