Sanity check please: current limiter using LED

Thread Starter

Robartes

Joined Oct 1, 2014
57
Hi,

as I'm just starting out in electronics and I had an old CPU power supply available, I thought to create everyone's favourite first project: an adjustable lab supply. As part of the lab supply, I want to build in a current limiter circuit. I came up with the below:

Current_limiter.jpg

I've tested this on a breadboard and it seems to work - with the value of R1 as stated (2.2 ohm, actually 3*6.8 ohm 1/4W in parallel) the current is limited to around 550mA when I short circuit the load. Of course, the transistor and the resistors get hot (although the power through them is still within spec for both) when they need to pass this current.

I know the classic version of this circuit uses two diodes instead of the single LED, but when I use the LED I get a nice indicator of the circuit pulling too much current.

Now to my question - does anyone see any glaring mistakes or caveats in this circuit? Thanks!
 

Thread Starter

Robartes

Joined Oct 1, 2014
57
How about the Vcc, Load, Led type V/I?
Vcc will be +12V maximum, the load can be anything (basically what's connected to the lab supply, so abstract it as say a 10K resistor). For the LED I have no detailed V/I curves -- it's a standard (for what that's worth) red LED which in my calculations I assumed drops 1.8V.

Edited to add: to measure is to know, as we say (trust me, it sounds snappier than that in my native language :) ) I measured the voltage drop across the LED when the transistor on, ie when the load is a short circuit - the LED drops exactly 2V in that condition, with Vcc being +12V.
 
Last edited:

Alec_t

Joined Sep 17, 2013
11,269
any glaring mistakes or caveats in this circuit?
No obvious mistake, and you're clearly aware of the power handling constraints. LEDs are commonly used as crude voltage references.
 

Alec_t

Joined Sep 17, 2013
11,269
When Q1 collector/emitter current is ~500mA the Vbe won't be as low as 0.7V. It will probably be nearer 0.9V (depending on the transistor).
 

#12

Joined Nov 30, 2010
18,210
It's a perfectly valid concept that will work as long as you don't go around exceeding maximum voltages or power limits.
Devil in the details and all that...
 
Top