NMOS - NPN current limiter circuit

Thread Starter

ssetiawan

Joined Nov 23, 2020
9
Hi, I am having a hard time trying to analyze the circuit below:
currentLimiter.PNG

It is said that the output (9V_currentLImited) is well 9V with a 30mA current limitation. capacitors C2, C3 and C4 are electrolytic capacitors while C1 is the non-electrolytic type. V1 is a 9V voltage source.

What I have tried to understand so far is that the NMOS (M1) source is supplying a slightly less than 9V due to the property of NMOS which then powers the Base of Q2. But, I have a hard time trying to calculate where does the 30 mA current limitation comes from?

Next question, what does the capacitors take part in this circuit?

Thanks!
 

ericgibbs

Joined Jan 29, 2010
16,364
hi ss,
OK, the 30mA limit only applies while charging the capacitor bank.
The current drawn from the charged cap bank is not limited, is this from a spark welding or etching unit.?

E
 

ronsimpson

Joined Oct 7, 2019
2,386
M1, the MOSFET cannot be turned on hard. The Gate turn on voltage must be subtracted from the 9V. I don't see how the output can be above 6V for that reason alone.
 

Thread Starter

ssetiawan

Joined Nov 23, 2020
9
hi ss,
OK, the 30mA limit only applies while charging the capacitor bank.
The current drawn from the charged cap bank is not limited, is this from a spark welding or etching unit.?

E
Hi Eric,

This is a part of a muscle stimulation unit which is capable to give out up to 60mA. So there is a DC/DC converter connected from the output (9V_currentLimited) which provides the needed current..

But, I'm still kind of confused, how do we get this 30mA? How can I calculate it based on the circuit to get 30mA as the current limitation?

thanks!
 

Papabravo

Joined Feb 24, 2006
19,266
Hi Eric,

This is a part of a muscle stimulation unit which is capable to give out up to 60mA. So there is a DC/DC converter connected from the output (9V_currentLimited) which provides the needed current..

But, I'm still kind of confused, how do we get this 30mA? How can I calculate it based on the circuit to get 30mA as the current limitation?

thanks!
The current limit is set by asking what current through the resistor will produce a voltage across the resistor that will allow the NPN transistor to turn on. This voltage, which is the same for virtually ANY silicon NPN transistor, is 0.66 Volts. So,

\( R_{limit}\;=\;\cfrac{0.66 V}{I_{limit}} \)

In practice you would pick the closest 1% value for the resistor.
 

Thread Starter

ssetiawan

Joined Nov 23, 2020
9
hi ss.
THis is your circuit powering up from zero time.
Note the 30mA limit.
E
View attachment 264259
Hi Eric,

thank you for your help with the simulation! I never knew about the "startup" spice directive before, now I know. And also with Papabravo's explanation:

The current limit is set by asking what current through the resistor will produce a voltage across the resistor that will allow the NPN transistor to turn on. This voltage, which is the same for virtually ANY silicon NPN transistor, is 0.66 Volts. So,

\( R_{limit}\;=\;\cfrac{0.66 V}{I_{limit}} \)

In practice you would pick the closest 1% value for the resistor.
I think I understand how it works now.. Thank you all for the kind explanation and time to simulate it even!

Have a great day :D

/SS
 

Thread Starter

ssetiawan

Joined Nov 23, 2020
9
M1, the MOSFET cannot be turned on hard. The Gate turn on voltage must be subtracted from the 9V. I don't see how the output can be above 6V for that reason alone.
According to the simulation, the output is around 7.8 volts. I always blindly assume that it has a 9V output.

Thank you for pointing that out!

/SS
 

ci139

Joined Jul 11, 2016
1,873
going complex (simplifed) https://tinyurl.com/ybhkzv49 allows higher charging currents
PS! -- the circuit proposed is not so practical as there are 2 ... 3 independent reference voltages and 3 amp.-s . . . a slight drift at any of these may induce a large difference at output current limit value !!!! . . . so the practical v. of it requires revising the "sensitivity" + minimizing the no. of independent references to preferably a single-source bond ref.-set. !!!
 
Last edited:

Thread Starter

ssetiawan

Joined Nov 23, 2020
9
No.
Why do you think it's in there if not needed?

M1 is what does the actual current limiting.
R1 and Q1 are what determine the value of the limit.

If you wanted to charge the capacitors to the full 9V, you could modify the circuit by using a P-MOSFET and a PNP transistor (below):

View attachment 264271
Hi, I have tried to understand your circuit and your explanations of your circuit library (here). So what I understood is, as I am using an N-MOSFET and a NPN transistor, the analysis is that the NPN will detect the voltage drop of the base-emitter Voltage when it is lower than 0.66V. Due to that, the voltage across the M1 gate-source reduces and cut off the MOSFET at the current limit point, which is 30mA in this case then..


Am I understanding it correctly?
 

crutschow

Joined Mar 14, 2008
30,774
Below is the simulation for the PNP, P-MOSFET circuit:

Note that the capacitor charges to the full 9V (green trace), not to about 7.5V as the NPN, N-MOSFET circuit does.

Also note the PNP emitter-base voltage during the charge (red trace).
For a 50k bias resistor, the emitter-base voltage at the current control point is about 600mV.

The PNP controls the P-MOSFET gate-source voltage (blue trace) to maintain the 30mA current limit (yellow trace) during the capacitor charge.

1649220801371.png
 
Last edited:
Top