Diode switching and LED lighting

[pjones78]

So, here is a question. In a circuit with diode clipping, say 2 1N4148 diodes with a dpdt switch. Is it possible to wire a LED in with them only for a visual conformation that the diodes are activated?

[induction]

Not very easily, no.

Diodes conduct when the applied voltage exceeds their intrinsic forward voltage (Vf). The 1N4148 has a Vf somewhere around 0.65 V, while the Vf of an led depends on the color and the part number, but is usually upwards of 2V or so. Thus the 1N4148's will conduct earlier, and will keep the signal voltage well below the onset of conduction for the diodes.

If you can find leds that have exactly the same Vf as your 1N4148's (not sure if they exist), then you can just replace the 1N4148's with the leds, and mount them externally. But be prepared for disappointment. Led's are often used for clipping, and experience shows that they will flicker a little bit at best. It might be visible, but probably won't be that useful or interesting. The small amount of current from clipping isn't enough to light them up very brightly, or for very long.

It is certainly possible to add a sidechain just for the function you want, but it would likely be a larger circuit than the rest of the pedal, and would require a great deal of fine tuning to get it to work properly.

That said, if you just want to explore the sound of clipping with visual confirmation of what you hear, build your circuit on the breadboard and use leds instead of 1N4148's. Increase the pre-clipping gain and make sure there's a volume control after the leds (higher Vf means louder output when clipping is reached, so turn down the output volume to get a similar volume to the same circuit with 1N4148's). If you make the pre-clipping gain variable, you can experiment with turning up the pre-clipping gain until the leds start to flicker and listening to the effect on the tone. You should simultaneously turn down the output volume because our ears often prefer louder sounds, so it's easy to trick yourself. Also, the louder output might overdrive the input of your amplifier, or whatever's next in your signal chain, which will confuse your observations.

[strungout]

Guess the easiest way would be to use a 3pdt, no?

[Markw5]

You can buy a very cheap oscilloscope on ebay for less than $30 and use that to look at clipping if you want a visual indication. Might be kind of fun too. However, while the bandwidth of those is enough to look at audio signals in a pedal, they are not for good enough troubleshooting high frequency oscillations or something like that. If you wanted it for that too, you would probably want at least 20MHz bandwidth.

[J0K3RX]

Am I missing something? Forgive me if I misunderstood...  You said dpdt switch for two 1N4148  clipping diodes and an LED indicator to indicate when the clipping is on or off, no?

[induction]

Am I missing something? Forgive me if I misunderstood... 

Maybe I misunderstood. I assumed that the diodes being 'activated' meant they are actively clipping. If the OP just meant that they are in the circuit, then ignore everything I said above.

[rutabaga bob]

That's what I thought, as well - one pole of the switch to kick in an indicator. 

[electrosonic]

If you want the LED to light when it is actually clipping - maybe this?



Andrew.

[amptramp]

You can do the mirror image of what electrosonic suggested using a PNP transistor to ground and an additional LED but you would have to have a -9 volt supply.  If the clipping diodes were referenced to a Vbb/2 supply, you could make it work without a second supply.

[pjones78]

That is what I was curious about. clipping diodes that when the switch it "on" there would also be a LED indicator for them to be on. I figured dpdt for duel gang and not a spdt?

[Markw5]

The transistor trick might not sound quite the same since it probably wouldn't have the exact same forward drop as the signal diode, and it has miller capacitance to the collector. Then again, you might not notice that stuff sound wise. You could also take the original circuit with 2 diodes to ground and instead of ground, use a virtual ground to an opamp with 2 LEDs in the feedback path. Then whatever current flows through the signal diodes will also flow through the LEDs without changing the original clipping threshold.

[idy]

my favorite is to use dpdt and switch two pairs of diodes and change the color of a 3 color LED. This sounds more "purple", that more "blue or green".... It' a kick to change color for functions.