I tend to design digital circuits and need to limit the current to a level that I consider is the Maximum current for the circuit. If the circuit needs only 200mA, then the supply will deliver many amps if the design is incorrect, this will let out the magic smoke.

This is why I would like a current limiter. I hope someone out there will have a clue or even better a schematic circuit for such a device.

Thank you all in advance.

 

Try this:

 

Try this:

View attachment 353211

Won't that try to force a fixed current into the load by changing the output voltage? That kind of defeats the regulated output voltage feature of the supply. Depending on the actual load it may or may not be possible.

 

Won't that try to force a fixed current into the load by changing the output voltage? That kind of defeats the regulated output voltage feature of the supply. Depending on the actual load it may or may not be possible.

You adjust Vin for the desired voltage (measured at the final output) with a constant voltage regulator.
Then the output voltage of the current limiter cannot exceed Vin. The output voltage can only go down, not up.

 

You adjust Vin for the desired voltage (measured at the final output) with a constant voltage regulator.
Then the output voltage of the current limiter cannot exceed Vin. The output voltage can only go down, not up.

So you only get Vin at one constant value of the load current. If the load changes you get the same current but at a lower voltage. So, what is the purpose of having a regulated voltage if the load is NOT constant in normal operation?

 

get yourself a DROK Buck Boost converter from Amazon https://www.amazon.com/dp/B089ND7PTH?ref_=ppx_hzod_title_dt_b_fed_asin_title_0_0 Output .6 to 30 volts current limit to 5 amps , columb counter,

 

So you only get Vin at one constant value of the load current. If the load changes you get the same current but at a lower voltage. So, what is the purpose of having a regulated voltage if the load is NOT constant in normal operation?

Assume that the LM317 plus the sense resistor R1 has a drop voltage of VD.
Vout = Vin - VD

At low load (high RLOAD), current output is small, and VD is small.
VD only starts to increase when the current limit is reached (low RLOAD). Then Vout drops.

If the load changes you get the same current but at a lower voltage.

That means that the current limit has been reached. That is how a constant current supply works.

 

The LM317 limiter will drop a couple volts, even when it is not limiting, so I don't see that as a viable circuit at the output of the power supply.
A two transistor limiter will reduce that drop to a maximum of about 0.7V near the limit, but that still may be excessive.

Much better to have the limiter as part of the supply or at the supply input.

Rod777, What supply are you using?

 

You could put the LM317 current limiter on the input side of the voltage regulator.

 

The basic assumption that you can protect a circuit with current limiting is false.

 

The basic assumption that you can protect a circuit with current limiting is false.

Please elaborate.

 

The basic assumption that you can protect a circuit with current limiting is false.

And that belief is categorically false.

We have to assume that the circuit is already in a fault condition.
The current limit prevents catastrophic failure.

That is the what a current limiting fuse does.

 

Well yes, the LM317 would work placed before the voltage regulator, but 1.25V is a lot to lose in any power supply.
Assuming here that it's a bench type test power supply, say 0(ish) - 30Vdc.
Being as 1.2 to 30V was mentioned, it may actually be using an LM317 for the voltage control?

And I reckon current limiting is better protection than no current limiting.

 

get a bench power supply and you will thank me later. make sure to pick one with digital controls (encoder) for both voltage and current limit. and get one with 4 digit displays. note that some trip when current limit is reached, others go into constant current mode. for day to day use, i prefer the later one. not sure why they not include switch so one can choose behavior as needed.

current limiting using LM317 is a low cost solution to get that functionality from regulated supply that does not already have current limit. but... as was mentioned before, it need to be inserted before voltage regulator. and since your PSU can go to 30V and most digital circuits are using not more than 5V, couple of volts dropped on a current limiter will not be a big loss. you can add switch to bypass current limiter for times when you do need higher voltage.

limiting current does save circuits. any new board i make is first connected to a bench supply where voltage and current are gradually increased. this helps find and fix any issues before things end up in smoke.

and since this is a linear regulator, make sure to use massive heatsink.

 

Please elaborate.

Do you really think current limiting can save that sensitive output you connected directly to a rail?

Other examples are virtually unlimited.

 

And that belief is categorically false.

We have to assume that the circuit is already in a fault condition.
The current limit prevents catastrophic failure.

That is the what a current limiting fuse does.

You added a parameter not explicitly expressed in the OP.

 

Many devices reach catastrophic breaking point when thermal dissipation exceeds device specifications (think resistors, transistors, etc.), i.e. we let out the magic smoke.

Current limiting can prevent that from happening.

 

You added a parameter not explicitly expressed in the OP.

That is exactly what the OP requested.

... then the supply will deliver many amps if the design is incorrect, this will let out the magic smoke.

 

That is exactly what the OP requested.

A single de-coupling capacitor connected backwards can "let out the magic smoke"

 

A single de-coupling capacitor connected backwards can "let out the magic smoke"

True. There are some things that can be saved and some things that cannot be saved.

If the device goes into short-circuit mode, sometimes it still can be saved. for example, CMOS latch-up, because limiting the supply current might still save the device.

That is why sometimes using an incandescent lamp or poly-fuse in series can save a circuit.