Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Current limit circuit design

Status
Not open for further replies.

hithesh123

Full Member level 6
Joined
Nov 21, 2009
Messages
324
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,298
Location
lax
Activity points
3,548
I am trying to design a discrete current limit circuit. So far I have come up with 3 ckts. I can't decide which one to go with -

1. Classic 2 NPN transistor current limiter
2. PNP + NPN + op amp current limiter
3. Mosfet current limiter

Ckt 1 is simple, but the drop across BJT is high compared to other 2 ckts. The drop across the sense resistor also has to be 0.7v at Ilim. So some power is wasted here. I haven't found a good design reference for implementing foldback limiting.
The problem with (ckt3) Mosfet is the load is connected to high side. I am not sure if this is a problem, say in a battery charging ckt like a cell phone charger.
Advantage being low Rds ON and Mosfet (IRF540) can dissipate insane amount of power - 130W. Even if the ckt is shorted, no problem!
For the Mosfet ckt shown in fig, the current is controlled by the zener at the gate. Gate voltage is limited to 3.9v, which corresponds to about 1A.
I can replace the zener with an op amp and add a current sense resistor for better operation.

Ckt2 has low drop across the PNP pass transistor but component count is higher. Problem I see here is output getting shorted. No foldback limiting.
 

Attachments

  • MOSFET current limit.jpg
    MOSFET current limit.jpg
    72.2 KB · Views: 261
  • PNP current limit.jpg
    PNP current limit.jpg
    99 KB · Views: 261
  • BJT current limit.jpg
    BJT current limit.jpg
    71.7 KB · Views: 220

The order of schematics is different from your numbering.
Ckt2 (left schematic) has various bugs:
- Output transistot 2N3906 is rated for 200 mA/350 mW and can't work in the intended current range
- R11 will cause a setpoint error varying with Vbe of Q1
- Q2 is missing a base-emitter load resistor
- The control loop will be possibly unstable due to insufficient compensation compensation
But the basic circuit can be modified to work correctly

For Ckt1 (mid schematic) I agree with your analysis.
MOSFET Ckt3 (right) isn't a real constant current circuit.
 

The order of schematics is different from your numbering.
Ckt2 (left schematic) has various bugs:
- Output transistot 2N3906 is rated for 200 mA/350 mW and can't work in the intended current range
- R11 will cause a setpoint error varying with Vbe of Q1
- Q2 is missing a base-emitter load resistor
- The control loop will be possibly unstable due to insufficient compensation compensation
But the basic circuit can be modified to work correctly

For Ckt1 (mid schematic) I agree with your analysis.
MOSFET Ckt3 (right) isn't a real constant current circuit.

Yes, ckts not in order.
Regarding ckt 2, I only wanted to see if it works. I put 2N3906 bcoz I wanted to try with that transistor on bread board.
R11 is the gain resistor for the op amp. Are you saying the gain will change set point variation?
R1 is the load (resistor).
What mods do I need to make this circuit to work correctly.

For the Mosfet ckt, I wanted to use a sense resistor. When the voltage across the sense resistor hits the limit, the Mosfet gate voltage should drop to 4v. Not sure, if this is a good design.

Can I also use current mirror as a current limit circuit?
 

The current error can be removed by changing the "controller" from P to PI action. Gain and bandwidth can be set independently this way. By adding two resistors, the output stage gets an almost current independant gain.
 

Can you explain how exactly will the RC feedback and the emitter resistor will help.
How do I pick those values?
 

The RC circuit implements a PI controller, which is generally first choice for all kinds of constant I or V regulators.

The emitter resistance is a means to make the OP output voltage to output current transfer characteristic more linear and reduce the gain, although the circuit still isn't optmal in this regard.
 

Thanks FvM. Can you tell how the response will be different compared to a proportional controller?
The emitter resistor is to reduce gain? But I can reduce gain just by reducing op amp gain.
 

The difference between P and PI is in the remaining error.

I told another reason for using an emitter resistor than reducing the gain.
 

Apparently, it's not a complete cicrcuit, only an ouput stage with current limiting circuit. The source controlling the output transistor isn't shown. But I don't like to guess about unclear publications.
 

... why do you need a diode at the op amp output?

"The source controlling the output transistor isn't shown" as FvM mentioned above. When the current limiting protection circuit is in the inactive state - without the diode - the opAmp used as comparator would output nearly its full negative supply voltage (≈GND) and thus override the control source.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top