On Sat, 14 Oct 2006 19:14:19 +0100 Palindr?me wrote: | snipped-for-privacy@ipal.net wrote: |> On Sat, 14 Oct 2006 16:12:53 +0800 Rheilly Phoull wrote: |> |> | What the average user does is to monitor the battery voltage and disconnect |> | the load at a preset level. I.e say 11v. Then if your use is light the |> | battery charge will last longer than if the draw is high. Rule of thumb for |> | long battery life would be never to discharge it below 80%. There are kits |> | and ready made devices available to do just that. The state of charge is |> | reflected by the voltage at the terminals which is what these devices do. If |> | battery life is not a priority then deep cycle batteries can be run down to |> | 10.5v to provide maximum usage at the cost of replacing it more frequently. |> |> I'm asking the question at a design level. I want to be able to draw a |> specific amount of current. How many batteries to I need to parallel? |> |> Suppose I want to make a battery power DC welder using several batteries, |> each having its own current limiting resistor, with enough of these wired |> in parallel to supply sufficient current. I want to make the system so |> robust it cannot be destroyed by a solid fault current (except maybe at |> the point where the fault exists). I can figure the rating I need for the |> resistor (bank) once I know the current I'll be expecting from each battery. |> But I don't want the battery to overheat from too high a current. So I |> need to know what is the safe maximum current from each battery to choose |> the resistors (maybe a light bulb bank). |> |> Suppose the answer for some battery is 60 amps. I could put in 12 60 watt |> bulbs 12 volt in parallel as the current limit for each battery. Then if |> I want a 300 amp capacity, wire 5 of these setups in parallel. |> | | At the design level, I would want to choose batteries that could meet | the load without the need to be paralleled.
I would up to a point. I'd rather limit the size of the individual ones to the sizes I can lift. I don't want to have to buy a forklift.
| If they did have to be, I would expect that identical batteries, from | the same manufacturer and having had the same history, would share the | load equally enough not to need any individual current equalisation | measure. But would ask the manufacturer - they are the experts on their | batteries - especially for unusual discharge curves.
I would expect the ones that are weaker to have a lower voltage and this push fewer amps through the current limiting resistance. That seems like it would not be a problem.
| If I did have to parallel where I needed to be able to control load | share (say between dissimilar batteries or supplies), I would use active | switches to vary the average current from each supply - using an | inductor to limit the peak current.
I will assume basically identical batteries, all bought new at the same time.
| At low currents, resistors can be used to effectively turn constant | voltage sources into constant current ones, which can then be summed by | paralleling them, but this tends to be extremely power-inefficient. | Series-connecting constant voltage sources automatically equalises the | current drawn from each battery - which is why series-connecting large | AH cells is better than parallel connecting large voltage batteries..
But I want the higher current, not the voltage.
I have thought about what it might take to make current limitation work on DC. Obviously the method most often employed with AC, inductors, is not going to work with DC. I have thought about solid state pulse width switching feeding a capacitor bank is one option. But I'm still faced with potentially high current issues discharging the capacitors if there is a solid short circuit, if they were at full charge. I suspect using smaller capacitance and higher switching frequency is the answer to that, as the short circuit current would then be very brief as the capacitors would be discharged very quickly. And I would need to know things about the capacitors that is not regular knowledge, like their long term and short term safe current rates (charging and discharging).
Have a ever tried to short circuit discharge a big power supply filtering capacitor (instead of a resistor)?