Hello everyone,

I'm working on measuring AC current using a current transformer (CT), the CT I’m using has a transformation ratio of 50:1, with a maximum current rating (Imax) of 50 A.
I’ve experimented with several burden resistors (4.7Ω, 1.2Ω, and 0.22Ω) and connected a logic analyzer to visualize the waveform and obtain the Vrms values.

Schematic:



Waveform (using a 4.7Ω burden resistor):


The waveform corresponds to an expected primary current of approximately 6.21 A. However, when calculating the primary current, I get:
Iprim = Isec *50 = (0.378/4.7) * 50 = 4 A
also tested another CT, the SCT-013, which has a 30A/1V ratio and comes with a built-in burden resistor. After repeating the same steps, I obtained this waveform:



So Iprim = Vrms * 30 = (212,806/1000) * 30 = 6,38
This result is much closer to the expected value.

Could anyone help me understand the discrepancy in measurements with the first CT? Any guidance would be greatly appreciated!

Best regards

 

the spec says the transformer is rated for 0.2VA and it looks like you are exceeding that by a large margin. sounds like core saturates and then output is of course less than projected.

since max output current is 1A, and maximum power is 0.2VA, then the maximum voltage output is just 0.2V (0.2VA/1A=0.2V). this is far less than your nearly 400mV. so burden resistor must not be more than 0.2 Ohm, I would probably go for 0.1 Ohm.

 

the spec says

What spec?

 

there is a link in the post with product specs

 

there is a link in the post with product specs

OK, thanks.
I missed that.

But the top spec says 50A(?).

 

that is the measurement range (50A input).
which scales down to 1A of output, at 0.2VA max

 

that is the measurement range (50A input).
which scales down to 1A of output, at 0.2VA max

Ok, I'm up to speed now.

 

since this is a low power device, and initially burden resistor was way too large, it is conceivable that CT may be damaged (at 30A input that is 3VA which is 1200% overload). and if burden resistor was ever disconnected while sensed current was present, that is almost a certainty. which is why many products have the resistor built in - no accidental disconnects.

 

So as you noted, the burden resistor should be selected so the output is no more than 200mVrms for the maximum expected current you want to measure.

 

Logic analyzer ?!?

 

the spec says the transformer is rated for 0.2VA and it looks like you are exceeding that by a large margin. sounds like core saturates and then output is of course less than projected.

since max output current is 1A, and maximum power is 0.2VA, then the maximum voltage output is just 0.2V (0.2VA/1A=0.2V). this is far less than your nearly 400mV. so burden resistor must not be more than 0.2 Ohm, I would probably go for 0.1 Ohm.

Yes, I agree with you. I've also tried using a 0.22Ω burden resistor. When measuring the same device with an expected current of 6.21A, I obtained a similar result:



Iprim = Isec *50 = (0.0285/0.22) * 50 = 6.495 A

However, the issue arises when measuring very low currents. For example, when there’s no load (no power consumption), the waveform looks like this:



And the calculated primary current is:
Iprim = Isec *50 = (0.006/0.22) * 50 = 1.26 A

This is much higher than expected! I encounter the same issue with other low current measurements as well, where the readings are significantly higher than they should be.

 

Logic analyzer ?!?
View attachment 333459

Logic analyzer is the device to visualize signals : Logic analyzer

 

Logic analyzer is the device to visualize signals : Logic analyzer

Do you have the logic pro 8? You need higher resolution and/or low-noise gain (with curve corrections) on the CT signal for good low current measurements.

 

also you are applying linear model to a nonlinear device.
even the "linear" region is not very linear.

so you probably want to use an amplifier to boost signal to desired level, as well as do some linearization in software.

 

also you are applying linear model to a nonlinear device.
even the "linear" region is not very linear.

For the best linearity, you can use a transconductance (current to voltage) converter, which will keep the burden voltage at essentially zero, but that would involve an op amp circuit with a 1A current buffer, and voltage supplies to power it.