Current limiting, the finer points
LoopyByteloose
Posts: 12,537
I have a note in my computer to run down options for current limiting.
Using an LM317 seems just about the easiest way to get what one wants... until you look at how much of a voltage drop you need and how much power goes up in heat.
And then there is a way to do this with one conventional transistor, but I suspect the voltage drop is pretty much the same. And for power operations, one might even be tempted to use a darlington pair.
So low voltage and low current, it seems that the resistor works best... especially when input voltage is actually regulated.
But I can't help wondering about zeners and Mosfets and other possible solutions. One size or solution certainly doesn't solve all, but there have to be some ways to optimize.
Why is current regulation important? Well, there is a whole world of battery charging and these days people are trying to squeeze the most out of solar panels as charger. It seems that about the best I can do with a 18volt solar panel (that actually delivers adequate current, say 450ma at 15 volts) is to charge a battery pack of 11 volts or so.
So there is a big slide due to regulation. Of course, this is a variable voltage supply and that makes conclusions a bit more dodgy. (Plus with a charge rate of 10/C and having to charge to 150% of capacity to get a full charge .. you likely need two sunny days in row to get a full charge. But that only becomes an issue after to regulate the current).
Using an LM317 seems just about the easiest way to get what one wants... until you look at how much of a voltage drop you need and how much power goes up in heat.
And then there is a way to do this with one conventional transistor, but I suspect the voltage drop is pretty much the same. And for power operations, one might even be tempted to use a darlington pair.
So low voltage and low current, it seems that the resistor works best... especially when input voltage is actually regulated.
But I can't help wondering about zeners and Mosfets and other possible solutions. One size or solution certainly doesn't solve all, but there have to be some ways to optimize.
Why is current regulation important? Well, there is a whole world of battery charging and these days people are trying to squeeze the most out of solar panels as charger. It seems that about the best I can do with a 18volt solar panel (that actually delivers adequate current, say 450ma at 15 volts) is to charge a battery pack of 11 volts or so.
So there is a big slide due to regulation. Of course, this is a variable voltage supply and that makes conclusions a bit more dodgy. (Plus with a charge rate of 10/C and having to charge to 150% of capacity to get a full charge .. you likely need two sunny days in row to get a full charge. But that only becomes an issue after to regulate the current).
Comments
Then move on to voltage drops and the associated waste of power.
Of course if you are dropping voltage (from battery or solar panel or whatever) down to some regulated level then the power lost is the same if you use a voltage regulator, transistor, darlington, resistor. It all goes out as heat somewhere and power = volts times amps. Where "volts" is the voltage drop across your regulating device.
The only way to win is to use a switching regulator. These devices are either on or off so the power loss is basically zero.
Now switchers are another big topic... not one I'm very familiar with.
Still, if you are limiting current that way you are still dumping a lot of power as heat.
That might be of concern driving LEDs from batteries, say, but when limiting charge rate to a battery from a solar panel it may not matter, after all where else are you going to put that lost energy?
Bean
Driving LEDs generally uses a resistor as a 'current regulator' regardless of battery or other source as the current is small and they cannot easily be regulated in parallel
There is a big world of current regulation out their -- battery charging, motor control, constant current sources for audio amps and instrument amps.
@Bean
I am not very clear what you mean by a shunt regulator. I found balancing charge demand with the total output of solar panel was a great way to deal with current limiting and using a high watt zener to make sure there wasn't an over-voltage was a good solution. I wonder if that is what you mean by a shunt regulator.
http://forums.parallax.com/showthread.php/141481-Batteries-running-SX28-don-t-last-all-night/page3?p=1113760&viewfull=1#post1113760
Battery charging from solar cells is pretty much a topic by itself since it involves converting the maximum power from a source with varying voltage and current to charge a device that requires varying current and voltage for best results.
Definitely no “one size fits all” for this.
Current limiting and constant current sources are not the small thing and serve different purposes as different tools.
And solar charging adds quite a bit of engineering to mere battery charging.
But battery charging is using Constant current and Constant voltage modes in their jargon, maybe these should be peak current and peak voltage.
The battery charging regimen is dependent on battery chemistry and physical construction so varies quite a bit between the various battery types. A lot of batteries have (or should have) a period of constant current and one of constant voltage during their charging period so some of the jargon really is valid.
Charging batteries from solar cells involves two distinct pieces of engineering. The first is to obtain the maximum power from the solar cells under varying levels of sunlight. The second is how to apply that power to charge the battery as efficiently as possible and at the same time avoid damaging it. To handle multiple battery types and chemistry along with getting maximum energy from the solar cells would need a very smart charger.