1
\$\begingroup\$

I am trying to make a bunch of LEDs fade out when they are turned off. The task is quite simple for a single LED by using a 1000uF capacitor to produce a time delay as in the following circuit: enter image description here

The other possibility is using Arduino analogWrite() function and get my work done but I don't want to use any microcontroller.

When I try to make it work for like 25 LED in parallel. The capacitance needed grows quadratically to store that much energy. Is there any other way of getting this effect to work which does not involve the usage of a lot of capacitors and neither a giant one.

Note: Please add tags to this question that are more relevant as I could not find many.

\$\endgroup\$
1
  • \$\begingroup\$ Use a smaller cap, then use a BJT as an emitter follower to power the LED's. Do be carefull with power dissipation in your transistor. I bet someone with more time will draw a circuit diagram in one of the answers. \$\endgroup\$
    – jippie
    Jun 19, 2015 at 19:15

1 Answer 1

Reset to default
2
\$\begingroup\$

If you don't mind experimenting around a little bit with the timing you can use something like that:

schematic

simulate this circuit – Schematic created using CircuitLab

R1 and C1 set the transient behaviour, R2 is only to limit the current over the switch, dont make it too large. If you don't care about losses you can also use a N-channel MOSFET instead. R1 and C1 have to be between G and GND, R2 and SW1 between D and G. The N-channel circuit gives you a better control about the fading (if you want to produce your circuit in large quantities) but you will always have a dropout voltage of Vth from D to S.

\$\endgroup\$
2
  • \$\begingroup\$ hey Chris. Thanks for your response. What would be the size of capacitor I would need to use at max if I go by your suggestion because the main issue for me is reducing the capacitor size? Secondly, do I keep on adding other LEDs in parallel to the shown? \$\endgroup\$
    – Muhammad Ali
    Jun 19, 2015 at 20:16
  • \$\begingroup\$ If you go with the PMOS, the relevant parameter for you is the transconductance of the transistor. The change of VGS times the transconductance gives you the change of drain current. VGS transient change rate is just given by the normal capacitor discharge curve VGS(t) = 12V*exp(-t/(R1*C1)). The tricky thing here is only, that the tolerance of the transconductance is very large, so you had to experiment around as I said. But yes, the capacitors value can be much smaller, as long as R1*C1 stays reasonably high. Give it a first try with 100uF and 100kOhm and play around a bit... \$\endgroup\$
    – christoph
    Jun 19, 2015 at 20:28

Not the answer you're looking for? Browse other questions tagged or ask your own question.