A charge pump, or switched capacitor voltage converter, uses capacitors as energy storage elements to generate an output voltage. For example, one basic charge pump circuit, the “doubler,” doubles the input voltage, using a single flying capacitor and four internal switches driven from a two-phase clock. In the first phase of the clock, a pair of switches charges the flying capacitor to the input voltage (VIN). In the second phase of the clock, a third switch connects the negative terminal of the capacitor to VIN effectively generating 2*VIN at the positive terminal of the capacitor. The fourth switch connects the positive terminal of the flying capacitor to the output capacitor. Under no load conditions, charge will transfer to the output capacitor on each cycle until the output charges to 2*VIN thus doubling the input voltage.
When an output load is present, the output capacitor provides the load current during the first phase, while the flying capacitor provides the load current and charges the output capacitor during the second phase. For charge transfer to occur, the output will regulate at a voltage slightly lower than 2*VIN. The charging and discharging of the output capacitor in the two phases of the clock generates an output ripple that is a function of the output capacitor value, the clock frequency and output load current.
All other charge pump circuit implementations follow from this basic scheme by adding/changing switches and capacitors as well as the number of phases of the clock. Charge pumps can double voltages, triple voltages, halve voltages, invert voltages, fractionally multiply or scale voltages such as x3/2, x4/3, x2/3, etcetera, and generate arbitrary voltages, depending on the controller and circuit topology. The efficiency of charge pumps can be quite good when near their ideal charge ratio.
In the doubler example above, the input supply will be equal to two times the output load current such that input power equals output power in the ideal case. In reality the efficiency will be slightly lower than ideal due to quiescent operating current and other losses. In reality, the efficiency will be slightly lower than ideal due to operating current and other losses. The versatility of charge pumps enables their usage in a wide variety of applications and market segments.
Charge pumps fill a niche in the performance spectrum between LDOs and switching regulators and offer a nice alternative to designs that may be inductor-averse. Compared to LDOs, charge pumps require an additional capacitor (a “fly” cap) to operate but do not require inductors, which are generally slightly more costly, have higher output noise levels and usually have lower output current capability.
However, they have many benefits over LDOs such as higher efficiency, good thermal management due to switching architecture and have more flexibility to step a voltage up as well as down, or generate negative voltages. When compared to conventional switching regulators, a charge pumps’ output current capability and efficiency are lower. However, they are simpler to design and do not require an inductor. Furthermore, advancements in process technology have enabled an expansion of charge pump input voltage range compared to previous generations.
Table 1 provides a comparison of key performance parameters between topologies.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.