Adding a delay to current limiting circuit

[Cval]

I am using a simple current limiting circuit design to limit the torque applied by a motor. The problem is; the initial start-up of the motor requires a lot of torque, but once it's rotating I would like to have a lower limit. How would I add a delay or bypass for the initial start-up?

 

[barry]

You might just put a cap across R1. That way, when you have have a large surge of current the cap will keep the Vbe of the transistor for increasing too rapidly. One problem with this method is that it basically slows down the current limiting response time, so if you got a locked rotor, for example, it would take a while for the current limiting to kick in.

 

[KerimF]

It could be done by using another PNP transistor but a resistor (R3) should be added between the actual PNP base (of U1) and the source node (of M1) first.

The emitter of the added PNP transistor (say U2) is connected to the U1 emitter.
The U2 collector is connected to the U1 base.
The U2 base is connected to 2 resistors (Rb1 and Rb2).
Rb1 is between the U2 base and the U2 emitter.
Rb2 is between the U2 base and the positive terminal of an electrolytic capacitor (Cb).
The negative terminal of Cb is connected to pin 2 (as R2 in the circuit).
Finally a diode (D1) may be needed to discharge rather fast the capacitor Cb when the board is off.
D1 could be connected between the positive terminal of Cp (D1 anode) and the U2 emitter (D1 cathode).

At the start, Cb looks as a short so Rb2 turns on U2. Vbe of U1 becomes close to zero (Vce saturation of U2) and the limiting circuit is disabled.
When Cb is fully charged, U2 is off and the limiting circuit returns to its normal function.

As you know, the values of the added components depend on the values that define the actual circuit and of the required boot time delay.

Hope this helps.

 

[E-design]

You can modify your circuit as below. You can change R3 for the delay you want.

 

[KerimF]

Hi E-design, I wonder if you also looked at the trace of Vbe.

 

[E-design]

Yes you are correct, Vbe breakdown problem must be prevented by adding a resistor and diode.

 

[KerimF]

You are right, E-design.
And I noticed that perhaps when the load is a motor, it is not necessary to discharge the capacitor faster than it is charged since, I suppose, the motor won't stop quickly when its currentt is cut.

Note:
I would like pointing out that by adding another PNP transistor, the delay capacitor would be much smaller for the same delay time.
But if the time delay is relatively short, the size of the capacitor is already small in your configuration.

 

[E-design]

The best is for the member to experiment with the circuit values and make adjustments to suit the application. Simulations only give us an idea what to expect. It can be quite different in a real application when other factors come into play.

 

[kam1787]

any danger of back/reverse EMF damagaing the circuit?

 

[Cval]

Thanks for the help. I tried it and it works well except that R3 is affecting the current limit. Also, why would R3's value affect the delay, and how exactly is this model causing a delay?

 

[KerimF]

The Vbe of T1 is reversed biased while C1 is charging till the emitter voltage becomes close to VS1.
R3 and R5 decrease the transient voltage gain V_r1/V_r2, they let the gain be more linear when the load current is at its limit.

Note:
For instance, to avoid affecting the original circuit, I suggested the solution on post #3.

 

[E-design]

What is the value of R2 in your original circuit and how much current limit do you want?