Hello folks, Im making a project that requires a pair of LEDs to glow and another pair to dim upon receiving an envelope signal from a synth. I have breadboarded a circuit that works where a npn transistor receives the signal allows a pair of LEDs to glow, then an LDR’s resistance drops shorting a second pair of LEDs making them fade out.
It works pretty well but seems like it might screw up the LDR quite quickly. Is there a better way of doing this?
my current diagram…
PWM is the best way to dim an LED. If you want to do it on a component level you might be able to use a triac.
I’d have thought the 10k in the 2’ndary path would produce some very dim LEDs, getting dimmer as the others get brighter. Without knowing more about the envelope signal (DC bias and pp levels) it’s hard to comment. one simpler way might be to use a circuit that’s similar too, but the compliment of, your primary cicuit. It would use a PNP such that when the envelope signal grows large, the PNP is biased more off. And when the envelope signal grows small, the PNP is biased more on. The opposite of the primary channel and what I think you’re looking for.
ps - details on the LEDs is needed for any design redo. Ditto for envelope circuit.
Thanks for the replies! And sorry for the lack of detail: The LEDs are 3v 20ma, and the envelope swings about 0 - 3v, the LDR is 5516 that swings between 0 (under LED light) and about 15k (when LED is extinguisher).
My ‘original original’ design worked with a PNP to get the opposite response but I couldnt work out how to set it up correctly.
I have made some modification and to tune the response and its working really well.
The resistor to the envelope signal is 10m
The resistor to the top leds is 100ohm,
the one in series with the bottom LEDs is 68k,
and the one in series to the LDR (and parallel to the bottom LEDs) is 47k
The LEDs are relatively dim but I think they should be ok for this application. So my concern really just is about if the LDR is ok being used in this way and that it wont die after building it up. I dont mind it being a bit of a cranky simple solution, (occams razor and all that) as long as it works and is robust. That said it would be very useful to know the PNP solution for future reference. Thanks again for looking at my problem!