tips to reduce tremolo LFO click noise?

[darron]

hey guys. i've searched around a bit for this subject on phasers and tremolos without much luck. in a separate thread link, i've posted a schematic for a tremolo link. it's optical and entirely passive as far as the audio signal is concerned. it just uses a duel opamp for the LFO. for my purposes it runs perfectly on battery. however, if you connect the negative supply of the power to the grounding the clicking goes through as the LFO changes direction. that's not a problem for me as there's no need to connect the two. but if i put a DC jack on this in the schematic and people built it, it wouldn't be long before someone uses a common power supply with other pedals in the chain, which are almost guaranteed to connect negative and ground.

i've tried putting a massive cap across the power which didn't help at all. i've only been playing with it on breadboard and battery though. this matter is urgent as i have a lucibox coming for the project, and i need to know whether to put in a dc jack! hehe

[smallbearelec]

When you actually lay it out on a board, keep the wiring as tight as possible. Keep the LFO entirely separate from any modulator circuitry. Ground LFO and modulator separately, and bring the grounds together only where the negative power supply lead enters the board (this last courtesy of R. G. Keen). This is usually enough to swamp any ticks, but sometimes it also takes a .47 to 1 mf. poly film cap across the LFO output to round off the edges of the waveform.

[darron]

When you actually lay it out on a board, keep the wiring as tight as possible. Keep the LFO entirely separate from any modulator circuitry. Ground LFO and modulator separately, and bring the grounds together only where the negative power supply lead enters the board (this last courtesy of R. G. Keen). This is usually enough to swamp any ticks, but sometimes it also takes a .47 to 1 mf. poly film cap across the LFO output to round off the edges of the waveform.

thanks for your response small bear (:

if i understand correctly, there's no modulator per-say as it's just a a LDR going between the signal and ground. i'll try playing with some small value film caps like you suggest, they might smooth things out here and there. i had only tried them on the power supply, but i'll try them on the lfo as you guest also. thanks (:

[gez]

As has been pointed out, keep the LFO ground separate from signal ground and only join the two at the -ve terminal of your supply.

Mark often outlines a trick that you see in some commercial pedals.  I'm going by memory here, but I think it involves a resistor from the - input of the Schmidt square generator to Vref and a cap from its output to its - input.  This slows down the on/off transition (trapezoid waveform).  Apparently, it causes minimal distortion of the triangle wave, though I've never needed to use it.

Other tricks: use low power amps (minimal current spikes); decouple supply close to the chip's pins; use CMOS inverters run at low voltage (again, minuscule current spikes).

The solution to your problem is obvious, however.  Simply disable your LFO in 'bypass' mode.  The resistance of the LDR will be 'infinite' and will simply be another parallel resistor in the chain - big deal!  The easiest way to do this would be to disable the LEDs with a switch.  Doing this would totally isolate the LFO circuitry: no clicks.

[gez]

PS  Your LEDs are the modulator, in this case.  When putting out a triangle output, there's still the possibility of spikes if they share the same ground line with the opamp LFO.

[gez]

The solution to your problem is obvious, however.  Simply disable your LFO in 'bypass' mode.  The resistance of the LDR will be 'infinite' and will simply be another parallel resistor in the chain - big deal!  The easiest way to do this would be to disable the LEDs with a switch.  Doing this would totally isolate the LFO circuitry: no clicks.

Early morning post, with a groggy head!  Above defeats the object as you'd need an on/off switch.  I'm talking b*llocks again...

[darron]

hey gez. thanks for the posts! i'll definitely start using a LM1458 instead, as it's general purpose (lower power) and has internal short circuit protection. i did like the jrc4558d mojo though

the pcb layout i have at the moment switches the power off to the circuit on bypass as well as removing the signal from the ldr. you are right, the LDR resistance is off my meter in darkness. bypassed isn't a problem. i'll play around with yours and smallbear's ideas on the trapezoid effect. this at least gives me a starting point to research further. like i said, as it is now there is no noise, but i foresee the problem and would like to be able to run it off my daisychain.

thanks again (:

darron