LDR noise

[jnee8871]

Hello,

I was wondering if anyone had any tips on making an LDR less "poppy". In a tremolo circuit using and LED/LDR. The LDR goes to ground. It is 100% the LDR because the astable mutlivibrator and the audio do not yet share the power supply.

If I use a LED that has less "brightness", or use a pot to lower the same, it gets softer. However, that isn't as ideal.

Any help would be great.

Thanks
Jeff

[Zoe_N_Iain]

I've seen them with a ~100pF cap strapped across them, if that helps any? I assume that's done for the same reason????

[jnee8871]

Tried

Hey

Thanks for the reply. I had tried that and it actually seemed to make the click more defined.

When I used a larger cap it cup overall volume.

There must be a solution...

[Zoe_N_Iain]

Could you run the LDR to ground through a hi-pass filter, perhaps? That might kill the click but not deck the trem signal. Just thinking aloud, of course. Don't know if it would actually work.

It's definitely a bit odd

[Mark Hammer]

It will depend on the specific circuit, but a great many LFO circuits used for tremolos, phasers, flangers, what have you, start out with a square wave, and shape that, via an integrator stage, into a triangle wave for modulation purposes. The trouble is that during the rapid swing from one extreme to another at the onset of the initial square wave, there is a huge current demand from the LFO, which can have an audible impact on the audio path, not unlike the good old days when the compressor on the fridge, or your mom's sewing machine motor would interfere with the TV or stereo sharing the same household power lines, creating audible static.

There are several ways to tackle this. One is to use a low-current op-amp for the LFO such that the current requirements to produce that square wave are low enough so as to not cause a disturbance elsewhere in the circuit. You will note the frequent presence of TL022 and LM358 dual op-amps in a bunch of modulation-pedal circuits. These are used for exactly that reason.

A second approach involves "decoupling" the power lines feeding each portion of the overall circuit. If the LFO portion of the circuit has its own isolated power lines from the 9v battery, then whatever happens to the LFO should leave the audio portion of the circuit unaffected. The simplest way to do this is to use something like a 10 ohm resistor from +9v to the power pin on the op-amp, and a 47-100uf cap from the power pin to ground. I should also point out that some modulation-pedal circuits, or at least their builders, attempt to save some space bypacking the LFO and part of the audio circuit into a single quad op-amp. The problem with this strategy is that the audio and LFO circuit segments will share the same power line and you can't decouple the one from the other once in the same chip. Here, using two dual op-amps is smarter and quieter than using a single quad.

The 3rd approach is a clever one I saw described by Nicholas Boscorelli in his now defunct Stompboxology newsletter. Thye standard 2-opamp LFO is modified so that instead of initially producing a square wave to turn into a triangle, it produces a trapezoidal wave. A small-value cap in the feedback loop of the first stage of that circuit introduces a tiny bit of lag in both the rise and fall times of that initial square wave. Just enough time that the swing is not instantaneous enough to be audible. Wish I had the schem to show you, but it isn't handy at the moment.

Any one, 2 or all three of these strategies used can be helpful in reducing audible ticking in most modulation circuits. The ticking is never from the LDR, but from what is driving the LDR.