3

Being developed by us is a kinetic energy harvester that harvests energy as its magnets move through a coil. It is our goal to measure the voltage across the coil when the real system is being moved by an actuator. In order to do so, we have prepared a test bed which moves the kinetic harvester at a particular frequency.

It is important that we sample the voltage across the coil at a particular frequency and send it over UART to MATLAB for further processing. We wish to use an Arduino for this purpose.

The voltage produced by the coil is not a constant voltage. It varies and also becomes negative as magnets move across the coils. I understand that the Arduino ADC cannot sample negative voltages.

I attach a photo of the harvester and the linear actuator that moves the harvester below.

enter image description here

Below is an image of the voltage waveform. Please note the voltage ranges from -3.3 to +3.3. We are able to confirm this with the help of the oscilloscope. I wish to sample at the highest frequency possible. The higher the accuracy the better the readings would be. We intend to use an Arduino Uno.

enter image description here

If all the voltages produced by the coil were positive the ADC can be directly connected to the coil. But it is definitely not the case. Perhaps some circuit / technique to make the output waveform positive is required. We cannot provide a DC shift to this coil as doing so may change the coltage induced by magnets effecting the system.

Whats required is a method we can sample the ac waveform varying from -3.3 to +3.3 with the help of the arduino ADC.

Any advice is a lot appreciated.

Denis
  • 1,245
  • 3
  • 17
  • 35
  • Several examples are presented here: http://electronics.stackexchange.com/questions/34071/how-do-i-correct-the-offset-voltage-of-op-amps-which-have-no-explicit-offset-nul – PDuarte Feb 26 '16 at 13:59

2 Answers2

1

At some stage you're going to have to rectify the signal to extract the energy so it makes sense to do it there, after the rectifier. Your problem is that you are working with low voltages so the diode drops become significant. I recommend that you go for higher voltage, lower current so that the losses become less significant. Of course, use Schottky diodes to minimise losses.

schematic

simulate this circuit – Schematic created using CircuitLab

Figure 1. Centre-tapped coil has only one diode in circuit at any time.

Use a centre-tapped coil so that one diode conducts on one half-cycle, etc. This is better than a single coil with full-wave rectifier as it always has two diodes in circuit at any time.

Add over-voltage protection to your ADC as you may get high spikes in the event of an impact / shock motion to the device.

Also, see William Bode-Tyrrell's answer to Increasing efficiency of a kinetic energy battery charger for some good calculations.

Community
  • 1
Transistor
  • 168,990
  • 12
  • 186
  • 385
0

schematic

simulate this circuit – Schematic created using CircuitLab

You might also need a scaler if the ADC can't go above 5V, but you didn't say and I'm not reading the datasheet for you. Likewise, you might want a bit more than 3.3V if the signal is not particularly constrained.

Alternatively: https://en.wikipedia.org/wiki/Precision_rectifier

Ecnerwal
  • 5,594
  • 17
  • 24
  • That's a first order high pass filter. It will create unwanted distorsions due to attenuation and phase shift. – PDuarte Feb 26 '16 at 13:55