No abstract

No abstract

No abstract

Harvesting energy from the environment by using a thermoelectric generator (TEG) or photovoltaic cells provides a solution for battery-free sensor networks or electronic healthcare systems. In these systems, the harvested energy is supplied at a very low voltages, requiring a low-startup-voltage power circuit for kick-start from low voltage. A previous sub-100mV-startup-voltage boost converter [1] was implemented by using a mechanically assisted step-up process that needs vibration at startup and the application is rather limited. In this paper, a 95mV startup voltage step-up converter without any mechanical stimulus extends the applicability of energy harvesting. The circuit converts a 100mV input to a 0.9V output with 72% conversion efficiency without any external clocks or mechanical switches. A capacitor pass-on scheme eliminates an additional external output capacitor that functions only at the startup.The minimum startup voltage of a step-up DC-DC converter in standard CMOS technology is limited by the oscillator since a clock signal is required for a charge pump (CP). Practically, the minimum supply voltage of an oscillator (V DDMIN ) is limited by PMOS-NMOS V TH unbalance caused by within-die and dieto-die V TH variations. The within-die V TH variation issue can be solved by increasing the channel width and thus it is easily solved in this application. The problem is the die-to-die variation, which can usually be adjusted by controlling the body bias of the MOSFET. In the startup circuit, however, the body-biasing circuit is not functional since the body-biasing circuit needs another oscillator, which does not function below V DDMIN . To solve this problem, post-fabrication V TH programming is applied to the oscillator. Figure 12.1.1 compares the conventional and our step-up converter architectures.In the conventional approach, two external capacitors are required. One (C OUT1 ) is used in the charge-pumping startup circuit and the other (C OUT2 ) is for the boost converter. C OUT1 is needed to separate the boost converter from the CP at the start-up. Otherwise the start-up CP needs to drive the boost converter control circuit, which is seen as leakage current (I Leak ) for the CP. In our scheme, C OUT1 is eliminated by using the output capacitor of the booster, C OUT2 , as the charge buffer of the CP as shown in Fig. 12.1.1(b). Once the C OUT is charged without leakage, C OUT is "passed on" with its charge inside to the boost converter. By doing so, the large external capacitor of C OUT1 is eliminated.The detailed block diagram of the proposed step-up converter and the timing chart are shown in Fig. 12.1.2. The capacitor pass-on controller contains a CP for PMOS super cut-off (PSC) and two voltage detectors D1 and D2. The CP for PSC provides the overdrive voltage to the gate of PMOS switch and reduces the leakage current. At startup, C OUT is charged by the CP and V START rises. When V CP1 (=V START ) is pumped to the preset trip voltage of D1, the node Y changes from low to high to pass on the C OUT to ...

No abstract

No abstract

No abstract

No abstract