The problem:
Produce any voltage 0 to thousands of volts DC without the need for a transformer.
This rechargable power-source must be resilient so that it can still function when damaged and is able to charge despite degradation of charging circuitry.
Intended audience:
Remotely
operated military or space applications. Once deployed these systems are expected to work for many years without maintenance. Failure of power source or degredation of charging circuitry is a common cause of satelite failure.
The Idea:
Today’s batteries are built from arrays of cells. The standard AA battery inside most consumer appliances is a single cell device. The configuration of these devices (either in series or parallel) determines how much voltage they produce or how much current can flow through them. In a traditional cell array, the failure of one battery will cause every other battery in series with it to fail, causing an appliance failure.
Imagine if rather than having a small number of relatively large cells, our battery is composed of an immense number of microscopic cells. The circuits joining cells would no longer be copper wires, but switchable microscopic tracks that can be controlled by a battery control system.
The controller keeps track of the state of each of the thousands (or millions) of micro-cells. At any time the processor will be aware of what that cell is doing and what it’s charge state is likely to be. At any time a cell is going to be:
* Charging up – from an external power source.
* Discharging – in order to provide power.
* Inactive
Each microcell would be designed to work at an optimised voltage. Even though the system can create a variety of voltages, each individual microcell would be required to produce a modest voltage for a short period of time (seconds) before becoming discharged.
Supposing we needed the system to produce a particular voltage, all it would have to do is configure enough microcells in series to produce that voltage. In practice, the system will need to configure a number of these arrays in paralell to produce the required current as well. After a few seconds an array of micro batteries will loose all it’s charge, and the processor will switch on another array that it has been lining up. Keeping a constant voltage requires constant shuffling.
Partially discharged batteries might get a 2nd go by being lined up in a longer array. In this way. The overall efficiency of this system will be limited by how cleverly the control system is able to ‘shuffle’ the microcells.
One of the advantages of this system is that it would be capable of re-charging itself from any AC or DC power-supply at much lower voltage than it is capable of outputting.
If the power routing system were sophisticated enough, such a battery could withstand severe damage (e.g. a meteorite hole right through it) and keep on working – albeit at a lower level of performance. The routing system would route around the problem. As long as the control system was intact (which itself could be distributed) some function would remain.
The system would also allow variable voltage charging. This would be handy when all the power is ultimately derived from something unpredictable like a solar-array. Charging can take place at the same time as dis-charging. All the controller has to do is route charging power to the optimum number and configuration of microcells to achieve the fastest rate of charging.
Problems:
We do not yet hae the technology to manufacture microscopic cells, or indeed the routing system. Using today’s DSP technology, the controlling system would be likely to consime as much power as the battery could produce.