Power circuit without using inductors

[sagnikbasu]

Most of the power circuit today use inductors for voltage conversion. Can we design one for negative dc -dc voltage conversion without using any kind of inductors ?

Can charge pump be considered for power circuit applications ?

[KhronX]

www.intersil.com/content/dam/Intersil/documents/icl7/icl7660.pdf

https://datasheets.maximintegrated.com/en/ds/ICL7660-MAX1044.pdf

... But don't expect a lot of power out of them 

That being said - http://electronics.stackexchange.com/questions/216094/why-are-charge-pumps-only-used-for-low-current-applications

[Codebird]

There are a few ways. You don't see them very often because they are just inferior in most ways to inductors.

Inductors: Compact, efficient, the modern solution in most circumstances.
Capacitor charge pumps: Efficient, but only practical for low current.

And the old-school ways, from before SMPS was practical:
Motor-generator sets: Bulky, noisy, inefficient, unreliable.
Battery switching: Bulky, inefficient, needs frequent servicing.
Vibrator-transformer: The archaic version of the SMPS, and it still uses inductors. Used to be used in a lot of car radios to power valve circuitry.
Electrostatic mechanical with motor drive: Used for generating crazy-high voltages at miniscule current.

[David Hess]

The common alternative is an inverter where the transformer is not used for energy storage; this has the advantage of being lower noise than a switching converter.  Big charge pumps using big capacitors and power transistors are also possible.

[Alex Eisenhut]

http://www.mmech.com/transformers

[jbb]

Note also that charge pumps can be less flexible than inductor-based designs.

Normally, a charge pump delivers some rational gain.  Typical examples are 2, -1, 3 etc. (minus some small voltage drops).  This is quite efficient.  You transfer energy from a capacitor with voltage VA into a capacitor with voltage VB, where A is a little more than B.  The average conduction power loss Pond will be related to the average current Iavg.  Pcond = Iavg * (VA-VB).

Hoever, if you really need arbitrary voltage output, it can be done (I think by adjusting duty cycles).  So VA >> VB.  This means Pcond is large, and the converter is less efficient.

If you need to operate in unusual environments (e.g. inside a large magnetic field such as MRI machine), it may be possible to use air-cored inductors.

It is also possible to create multi-level switched capacitor converters which can deliver gains in the form of N / M, where N and M are integers.  This allows for more output voltage levels with adequate efficiency, but requires more complex circuitry.