555 switching voltage regulator

Thread Starter

ege124

Joined Dec 4, 2016
3
I'm trying to build a simple 5V to 3V switching voltage regulator. I thought to adjust the voltage between 2.8V and 3.5V by changing the duty cycle. Well, my main reason is to drive a LED(two 1W LEDs in parallel) and I want to adjust its brightness with a potentiometer. I know there are a lot off constant current led driver circuits that can do this but in addition, I'll pulse the led using another 555 circuit. I can just use a resistor in series with led but in that case, the power consumption of the resistor is nearly equal to LEDs'(0.8W). I need to make this circuit as efficient as possible so it's a great power loss to use a resistor there. I designed a circuit but I wasn't able to smooth the PWM signal into DC(Top left corner some low pass filter tries without using a resistor).

leddriver.PNG

Aside from this, I would like to know problems in the circuit or noob errors. Also if you have some pro tips for designing better circuits, I would love to know.

Thanks a bunch.
 

Dodgydave

Joined Jun 22, 2012
11,395
Why don't you just use the pwm output to light the led for brightness control, you only need one 555?



ne555-pwm-led-dimmer-circuit.gif
 
Last edited:

crutschow

Joined Mar 14, 2008
38,508
Why don't you just use the pwm output to light the led for brightness control, you only need one 555?

View attachment 126498
There's nothing to limit the peak LED current in that circuit other than the LED and MOSFET ON resistance, so that will likely zap the LED.
You need an inductor in series with the LED as the OP's circuit has.

Edit: Dirty pool, DD. You edited your schematic after my post.
But your modified circuit still does not give any better efficiency than a a linear current source and the op stated he wanted higher efficiency.
 
Last edited:

AlbertHall

Joined Jun 4, 2014
12,625
I think you will have problems driving those two MOSFETs with the NE555. The output doesn't get all that close to the supply rails so the available drive voltage is going to be a good bit less than 5V. One or other FET may not turn on fully or one or other of them may not turn off fully all depending on the particular gate threshold of the FETs you use as there is quite a spread in the specification. You would probably do better using the LMC555.
 

ian field

Joined Oct 27, 2012
6,536
There's nothing to limit the peak LED current in that circuit other than the LED and MOSFET ON resistance, so that will likely zap the LED.
You need an inductor in series with the LED as the OP's circuit has.
There is a 2 transistor circuit that does the job nicely. It uses an inductor and switches on a current threshold, the inductor energy is recovered during the recovery period.

Its near the back of an easy to find PDF book; something along the lines of "101 diode circuits" type thing.

The one I built survived various attempts to wreck it......................
 

Thread Starter

ege124

Joined Dec 4, 2016
3
Why don't you just use the pwm output to light the led for brightness control, you only need one 555?
There is a paper on driving LEDs at 60Hz and %5 duty cycle causes them to seem brighter. I want to test it if it's true or not. Because it'd be a great benefit if it's true as I want to make this project low power as possible.

I think you will have problems driving those two MOSFETs with the NE555. The output doesn't get all that close to the supply rails so the available drive voltage is going to be a good bit less than 5V. One or other FET may not turn on fully or one or other of them may not turn off fully all depending on the particular gate threshold of the FETs you use as there is quite a spread in the specification. You would probably do better using the LMC555.
Thanks for the advice! I was going to use LMC555 anyways but since there is the only NE555 model in LTSpice I ended up putting it there. But I didn't know that output doesn't get the same as supply in NE555. Good to learn.

There is a 2 transistor circuit that does the job nicely. It uses an inductor and switches on a current threshold, the inductor energy is recovered during the recovery period.
Can you give a schematic or a link?
 

ian field

Joined Oct 27, 2012
6,536
There is a paper on driving LEDs at 60Hz and %5 duty cycle causes them to seem brighter. I want to test it if it's true or not. Because it'd be a great benefit if it's true as I want to make this project low power as possible.


Thanks for the advice! I was going to use LMC555 anyways but since there is the only NE555 model in LTSpice I ended up putting it there. But I didn't know that output doesn't get the same as supply in NE555. Good to learn.


Can you give a schematic or a link?
Its not hard to find online - I gave enough info to search with.
 
Top