Society of Robots - Robot Forum

Electronics => Electronics => Topic started by: sherby on July 10, 2013, 05:15:35 PM

Title: Power supply
Post by: sherby on July 10, 2013, 05:15:35 PM
I had another question regarding a power supply.
If i have a 9V battery, then there is a limit to the maximum current o/p it can give.

Q1.Is there a way to increase the current,anyhow?

Q2. Also, if i have a 16 V Lithium Polymer battery, with high current rating, say 5A, and i want to limit it to 2A, so that maximum current that can flow in the circuit is 2A, without breaking it, then how can i do so?
Do you think, a resistor in parallel to the load device, do the trick, so that the current splits there, mantaining the same voltage?
I am aware that the current depends on the device which draws it, but still is this possible, incase i do not want it to draw too much current.

Thanks for any help.
Title: Re: Power supply
Post by: waltr on July 10, 2013, 07:28:17 PM
A resistor parallel to the load increases the current draw. Ohm's Law and parallel resistor equations tell you this.
Rparallel = (R1 * R2) / (R1 + R2)
Ohm's Law: E = I * R or I = E/R. decrease the value of R and I increases.

To limit current put a resistor in series with the power/load. Rs in series add so in the above equation when R in larger I is smaller. I real circuits this may not work since there is a Voltage drop across the series resistor (E = I*R) which will decrease the Voltage at the load.

There are active circuits to limit current but are not trivial to explain. Google "current limiting circuit" to learn more.

As for the 9V battery, it maximum current output is limited by its internal series resistance. Put a Voltmeter on the battery and then different value resistors. You'll see that the measured Voltage is lower with lower value resistors.
Title: Re: Power supply
Post by: jwatte on July 10, 2013, 09:37:22 PM
You can increase the current capacity of a 9V battery by adding several more 9V batteries in parallel. It won't really be economical to drive anything strong this way, though -- you'd need A LOT of 9V batteries.

Current isn't "pushed" by batteries. Current is "drawn" by a circuit (such as a device.) The circuit draws current based on its equivalent resistance, and the voltage applied. So, if a circuit has 100 Ohm resistance, and you apply 16 V, it will draw I = U / R = 0.16 Amps, or 160 milli-amps. Something in parallel with that circuit will not actually change how much current that device draws -- it will just serve to draw more current from the battery.

If a device is rated for 9V, and you only have 16V available, then what you need is a voltage regulator, that regulates the voltage to 9V into the circuit. If the circuit is then designed correctly, it won't draw more current than rated. Just make sure the regulator itself can deliver sufficient current. Most linear regulators (78xx series, 1117 series, LF series, etc) top out below the 2 A you suggest. Also, linear regulators generally need lots of heat sinking, because they dissipate the difference between supply voltage and draw voltage as heat.

The best way is to find a switching power converter that can take your input voltage (range) and output the voltage you need, at the number of amps you expect to draw at max. In hobby circles, this is known as an "UBEC" and in electronics design, this is known as a DC-DC converter.
Title: Re: Power supply
Post by: johnwarfin on July 11, 2013, 08:09:07 AM
Q1.Is there a way to increase the current,anyhow?

a dc-dc converter will allow higher current draw but voltage will be lower. for example triple current at 3v or about double current at 5v. a large cap in parallel will also allow increased current but only for brief periods.

Q2. Also, if i have a 16 V Lithium Polymer battery, with high current rating, say 5A, and i want to limit it to 2A, so that maximum current that can flow in the circuit is 2A, without breaking it, then how can i do so?
Do you think, a resistor in parallel to the load device, do the trick

no. nor is a voltage regulator the correct method. best way is with a CURRENT regulator. it is possible to use a vr ic to do this cheaply and reliably with a couple additional components. googling scott henion might do it.
Title: Re: Power supply
Post by: jwatte on July 11, 2013, 10:10:38 AM
It's true that a current regulator will limit current to a particular value. If that is, actually, what you want, then the current regulator will increase and decrease the voltage to try to keep the current at 2A. You can easily build a current regulator from a linear regulator such as the LM350 by tying reference and output together with a current sense resistor. Again, the warning about heat sinking the regulator (and the resistor!) applies.

TO make a LM350 (1.25V reference, max 3.5A rating) regulate to 2A, tie your input voltage to input, tie a (1.25 / 2 ==) 0.625 Ohm resistor from output to reference. Then tie the reference to your actual circuit. The regulator will attempt to adjust the voltage such that 2 Amps flow through the circuit. The drop-out will be approximately 3.25V (2V from the regulator, 1.25V from the reference.) The resistor will need to be rated for at least (2*2*0.625 ==) 2.5W.

If what you want is a simple safety mechanism, a 2A fuse seems like an easier option :-) And if what you want is to adjust a circuit so it draws a reasonable amount of current, you typically do that by designing the voltage of the supply to the appropriate level for the circuit. This is why voltage regulators are much more common than current regulators, because that's what you "typically" actually want except for certain special cases.

If you share what you actually are trying to accomplish and why, you may get more specific advice.
Title: Re: Power supply
Post by: johnwarfin on July 11, 2013, 12:43:27 PM
op described the need to limit current "without breaking it" which sound like he just wants it to shut down on overcurrent then recover. contrary to popular belief a 2amp fuse does not limit current. ive measured 10 amps briefly through a 1 amp fuse before it blows. probably a lot more in some cases. not even close to fast enough to protect solid state devices. and of course constant replacement can get costly if limit is exceeded often.

the version i described will actually do the job better. for really low cost a simple transistor and resistor also limit current quite well but not as stable as vr type over wide temperature swings.