I am trying to charge a capacitor using repeatable damped sinusoid current signals from a piezoelectric bender - Result of the free vibration of the piezo beam(Current source plus piezoelectric capacitance in parallel). The current from the piezo bender is rectified and used to charge a capacitor. I have simulated it Matlab/Simulink [See model in Figure 1] but it yields a weird behaviour when the input is a damped sinusoid current signals. The voltage across the capacitor never reaches the open circuit voltage but grows indefinitely.
[See Figure 2]- >This is the expected behaviour (From a sinusoidal no-damped current input): Graph1: Voltage Across Cp1(Piezo Bender) Graph2: Voltage Across Capacitor that is being charged.
[See Figure 3]->However, this is what I get (The voltage across the capacitor never reaches the open circuit voltage but grows indefinitely): Graph1: Voltage Across Cp1(Piezo Bender) Graph2: Voltage Across Capacitor that is being charged.
Would you have any suggestion or hint about what is happening here? Thank you very much for your help in advance.
Best wishes, Lucas
Hi Lucas Queiroz Machado ,
as far as I can see, the problem is not the voltage at the capacitor in Fig. 3 because if the voltage delivered by the piezo bender would increase indeed as in graph 1 then the voltage in graph 2 were to be expected.
So, the question is: Why does the bender deliver not just pure AC but AC with an increasing DC (resp. low frequency) component?
Do you get the same behavior when the piezo bender is disconnected from the rectifier? Can you adjust the parameters of the rectifier in order to get an ideal rectifier (reverse current exactly zero, reverse recovery time virtually zero)? If the simulation uses non-ideal elements: Could it be that you choose a rectifier suitable for 50/60 Hz but the frequency of the piezo bender is much higher resulting in always (more or less) conducting diodes?
Hi Joerg Fricke ,
Thank you very much for your answer. Yes, that's exactly my question: Why does the bender deliver not just pure AC but AC with an increasing DC?
Answering your questions:
1. Do you get the same behavior when the piezo bender is disconnected from the rectifier?
R: Yes, I do also get a AC plus an increasing DC. [See Figure 1 of this reply].
2. Can you adjust the parameters of the rectifier in order to get an ideal rectifier (reverse current exactly zero)?
R: Yes, it's modelled as an ideal rectifier.
3. Could it be that you choose a rectifier suitable for 50/60 Hz but the frequency of the piezo bender is much higher resulting in always (more or less) conducting diodes?
R: The frequency of the piezo beam is about 25 Hz and the rectifier is assumed ideal.
The input current in Figure 3 is as shown in Figure2_Answer_Current_Input, attached here. As you can see, the input is a periodical series of damped current signal resultant of the free vibration of the beam. The original input does not increase in magnitude overtime, but is only periodically repeated. The interesting thing is that the voltage does not increase like in Figure 3 when the input signal is just a regular sinusoidal function with no damping (as shown in Figure 2 of the question).
So, the only reason I can find is the damping characteristic of the signal, but why?
I very much appreciate your help.
Thank you,
Lucas
Hi Lucas Queiroz Machado ,
obviously, the problem originates in the controlled voltage source. Here in Fig. 1, each burst begins not at the mean value during the oscillation but at a negative value, relative to the mean value. But each burst ends with the mean value.
Unfortunately, I'm not familiar with Matlab/Simulink, so I don't know whether this behavior is due to a bug in the voltage source or can be remedied by adjusting one or more parameters.
I guess the voltage source (25 Hz) is controlled by another voltage source of about 2.5 Hz. As a first step, I would try to vary the frequency of the 2.5 Hz source by a very small amount. By increasing the period of the 2.5 Hz source by about 10 ms = 1/4 period of the 25 Hz source, perhaps the starting voltage of each burst is shifted to the mean voltage.
If it works, admittedly this is only a workaround.
- Badges
- Science topic
- Similar topics
- Force