The Harvester circuit
The proposed energy harvester operates in three separate phases: pre-charge, harvesting, and recovery. The variable capacitor is first pre-charged to the battery voltage via a quasi-lossless inductor, as shown in Figure 3. The pre-charge phase is therefore decomposed into a sequence of two steps. First, energy is transferred from the Li-Ion battery into the inductor by superimposing the battery voltage across the inductor with switches S1
. The inductor current then increases linearly, and when sufficient energy is stored, S1
open. The second step is to drive the stored energy into the variable MEMS capacitor by connecting the inductor to the capacitor with S2
. The inductor current ultimately charges the variable capacitor until its voltage reaches that of the battery, at which point S2
Figure 3. Pre-charge phase
During the harvesting phase, the MEMS capacitor is connected to the rechargeable battery with S5 so that charge can be driven into the battery (Figure 4). Since the battery is a low impedance source, the voltage across the battery is practically constant and the decreasing capacitance therefore produces a harvesting current (I = C dV/dt + V dC/dt ≈ V dC/dt). Although this current charges the battery, the resulting increase in voltage is minimal because the magnitude of the current is low, which is why the capacitor voltage is still constrained.
Figure 4. Harvesting phase
To complete the cycle, after the minimum capacitance is reached and harvesting ends, the energy remaining in the capacitor is recovered with the same inductor used in the pre-charge phase (Figure 5). In essence, the pre-charge sequence is reversed: the residual energy is transferred into the inductor with S2 and S4 and later delivered to the battery with S1 and S3. After the capacitor is fully discharged, its electric field no longer exists and its plates are free to move and return to their minimum separation state without requiring any electrical energy.
Figure 5. Recovery phase