Using split CT termination for better THD performance
In the ADS131M02 datasheet, there is a following schematic snapshot for interfacing the ADC differential input channel with the Current Transformer:
Furthermore,
Two identical burden resistors in series are used with the common point being connected to GND instead of using one burden resistor for best THD performance.
How does this work - how does it improve THD performance?
2 answers
Your question is answered in the very next sentence in the datasheet after the one you quoted:
This split-burden resistor configuration ensures that the waveforms fed to the positive and negative terminals of the ADC are 180 degrees out-of-phase with each other, which provides the best THD results with this ADC.
Any measurement system will have some non-linearities. Such non-linearities create frequencies in the result that were not in the input. We call those fictitious frequency components "distortion".
The non-linear effects are probably not symmetrical about the analog input midpoint. By driving both inputs with an equal but opposite signal, the result does become symmetric, cancelling out some of the non-linear effects.
Added
I see Andy's answer mentions another effect. His answer isn't technically about THD (total harmonic distortion), but is nonetheless another issue you should consider. You want to ignore common mode signals caused the the capacitance between the primary and secondary in the current transformer, while measuring only the differential mode signal.
The suggested circuit addresses both issues simultaneously (THD and common mode noise rejection).
The part I don't "see" is how split-burden configuration ensures that the waveforms fed are 180 degrees out-of-phase with each other
The transformer secondary puts out a voltage that the A/D is supposed to measure. It doesn't care how you interpret this voltage.
If you ground the bottom end, then you can consider the top end to be the output voltage. If you ground the top end, then the bottom end is the output voltage, although flipped in sign compared to the previous. You could say that grounding the bottom gives you the full positive voltage at top, and grounding the top gives you the full negative voltage at bottom.
If this secondary had a center tap and you grounded it, then you'd get half the positive voltage at top and half the negative of that at bottom. Splitting the burden resistor and grounding its center point achieves the same thing in this case. Half the secondary output will be across the top Rb, and the other half across the bottom Rb. If you consider the center 0 V, then the top will be positive and the bottom negative. They will have the same magnitude, but negative of each other. That's the same as 180° out of phase.
How does this work - how does it improve THD performance?
As with any transformer connected to a live wire, there will be an AC voltage common mode signal transferred by winding capacitance to the secondary. If you want that common-mode signal to be balanced (and hence easily cancelled by the ADCs differential input), you would use a split burden resistor.
If your application isn't that important then you would use a single burden resistor and imbalanced ADC input. That would let a certain amount of unwanted common-mode signal into your desired signal chain. That unwanted signal will be unrelated to your true CT signal hence, it's a contaminate and can be "lumped in" as a distortion artefact. Maybe that's a factor they are talking about?
0 comment threads
0 comment threads