Hey guys,

I'm working on a custom PCB to measure the current with a CT-sensor and a Arduino Giga.
I used this article as my reference: https://docs.openenergymonitor.org/electricity-monitoring/ct-sensors/interface-with-arduino.html

This is how my setup look right now:

Connections:
- A6 = analog 6 on the arduino
- P1 = connector to the CT-sensor

Are there any improvements I can implement or does this setup looks good to you guys?

Kind regards,

Rens

 

The burden resistor’s value is WHOLLY dependent on the particular current transformer you choose. Have you selected the CT brand and model yet?

If not, choose one first , and then read the datasheet to determine the correct burden resistor value.
Some CTs already have the burden resistor integrated into the assembly.

 

The 10µF filter capacitor acts as a low-pass filter with a 3.2Hz corner frequency.
I assume you would want it closer the the 50-60Hz line frequency.

 

I assume you would want it closer the the 50-60Hz line frequency.

I disagree. There should be no current in the capacitor. Only uAs. None of the CT's current glows through the C or Rs.
The 2R & C are to make a 1/2 supply reference that has no load.
I want the RC time to be much lower than 50 or 60hz.

 

The burden resistor’s value is WHOLLY dependent on the particular current transformer you choose. Have you selected the CT brand and model yet?

If not, choose one first , and then read the datasheet to determine the correct burden resistor value.
Some CTs already have the burden resistor integrated into the assembly.

Hello Schmitt,

Great comment, for this setup I chose the following burden resistor: https://www.amazon.nl/BAIRU-Sct-013-000-Open-Close-Stroom-Sct013000/dp/B0C85ZJXFV

Specifications:
- Windings/turn = 1:1800
- 100A primary current
- 50mA secondary current
- Supply voltage = 3.3V from the arduino

This should be good right with my shunt resistor. I calculated as in the page: https://docs.openenergymonitor.org/electricity-monitoring/ct-sensors/interface-with-arduino.html

 

I can read some German, but Dutch makes it more complicated.

EDIT: never mind. I see you selected the first one

Which exact transformer variant do you got? These transformers you showed already have the burden resistor included. And the output voltage is 1 volt at its rated current. If you add an external burden resistor it will produce a lower uncalibrated voltage.

From the list below, copied from the description, one are you using? Only the first one would require a burden resistor.

Kenmerktypen: 100A 50mA / 30A 1V / 50A 1V / 10A 1V / 20A 1V / 60A 1V / 100A 1V (optioneel)

 

I can read some German, but Dutch makes it more complicated.

EDIT: never mind. I see you selected the first one

Which exact transformer variant do you got? These transformers you showed already have the burden resistor included. And the output voltage is 1 volt at its rated current. If you add an external burden resistor it will produce a lower uncalibrated voltage.

From the list below, copied from the description, one are you using? Only the first one would require a burden resistor.

Kenmerktypen: 100A 50mA / 30A 1V / 50A 1V / 10A 1V / 20A 1V / 60A 1V / 100A 1V (optioneel)

Haha, yes sorry for that.
But are the values OK now?

 

I disagree. There should be no current in the capacitor. Only uAs. None of the CT's current glows through the C or Rs.
The 2R & C are to make a 1/2 supply reference that has no load.
I want the RC time to be much lower than 50 or 60hz.

The CT doesn't contribute any current to the half-supply divider, so decoupling isn't strictly necessary.
If you have a decoupling capacitor then you need a series resistor between the CT output and the input pin. Otherwise the first current spike that comes along will blow up the input pin.
Without a decoupling capacitor the half-supply divider resistors would limit the current.

 

Yes, 22 ohm should give 1.1 VRMS, which is equal to +/-1.54 Vpk-pk, which will fit nicely in an Arduino powered by 3.3 VDC.

 

I disagree. There should be no current in the capacitor.

Correct.
I misread how the circuit works.

 

Whilst this is probably not a "normal" way to do the job, here is the circuit of half of the 4 analog inputs of an industrial controller that we designed and has been successfully used for many years.
As shown, the inputs are set for 4-20mA, using the 10R termination resistor. But, if you leave that off, and use the correct burden resistor, the opamp runs as an active rectifier for a current transformer and its gain can be adjusted to give a calibrated DC reading for the AC current. These have seen very reliable service in pump control for example.
It is just a thought.

 

anything that connects to mains should have reasonable protection.