enter image description here

Hello all. The above is a circuit schematic (sorry for poor hand drawing) for monitoring power from 3-phase AC generator. The main goal of the circuit is to see how the power output varies depending on the load resistance that control the generator torque. Here are my questions.

  1. I though that when I control the PWM duty cycle, I could control the value of resistance. I verified that the resistance actually changes. But the current output is really oscillating, whereas the voltage output is pretty stable. Some people saids that connecting capacitor would reduce the fluctuation, but it seems not to work. Once I average the current value over, say 60sec, I could get the kind of representative value. However, I really wonder if there is a certain way to regularize the current output?

  2. I am always having difficulty to understand the role of ground in a circuit. In the above figure, do I need to connect the negative pin of the generator (rectifier) to the ground pin of the Arduino board? I am really reluctant to do that worrying about something wrong will happen.

  3. Finally, is there any way to monitor power from 3phase generator more efficiently? I am feeling that using rectifier causes some energy loss as well as block the natural relationship between the load resistance and the output power.

Thank you very much.

  • \$\begingroup\$ In no way is this a power monitoring circuit. Why not tell us what you really think you are trying to accomplish? \$\endgroup\$ – Andy aka Dec 3 '14 at 8:27
  • \$\begingroup\$ This is circuit is for controlling an experimental wind turbine. I am trying to control blade pitch angle, yaw angle and the generator torque through the variable resistance. For every possible combinations of control inputs, I am trying to measure the power output from the wind turbine. All these process are remotely executed. \$\endgroup\$ – Jinkyoo Park Dec 3 '14 at 10:39

Your problem is that your current sensor is fairly fast (but not too fast), as is your Arduino A/D converter. So the current is being sensed at different points in the PWM cycle and you get wildly varying current values. What you need to do is slow your current sensor down so it only sees the average current. This can be done simply enough by putting an RC between the current sensor output and the Arduino. The cutoff frequency of the RC network should be at most 1/10 the PWM frequency. If you don't want to do this, you can low-pass in software - in other words, take a running average.

If the current sensor were faster, you would only get two values, 0 and full current, corresponding to the on and off states of the MOSFET. If the sensor were slower, it would do the averaging for you.

I suspect you think that, by putting a capacitor on the gate drive to the MOSFET you can produce a DC level at the gate which will produce a DC resistance in the MOSFET. This can sort of work, but you need a more sophisticated gate filter than just a cap. Furthermore, effective resistance is quite sensitive to gate drive, so you'll find it difficult to do fine control. Plus, the resulting resistance will be fairly temperature-sensitive with respect to self-heating, and somehow I doubt you've put a big heat sink on the MOSFET.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.