Copying Behringer level/ signal indicator LED circuit

[diydidi]

I want to add some "signal present" LED's to some of my home brew test gear/ amps etc.
I decided to have a look to see how Behringer do it in some of their mixers.
Here attached is a section from one of their SL range mixers. I just decided on Behringer, because I have so many schematics on my PC.
Seems like they soft short the LED with transistor T1 in normal, no signal operation. When a signal appears from the previous stage opamp, T2 pulls T1 base to ground, switching the LED on.
Why would they use this logic? Wouldn't it have been easier to use one transistor in series with the LED and use a potential res divider in the base?
Any insight would be appreciated.

[J M Fahey]

Seems like they soft short the LED with transistor T1 in normal, no signal operation. When a signal appears from the previous stage opamp, T2 pulls T1 base to ground, switching the LED on.
Why would they use this logic? Wouldn't it have been easier to use one transistor in series with the LED and use a potential res divider in the base?

Just one way to see it, there may be others for sure:

1) in the simpler 1 transistor version, audio peaks at transistor base would turn it on (and light the LED), but for brief intervals.
Narrow pulses (say, a cymbal crash or peaks while strumming an acoustic guitar or transients while playing the piano) would distort in an annoying way, yet would be barely seen.
It's common to add a "pulse stretching" circuit, so a , say 1 ms transient lights the LED for, say, 100ms or more.

2) one way to do it would be to add a diode for rectification, a capacitor to store the detected peak, a couple resistors to establish attack and release times.
A simple Si diode detecting Audio is a crude rectifier, needs some power to drive it (which must come from the signal itself) , is somewhat "sluggish" because the current charging the capacitor comes straight from the Audio signal, etc.
We also lose an extra 0.7V (to turn diode ON) besides the 0.7V needed to turn transistor base ON .
May and will work, but has small problems as mentioned above.

3) now to the circuit used by Behringer (and definitely NOT invented by them):
* T1 is always ON
* its base current comes from the supply , not the audio signal
* it has constant 0.7V at the base, probably around 1V at collector, not enough to turn LED ON
* the audio signal only "sees" 22k or 12k to ground, easy to drive.
* T2 does double duty: its base acts as a rectifier, will turn on when peaks reach 0.7V , it's a relatively high impedance point (base current is very low because it's collector current divided by Hfe, say between 100 and 500) so it's easy to scale the trigger point to what's needed by varying 12k/10k ratio *and* it acts as a fast peak rectifier *and* it also acts as a very fast current sink to discharge 2u2 cap in microseconds.
Which will be recharged through 6k8 in a much longer time ... "stretching the pulse length".
* so by adding a cheap transistor T2 now we have:
* a very fast peak detector
* easy to drive
* threshold 0.7V peak, easy to scale to other values, instead of 1.4V minimum and with sluggish response ...what's not to like?

FWIW I LOVE this kind of very simple minimalistic approach, my own IOC type limiter which I fit to my Power Amps , which detects the power amp reaching clipping and only then limits, isn't fooled by varying rail voltages or load impedance and detects subtle distortion (no need to reach clipping) if the amp struggles to drive a reactive load, yet has a total parts count of 4 (four) , none of them an IC (which would be sort of cheating) and allows some 5% to 10% distortion (but no more) when acting, so my MI power amps are *just* a little dirty, what the Doctor ordered for MI duty.

When tested side by side with, say, a Peavey (DDT compression) , Fender (Deltacomp) or Ampeg/Crate (optocoupler) , mine is grungier , definitely less polite in a good way, yet never buzzes or farts.

[diydidi]

JM, Thanx for the great reply.
So you're saying this is how you would have done it?
Please if you may, would you mind sharing your IOC type limiter with me.
My email is riaandd@gmail.com

[patlaw]

Since current is always flowing through the 1.5K resistor, you would not use this circuit for a battery operated device. A PNP could be used for T1 such that current flows only when the LED is on and the capacitor is charging. Using a Darlington PNP would would allow you to use a smaller value capacitor for the same time constant. That being said, this circuit works and works well. If you like it, use it.