Possible way to reduce LED pops

[earthtonesaudio]

I haven't tried this yet, but I was thinking of a way to help reduce pops when switching the status LED on and off.

Basically you have one resistor+LED, always connected to the battery and ground.  When you want to turn it off, you put another LED in parallel with the first.  If the second has a lower Vf, it will dim the first.  The current drawn by the other LED is similar to that drawn by the first, so there's not such a huge change as you'd get from simply open-circuiting the LED (or shorting it).

Of course a smoothing cap would further help.  I hope to try this soon.

[R.G.]

Even simpler and cheaper: drive the LED from a two-transistor differential amplifier that's emitter biased. A diffamp's current is controlled by the emitter bias. All the bases do is switch the current from one transistor to the other. One transistor's collector has the LED in it, the other is a short to the supply. Hold one base at a fixed voltage, move the other base something like 100mV above and below the bias voltage to switch the current from LED to not-LED. Current is constant at all times. No click.

I've done this many times, works great. Costs you two $0.05 transistors and a few resistors though, instead of another LED.

[Barcode80]

Even simpler and cheaper: drive the LED from a two-transistor differential amplifier that's emitter biased. A diffamp's current is controlled by the emitter bias. All the bases do is switch the current from one transistor to the other. One transistor's collector has the LED in it, the other is a short to the supply. Hold one base at a fixed voltage, move the other base something like 100mV above and below the bias voltage to switch the current from LED to not-LED. Current is constant at all times. No click.

I've done this many times, works great. Costs you two $0.05 transistors and a few resistors though, instead of another LED.

Sorry to revive an old topic, but R.G. could you elaborate on this a bit? I'm having some pop troubles with some high current LEDs but I want them to stay as bright as they are so I don't want to change the current limiting resistor. Sounds like this could be an alternative.

[merlinb]

Yet another option is to supply the LED from a constant current source. It could be made from a couple of BJTs, a single JFET or even a current regulator diode. The footswitch simply shunts the LED to switch it off.

(On the other hand, most FX run the LED at barely 1mA. If anyone gets popping problems at those current levels then I would suspect some major shortcoming in the PSU...)

[Barcode80]

I'm not sure what circuits you are referring too, but a large number of circuits run off significantly more current than that. The EHX Holy Grail, for example, taps almost all of it's included 500mA power supply.

[armstrom]

He's talking about the current through the LED, not the current of the entire pedal. Most people run a 2k2 resistor as a current limiter. If you assume ~2V forward voltage for a red LED and a supply voltage of 9V that means the LED circuit is only drawing ~3mA. Your typical 5mm red LED from radio shack can handle up to about 20mA.

[Barcode80]

ah, gotcha. but super brights pull a lot more i'm guessing? every time i use them with about a 10k they still pop.

[CynicalMan]

Nope, super-brights pull slightly less current because they have a higher forward voltage. Also, because they are brighter for the same current, you can often run them with a higher current limiting resistor. I run super-brights with a 4.7k resistor, at around 1.5mA, and that's more than bright enough.

Edit: I did some tests a while ago and, using a 4k7 resistor at 9.5V, a blue super-bright LED drew 1.47mA, while a normal red diffused LED drew 1.68mA. The super-bright had a forward voltage of 2.74V and the red LED had a forward voltage of 1.69V.

[Eb7+9]

I haven't tried this yet, but I was thinking of a way to help reduce pops when switching the status LED on and off.

Basically you have one resistor+LED, always connected to the battery and ground.  When you want to turn it off, you put another LED in parallel with the first.  If the second has a lower Vf, it will dim the first.  The current drawn by the other LED is similar to that drawn by the first, so there's not such a huge change as you'd get from simply open-circuiting the LED (or shorting it).

you might want to consider the advantages of using an active current source to minimize the impact further

one approach I took in my 360+ Bass pres runs LED's off a dual current mirror where a switch shunts to ground either diode - resulting in a near constant-sum change in overall current state ...

http://www.effectsdatabase.com/model/vivaanalog/360/plus

another way, similar to what you're talking about and what I use in my Octave Amp to provide pop-less indication, is to drive the indicator LED by a cascoded jFET current source and parallel across it with a slightly smaller turn-on voltage LED so as as a means of incuring a negligible current variation in the rail ... the combination of very high Zout of the current source and near voltage of the LED's brings us very close to zero state change ... a trimmer in series with the shunting diode allows for a partial "bypass" glow in the LED also

a simple cascoded current source is all you need - nothing fancy