Hello all. I am having some problems picking a current limiting circuit (0-3A) for the attached 0.6-5V DC-DC Converter.
I am creating a multiple output voltage/adjustable power supply and I need to be able limit the current (short circuit protection etc).
1. Initially I thought of using a linear regulator rated at a lower current (i.e 1A) and then a series resistor to limit the current but the resistor might get too hot?..this might also help with reducing ripple from my switching reg
1. I also thought of using one of the many different transistor circuits similar to the ones here: https://en.wikipedia.org/wiki/Current_limiting
However, I am not sure which one works best in terms of precision, lower voltage drop (My dc-dc converter is capable of producing 6V maximum so I would need less than a 1V voltage drop to produce a 5V output (I would like the load to be connected to gnd and not floating/connected to the transistor's collector).
.
3. Another option is to buy something off the shelf to do the current limiting. I have seen a lot of LM317's used but that does not go down to 0.6V. If anyone can recommend me an alternative to this that would be great!

I would really appreciate any help on this like either a

 

That circuit already has current limit protection.
Why do you want to add more?

 

Talk about coincidence !! I am currently trying to do the same thing to his bigger brother - a TI PTN78060 ... to answer above .. I (we) want to set a maximum current and have it lower the voltage so as to not exceed it. ie charging batteries
The problem / difficulty with these devices, unlike a simple LM317 etc .. where you can simply use an NPN transistor to drive the Set lower is -
It is an inverse function - non-linear - and not a simple c/current source out of the adjust pin. Plus you have to break into the the V_Set potentiometer.
I really don't want to have an external current limit circuit (either?), just a waste of V and I, seeing as these are adjustable.
OP - it is doable, likely more than just 2-3 components. (Note: you can't get down to 0V).
As for the HF ripple, a series HF inductor + MLCC ceramics, 0u01 / 0u1 / 10uF works well. Check application notes.

 

That circuit already has current limit protection.
Why do you want to add more?

I would like to adjust the max current limit just like you would on a normal bench power supply. For example lets say I want 100mA max and if the circuit exceeds this, it would lower the voltage accordingly.

 

The reference voltage adjustment is 0.592V so you could likely use a small series resistor to sense current and use a rail-rail op amp to sense that voltage and lower the output voltage when the set current limit is reached.

The main limitation would be that the minimum output voltage when limiting current is 0.592V, so it couldn't limit the current for a short circuit.

If that's not acceptable then you could add the op amp with a series P-MOSFET at the input to the DC-DC converter to limit the current before it goes to the converter.
This will require a heatsink for the MOSFET.

So what's your choice?

 

The reference voltage adjustment is 0.592V so you could likely use a small series resistor to sense current and use a rail-rail op amp to sense that voltage and lower the output voltage when the set current limit is reached.

The main limitation would be that the minimum output voltage when limiting current is 0.592V, so it couldn't limit the current for a short circuit.

If that's not acceptable then you could add the op amp with a series P-MOSFET at the input to the DC-DC converter to limit the current before it goes to the converter.
This will require a heatsink for the MOSFET.

So what's your choice?

Thanks for your help.
I am struggling to visualise what you mean, could you please provide a rough sketch maybe of what the circuit might look like for both options?

 

Would it be acceptable to put a small shunt resistor (10mΩ) in series with the ground side of the load?

 

Would it be acceptable to put a small shunt resistor (10mΩ) in series with the ground side of the load?

Yes that would be okay

 

After some additional thought I believe the circuit below can give you a nearly full range current limit without any added power dissipation.
The circuit senses the voltage across the 10mΩ shunt resistor in the ground side of the load.
The Trim connection is the Trim input on the DC-DC converter.
(There is also a resistor to ground from the Trim input to determine the normal output voltage).

The load current across the 10mΩ shunt resistor is amplified by U1.
When this voltage (buffered/isolated by U2 and D1) exceeds the Trim reference voltage of 0.591V it will override that voltage, causing the DC-DC converter to reduce its voltage, thus reducing the current.
The adjust range is from ≈100mA to the current limit of the DC-DC (no added limit).

The one disadvantage of this circuit is that the output limit current is a very non-linear function of the pot position.
A logarithmic (audio taper) pot would thus be better for U3.

V1 is the input power to the DC-DC.

Note that the ground side of R_Shunt should be connected directly to the ground terminal on the DC-DC converter.

 

After some additional thought I believe the circuit below can give you a nearly full range current limit without any added power dissipation.
The circuit senses the voltage across the 10mΩ shunt resistor in the ground side of the load.
The Trim connection is the Trim input on the DC-DC converter.
(There is also a resistor to ground from the Trim input to determine the normal output voltage).

The load current across the 10mΩ shunt resistor is amplified by U1.
When this voltage (buffered/isolated by U2 and D1) exceeds the Trim reference voltage of 0.591V it will override that voltage, causing the DC-DC converter to reduce its voltage, thus reducing the current.
The adjust range is from ≈100mA to the current limit of the DC-DC (no added limit).

The one disadvantage of this circuit is that the output limit current is a very non-linear function of the pot position.
A logarithmic (audio taper) pot would thus be better for U3.

V1 is the input power to the DC-DC.

Note that the ground side of R_Shunt should be connected directly to the ground terminal on the DC-DC converter.

View attachment 158354

Thank you for this!

 

I would use LM2576.

 

If any instability or oscillations are found when using the circuit in post #9, try adding a capacitor (say 0.1μF-10μf) in parallel with pot U3.