constant/limiting current simple circuit or device

[marcusone]

I have read a number of posts regarding limiting current; but I am still at a loss of how to go about getting the correct components; assuming I found a circuit that is what I need:

Some seem to show a two transistor circuit:
http://www.radio-electronics.com/info/circuits/transistor_current_limiter/transistor_current_limiter.php

However, as stated above; i'm finding it hard to determine what parts I need.

Here is what I need.

Input Voltage will be 9-12v
I want to limit the current to 2Amp max (where it would allow a constant current of 2Amp max, for about 400ms to 1second, or forever is ok as well... key is to limit the current to 2Amps).

I would welcome a ready made device as well if anyone has suggestions.  Or a circuit diagram with parts/values.

Thanks!

[uncle_bob]

Hi

Ok, you have 12V and 2 A. Your current limiting transistor will need to dissipate 24 watts. You thus need to decide about a heatsink or a heatsink plus a fan. Either way the transistor is likely to be rated above 48W without the heatsink.

Next you need to think about the accuracy of the current limiting. Is "about 2A" ok? If it starts limiting at 1.2A under some conditions and only limits at 2.4A under other conditions .. is that ok? What are the conditions you will use it under?

Is this a one off circuit, or are you going to make these in volume? For a one off, a constant current power supply is probably your best bet.

Lots of questions.

Bob

[marcusone]

Hi

Ok, you have 12V and 2 A. Your current limiting transistor will need to dissipate 24 watts. You thus need to decide about a heatsink or a heatsink plus a fan. Either way the transistor is likely to be rated above 48W without the heatsink.

Next you need to think about the accuracy of the current limiting. Is "about 2A" ok? If it starts limiting at 1.2A under some conditions and only limits at 2.4A under other conditions .. is that ok? What are the conditions you will use it under?

Is this a one off circuit, or are you going to make these in volume? For a one off, a constant current power supply is probably your best bet.

Lots of questions.

Bob

Thanks... about 2Amps is fine, but should always limit at about that level.
Power source is likely a LiPo pack or LedAcid battery; why I want to ensure I limit the amps.

It is a one off circuit (why if I can buy something "off self" that can be powered by battery I would like that).

I will be making repeated "short-circuit" like events, that will NOT last longer than 400 milli seconds, but likely will be much shorter in duration, and will NOT happen more than once per second (likely longer, as I have a 1-2 second rest period between events that should be enforced).

[IconicPCB]

An LM317 regulator connected in constant current configuration will do the trick. total cost a couple of dollars.

And to get You on the way http://www.bristolwatch.com/ccs/LM317.htm

[uncle_bob]

Hi

Ok, you have 12V and 2 A. Your current limiting transistor will need to dissipate 24 watts. You thus need to decide about a heatsink or a heatsink plus a fan. Either way the transistor is likely to be rated above 48W without the heatsink.

Next you need to think about the accuracy of the current limiting. Is "about 2A" ok? If it starts limiting at 1.2A under some conditions and only limits at 2.4A under other conditions .. is that ok? What are the conditions you will use it under?

Is this a one off circuit, or are you going to make these in volume? For a one off, a constant current power supply is probably your best bet.

Lots of questions.

Bob

Thanks... about 2Amps is fine, but should always limit at about that level.
Power source is likely a LiPo pack or LedAcid battery; why I want to ensure I limit the amps.

It is a one off circuit (why if I can buy something "off self" that can be powered by battery I would like that).

I will be making repeated "short-circuit" like events, that will NOT last longer than 400 milli seconds, but likely will be much shorter in duration, and will NOT happen more than once per second (likely longer, as I have a 1-2 second rest period between events that should be enforced).

Hi

Well, to be more specific, if the first short limits at 2.4 amps and 10 seconds later you are limiting at 1.2 amps .. is that ok? If the voltage starts to drop as it goes into limiting, how much drop is "ok" before you are in trouble?

Bob

[marcusone]

An LM317 regulator connected in constant current configuration will do the trick. total cost a couple of dollars.

And to get You on the way http://www.bristolwatch.com/ccs/LM317.htm

Doesn't the LM317 max out at 1.5A?  Can I just use any adjustable regulator in this configuration?

[marcusone]

Hi

Well, to be more specific, if the first short limits at 2.4 amps and 10 seconds later you are limiting at 1.2 amps .. is that ok? If the voltage starts to drop as it goes into limiting, how much drop is "ok" before you are in trouble?

Bob

Likely doesn't matter... As I said, it will never be left in limiting state for more than 400ms... I need the current more than care about the voltage.

[uncle_bob]

Hi

Well, to be more specific, if the first short limits at 2.4 amps and 10 seconds later you are limiting at 1.2 amps .. is that ok? If the voltage starts to drop as it goes into limiting, how much drop is "ok" before you are in trouble?

Bob

Likely doesn't matter... As I said, it will never be left in limiting state for more than 400ms... I need the current more than care about the voltage.

Hi

Ok, well if a 2:1 drift in the current limit is ok and you can "use up" about 4 V of the available supply, there are a wide range of options. Effectively you would have:

Input at no load 9V
Output voltage at no load ( 100 ma) 5V

Current limit on first short 2.4A
Current limit on second short 2.1A (2 seconds later)
Current limit on third short 1.8A (total of 4 seconds later).
Current limit on fourth short 2.4A (20 minutes later).

If instead you need:

Input voltage 9V
Output at 1.9A 8.9V

Current limit on first short 2.1A
Current limit on Nth short 2.0 A

That is a very different circuit. It also is a much more common requirement.

How fast do you need the limiter to act?

(It might be quicker if you described a bit of what your entire design looks like ..)

Bob

[marcusone]

Input voltage 9V
Output at 1.9A 8.9V

Current limit on first short 2.1A
Current limit on Nth short 2.0 A

That is a very different circuit. It also is a much more common requirement.

How fast do you need the limiter to act?

(It might be quicker if you described a bit of what your entire design looks like ..)

Bob

This sounds more like what would work.

My circuit is this:

1. Battery powered device.
2. Event occurs (I short out a wire to produce heat, start a motor, charge another battery, whatever)
3. I don't want power to shut-down on short; but don't want the battery to get damaged, and therefore want to limit the power to 2Amps (or whatever, but 2Amps should work).

Thanks for the help!

[uncle_bob]

Input voltage 9V
Output at 1.9A 8.9V

Current limit on first short 2.1A
Current limit on Nth short 2.0 A

That is a very different circuit. It also is a much more common requirement.

How fast do you need the limiter to act?

(It might be quicker if you described a bit of what your entire design looks like ..)

Bob

This sounds more like what would work.

My circuit is this:

1. Battery powered device.
2. Event occurs (I short out a wire to produce heat, start a motor, charge another battery, whatever)
3. I don't want power to shut-down on short; but don't want the battery to get damaged, and therefore want to limit the power to 2Amps (or whatever, but 2Amps should work).

Thanks for the help!

Hi

Ok, so you don't have a need for constant current?

Put another way: A circuit that simply disconnects things would do the job?

A fold back current limit is one example of a protection circuit that does not go into constant current.

Bob

[marcusone]

Bob; I do need constant current; because I do NOT want the power to shut off during a "short" (i.e. high current draw), I just want it to limit to a max of 2Amps.

[IanB]

My circuit is this:

1. Battery powered device.
2. Event occurs (I short out a wire to produce heat, start a motor, charge another battery, whatever)
3. I don't want power to shut-down on short; but don't want the battery to get damaged, and therefore want to limit the power to 2Amps (or whatever, but 2Amps should work).

Thanks for the help!

The simplest possible approach to this is to wire an incandescent lamp in series with the battery. Let's suppose you use a 12 V 24 W lamp. If you short it across a 12 V supply the current will be 2 A, giving a filament resistance of 6 ohms. However, if you only try to draw a little current the filament will be cold and the filament resistance will be about 0.5 ohms giving little voltage drop. So you have a very crude current regulator.

Depending on your needs you can tune the size of the lamp to adjust the current limit. Automotive bulbs are good for this application, or perhaps halogen lamps.

[uncle_bob]

Bob; I do need constant current; because I do NOT want the power to shut off during a "short" (i.e. high current draw), I just want it to limit to a max of 2Amps.

Hi

No matter how you do it, you run out of power when you hit 2A. The voltage quickly drops to zero so the current x volts goes to zero. A fold back simply drops current as the voltage goes down.

Probably your best bet is a high side monitor IC with a good shunt attached to it. Let that drive a control loop to stabilize your current limiting device.  You would need a control loop to keep the thing from bouncing up and down with load. All of the packaged parts out there do some sort of fold-back / timeout approach.

===edit:

Starting with something like this:

http://www.linear.com/solutions/5517

As the high side monitor gives you a linear signal to detect the current.

Then an op amp or two to do the control loop with threshold and a MOSFET to pass the current. The trick would be getting it to act fast enough in a dead short to protect the MOSFET.  Something like:

http://www.onsemi.com/pub_link/Collateral/BBS3002-D.PDF

Should be pretty tough to kill.


Bob

[marcusone]

Thanks Bob!... although that seems to be getting more complicated than I hoped for.

is there a simpler way to do it, like the transistor circuit I linked (I just don't understand how to do the math for the parts I need so that I get the voltage and amps I want).

Also, found a couple spots that mention you could do it via PWM, which sounds like it would work, but I don't have a clue where to look for a solution like that.

[IanB]

is there a simpler way to do it?

Well, gosh, I dunno...

[marcusone]

is there a simpler way to do it?

Well, gosh, I dunno...

Sorry, I saw your post; however, not really practical to place a bulb in what I would rather have as a small portable solution.  Also, I tested it out a bit to see what kind of results I would get, and it appears I will need to do some playing around to see if I can find the right bulb.

[IconicPCB]

Yes thereis a simple way.. use the LM317 approach it is close enough to 2 Amps as per your specification.. or are we now chasing goal posts?

[marcusone]

Yes thereis a simple way.. use the LM317 approach it is close enough to 2 Amps as per your specification.. or are we now chasing goal posts?

I'll certainly give this a try; I did ask if any adjustable power supply can be used this way? as I have a couple that are up to 5A I think lying around... have to order in an LM317.

[uncle_bob]

Yes thereis a simple way.. use the LM317 approach it is close enough to 2 Amps as per your specification.. or are we now chasing goal posts?

Hi

While we were going back and forth the voltage drop and limit issues on the 317 pretty much got ruled out ...

Bob