Thanks,

This mosfet is ok too for the 12v supply ?

 

.............This mosfet is ok too for the 12v supply ?

Yes.
Its maximum drain-source and gate voltage is 30v so it will work fine on 12V.
In general a logic-level type MOSFET will also work fine at a higher Vgs (usually will have a slightly lower ON resistance).

Note that for proper operation at lower voltages you should change R3 to 5kΩ.
Here is a simulation of the circuit operating at both 5V (green) and 25V (yellow) so it should also work at any voltage in between.

 

Thanks alot for your help !

I'm going to buy the parts (will take a few weeks to arrive).

 

Let us know how it worked.

 

Hello,

I am back, the mosfets arrived at last.

So i tested but when i create a short-circuit the voltage goes down to zero but dont stay (there is no big current flow at the input so that's good)

Maybe i've didnt connected correctly.

Here is the switch i use for test :

Here is the schematic, no idea if i did it right :

Is it the collector of the 2N3906 transistor who goes to the input + ?

Thanks !

 

.............
So i tested but when i create a short-circuit the voltage goes down to zero but dont stay (there is no big current flow at the input so that's good)

Maybe i've didnt connected correctly.

Here is the schematic, no idea if i did it right :

Is it the collector of the 2N3906 transistor who goes to the input + ?

Thanks !

I don't understand the problem you have.
Do you mean it shuts off the current but then resets itself?

No, Q1's emitter goes to the plus input, as shown.

 

Thanks for your reply

No, Q1's emitter goes to the plus input, as shown.

It doesn't change anything (i even change the transistor to be sure)

I use 12V for tests

When i make a short-circuit at the output the voltage goes down and then i take off, the voltage raise up without pressing the switch, like a normal short-circuit, the button is useless (but no sparks because it's looks like the mosfet is stopping the short so no high current flows at the input)

I changed the R1 resistor but it doesn't change anything (064V / R1).

I have about 11V at the switch pin

(didn't had the capacitor but i won't think it would be that)

Thanks

 

Yes, that's an error on my part.
The circuit will go to a very low current when it trips but it does automatically reset when you remove the short.
You don't need the reset switch.

It does have the advantage over a simple current-limit circuit in that it doesn't dissipate any significant power when it reaches the limit current.

 

How about a resettable fuse ?

I have a amplifier test bed.
I use a 2 amp transformer but don't use any fuses.
I know there is a short if the transformer rumbles on power up.
The 2 amps is sufficient to run the amplifier to low volume but isn't too strong that it fries output transistors.
Transformers are hardy and current limiting anyway so it might not be a problem.

 

Yes, that's an error on my part.
The circuit will go to a very low current when it trips but it does automatically reset when you remove the short.
You don't need the reset switch.

It does have the advantage over a simple current-limit circuit in that it doesn't dissipate any significant power when it reaches the limit current.

Ah ok, nevermind but where the problem does come from then ? Looks like the 2N3906 is useless then

Anyway it's much easier.

Wouldn't it be a problem to have loads like LEDs, battery charging module, 12v car stuff etc...

You've said it has the advantage of a simple current-limit circuit so can it drive led's ?

Does the mosfet will get warn for 12V 1.8A ? does it need an heatsink ? Because i like to fit the mosfet into my 12V supply box

Thanks alot for your help

 

Ah ok, nevermind but where the problem does come from then ? Looks like the 2N3906 is useless then.............
Wouldn't it be a problem to have loads like LEDs, battery charging module, 12v car stuff etc...

You've said it has the advantage of a simple current-limit circuit so can it drive led's ?

Does the mosfet will get warn for 12V 1.8A ? does it need an heatsink ? Because i like to fit the mosfet into my 12V supply box

Why do you think you don't need the 2N3906?
That's what shuts off the MOSFET when the limit is reached.
Without it, there is no current limiting.
The problem is in my understanding of the design. I missed the fact that it would reset when you removed the load. Otherwise it still works like a fuse.

I don't understand your concern about loads like LEDs, etc.?

I said this is unlike a current-limit circuit which just maintains the current.
This circuit causes the current to go to a very low value when the limit is reached.

The MTP50P03HDLG MOSFET you selected has a 25 milliohm ON resistance so will dissipate only 81 millwatts at 1.8A. It will barely get warm.

 

Why do you think you don't need the 2N3906?

Because it worked the other way around, i will try tomorow (late in France)

I said this is unlike a current-limit circuit which just maintains the current.
This circuit causes the current to go to a very low value when the limit is reached.

Ah ok

Thanks

 

Because it worked the other way around, i will try tomorow (late in France)
...........

It may work the other way around because a BJT transistor will still work with the emitter and collector interchanged (albeit with much lower current gain).

 

Hello,

So i've made a test (on a breadboard)

But i have one problem :

I calculate R1 for 1A (0.64/0.6=1.06A) so 0.6ohm

I chose a 5v input, i had a USB port at the output with a USB voltmeter and a USB load resistor

1A load of the USB load resistor is actually 0.9A and 2A about 1.8A

USB voltmeter :

USB load resistor :

When i power everything and add the load the usb voltmeter goes off but the USB load is on (we can see the led on but it's very dim)

The USB load at 1A consumes 0.56A instead of 0.9A (that's why the led is dim)

If i chose 2A on the usb load it consumes 0A because i've set the cut-off to 1.06A

So what is the problem ?

PS : I've found how to use the switch I had to change the R3 5K to 1K.

Thanks

 

Can it be the voltage loss trough the resistors or it's because i'm using a breadboad ?

 

You need to post the exact circuit configuration you are using.

 

You need to post the exact circuit configuration you are using.

What do you mean ?

At the input i use a usb charger power supply 2.1A

 

What do you mean ?

At the input i use a usb charger power supply 2.1A

I mean I need to see exactly all the connections you are using that are giving you a problem.
I can't tell from your description how everything is connected.

 

I found the problem

I was using a long USB cable and that was limiting the voltage

I will try tomorrow with a shorter USB cable

 

You don't necessarily need an op amp for the fuse.
Below is the LTspice simulation of a two transistor electronic fuse (one BJT and one MOSFET).
The simulated load is varied from 20Ω to zero and back to 20Ω to test the current limit.
The fuse triggers when the current increases to about 1.8A and the current then stays at zero until the reset switch momentarily closes at 30ms.
The P-MOSFET can be just about any that has a 20V or greater rating and an ON-resistance of 0.1Ω or less.
The PNP can be any small signal BJT with at least a 20V Vce and 100mA collector current rating.

If you have trouble with nuisance tripping on short load transient currents, you can add a small capacitor from the MOSFET gate (G) to ground. You then should also add a 10Ω resistor in series with the reset switch to limit the surge current through the switch.

The main disadvantage of this simple circuit is that it has a 0.35Ω resistor in series with the output and this will cause some voltage drop at higher currents.
If that is too much voltage drop for you then a more complex circuit with a op amp or comparator is needed.
Alternately if you can put the fuse circuit in front of the power supply regulator then the voltage drop shouldn't be a problem.

View attachment 103911

Not a latching circuit. On over current the MOSFET is supposed to turn off, which brings current back into normal limits, which brings back the overcurrent condition ... a saw tooth generator. It may limit the current to a specific level, but it does not remove power until a button is pushed.
A flip flop is suggested that sets on overcurrent, turns off the MOSFET and keeps it off until the overcurrent condition is corrected. Then the switch can be pushed to restore power/ An electronic fuse as opposed to a current limiter.
LM358 may not be the best choice is the LM358 is powered by the line it is monitoring. Maybe some op amp with rail-to-rail input capability.