## current limiting

22 posts / 0 new
Author
Message

hi
i want to build current limiting circuit
i found this in Google

which one should i use?
i have 300 watt power supply to limit its current

Your first two links don't seem to work.

how many volts for your power supply? What current do you want to limit. What do you want the circuit to do when the current limit is reached?

That's not the way image linking works on this site!

See the instructions here: https://www.avrfreaks.net/wiki/em...

Then you get this:

http://www.maximintegrated.com/e...

http://cdn.instructables.com/FSQ...

http://i.stack.imgur.com/3mIg3.p...

But, as Kartman says, you haven't provided enough information to be able to be able to give any advice.

On the pages where you found those images, was there not any explanatory text?

Have you done any general study/research/reading on the topic?

i have 300 watt power supply

does it not have its own current limit?

Top Tips:

1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
Last Edited: Sat. Sep 5, 2015 - 11:27 AM

Kartman wrote:

Your first two links don't seem to work.

how many volts for your power supply? What current do you want to limit. What do you want the circuit to do when the current limit is reached?

between 0 - 30 voltage and 0 - 10 a current

when it reach the current it will decrease the voltage

The circuits you provided are for constant current. The concept will work, but these circuits will need to be scaled to handle 10A. You also need to cope with up to 300W of heat. What sort of power supply are you considering and how do you vary the voltage?

Last Edited: Sat. Sep 5, 2015 - 11:43 AM

awneil wrote:

navidrct wrote:

That's not the way image linking works on this site!

See the instructions here: https://www.avrfreaks.net/wiki/em...

Then you get this:

http://www.maximintegrated.com/e...

http://cdn.instructables.com/FSQ...

http://i.stack.imgur.com/3mIg3.p...

But, as Kartman says, you haven't provided enough information to be able to be able to give any advice.

On the pages where you found those images, was there not any explanatory text?

Have you done any general study/research/reading on the topic?

i have 300 watt power supply

does it not have its own current limit?

thanks ill do that for images.

i search . but i want to know your opinion . cause it will be safer for me to do

Kartman wrote:

The circuits you provided are for constant current. The concept will work, but these circuits will need to be scaled to handle 10A. You also need to cope with up to 300W of heat. What sort of power supply are you considering and how do you vary the voltage?

vary voltage with buck - boost regulator

i use llc for power supply

You would want the current sense to work into the regulation loop of your power supply so you don't have to cope with large losses.

Kartman wrote:

You would want the current sense to work into the regulation loop of your power supply so you don't have to cope with large losses.

then i should sense current and relative to that change PWM ?

thanks

Hello,

As some folks have already said and I'll state further - if you're trying to limit a 300w power supply there are special considerations.  Limiting a few mW is trivial, 300w requires specialized power devices (large BJT or more likely MOSFET).

Does the supply have any limiting now?  If so you'll get a time margin, if not and a short will quickly destroy the output devices, the hardware and software constraints are much tighter.

Kartman wrote:

You would want the current sense to work into the regulation loop of your power supply so you don't have to cope with large losses.

then i should sense current and relative to that change PWM ?

Yes.  I can't tell how much PWM you understand but that's the idea.  However at that point you're designing a new dc-dc power supply to go after the originating power supply, so you'll need to figure out topologies - buck, buck/boost, sepic, etc. along with filtering.  Not hard but at 300w definitely some specialized design.  The surrounding drive circuitry to adequately drive a large power mosfet is significant, you can't hook one of those bad boys up to a logic pin (to a first approximation) and get fast charge/discharge rates, they require a couple of amps of instant current before the gate is charged to zero current.  If you charge them slowly you get too much dissipation in the device.

Some power supplies have an ENABLE input or DISABLE input, if your 300w unit does you can use that for control so no additional PWM, just current monitoring.

Note that with high current monitoring the voltage differences you're trying to monitor change very little because you can't have much drop in the sense line.  As Kartman noted many dedicated modern PWM control chips in buck mode use the drop in the upper mosfet for current sensing; that requires knowing your component characteristics tightly but it involves no sense resistors.  It does require understanding the dV/dT characteristics of your inductor/mosfet combination.

There are so many good PWM control chips out there, download some data sheets and check out their topologies and reference circuits as they know how to do it well.  Most of them don't do a dead short very well despite what they claim, that's why when I go off-circuit and need short protection I add a super-low rDS(on) power mosfet in constant conduction; the micro then monitors that and shuts it off.  Plenty quick enough because the dedicated converter chip does the first tens of ms, the AVR kills it after that.  Which, in your case, may be adequate - if the existing supply can handle a short for a small time and you can then shut it off, you may be able to forgo pwm.  It will probably require a boost circuit to drive the mosfet gate several volts above the output voltage assuming N-channel enhancement mode, sometimes a depletion mode can work but at 0vGS there's often too much dissipation and you still need the boost.  P-channel enhancement can also sometimes work without boost, depending on several factors.  However this is only for on/off, not PWM, so it won't do a current limit, only a "too much, turn it off".

One more possibility in your case not using PWM - depending on your voltage and current requirements you can use an SSR to turn on/off the output without PWM.  At 300w that may be too expensive but they interface directly to a uC.  I've used them in a bunch of applications with a couple of amps or less, work great.  Don't believe the data sheets, double the current requirement.  However this is only for on/off, not PWM, so it won't do a current limit, only a "too much, turn it off".

Without knowing your requirements I'm just throwing you ideas.  I hope the above is not too negative, it's intended to suggest that you have much research to do.

Hugh

Think, for a moment, please. Here is a thought experiment.

Suppose that you have a 1 ohm load. With current limited 10A through it, it will have 10A * 1 ohm = 10V across that resistor. The current limiting device will then have (30V - 10V) = 20V across it. 10A through that limiting device will generate 200W of power. Have you ever tried to to hold on to a 200W light bulb?? VERY  HOT! Even worse if that load is a good short circuit, and the current limiter has 300W dissipated in it. EVEN HOTTER!

Now, that is what you have to deal with when you use ANY linear current limiter.

Maybe it is time to rethink your requirements?

Jim

Until Black Lives Matter, we do not have "All Lives Matter"!

Google "foldback current limiting"

Greg

thank you all

maybe i should just turn it on and of in higher power and limiting current in lower power

like limit in aspect of 1 amp or even more and disable if it exceed 5amp

what you think?

again thank you

ka7ehk wrote:

Think, for a moment, please. Here is a thought experiment.

Suppose that you have a 1 ohm load. With current limited 10A through it, it will have 10A * 1 ohm = 10V across that resistor. The current limiting device will then have (30V - 10V) = 20V across it. 10A through that limiting device will generate 200W of power. Have you ever tried to to hold on to a 200W light bulb?? VERY  HOT! Even worse if that load is a good short circuit, and the current limiter has 300W dissipated in it. EVEN HOTTER!

Now, that is what you have to deal with when you use ANY linear current limiter.

Maybe it is time to rethink your requirements?

Jim

thank you

i want to limit current with change of PWM and voltage cause i have buck boost regulator.

any idea?

Again, you don't give us much to go on. What 'buck-boost' regulator? It would help if we knew exactly what regulator you were talking about. Did you Google 'foldback current limiting'? Did it give you any ideas?

Last Edited: Sun. Sep 6, 2015 - 08:45 AM

Kartman wrote:

Again, you don't give us much to go on. What 'buck-boost' regulator? It would help if we knew exactly what regulator you were talking about. Did you Google 'foldback current limiting'? Did it give you any ideas?

my buck boost is lm5118.

yes i read and its good.

i will test it.

this foldbacl is like that i told before to sense current and change PWM to change voltage .

thank you.

of course there will be a lot issue . but i will test it . its the only way for now

You don't touch pwm - you use op-amps to alter the feedback signal to the lm5118 so it drops the output voltage when the current exceeds the set-point.

Kartman wrote:

You don't touch pwm - you use op-amps to alter the feedback signal to the lm5118 so it drops the output voltage when the current exceeds the set-point.

for changing feedback we can use PWM and RC filter to decrease Voltage.

But i think of Changing Resistor divider on feedback on low side with digital resistor.

what u think on that?

By the time you've changed the voltage with the digital resistor, your load has gone up in smoke. That's why I'd do it in the analog domain.

Kartman wrote:

By the time you've changed the voltage with the digital resistor, your load has gone up in smoke. That's why I'd do it in the analog domain.

ok.then i cant use any digital processor cause its slow.

im not expert in analog domain. any suggestion ?

thank you

navidrct wrote:

Kartman wrote:

By the time you've changed the voltage with the digital resistor, your load has gone up in smoke. That's why I'd do it in the analog domain.

ok.then i cant use any digital processor cause its slow.

I didn't say that. How fast will your system react?

My suggestion was to use op-amps. It's up to you to design it or find a suitable design.

Kartman wrote:

navidrct wrote:

Kartman wrote:

By the time you've changed the voltage with the digital resistor, your load has gone up in smoke. That's why I'd do it in the analog domain.

ok.then i cant use any digital processor cause its slow.

I didn't say that. How fast will your system react?

My suggestion was to use op-amps. It's up to you to design it or find a suitable design.

i think of that i use analog stuff for short detect and digital(current sense) for giving option to user to change current limit .