Help with noise in an optical phaser

[RLawlor]

I've been working on this optical phaser for quite a while and I've got it sounding how I want... but no matter what I try I can't seem to get a feedback control working without it just creating noise. The top schematic is what I am currently using and works well but when I try the second (which I would use so I can feedback to the non-inverting input of the first opamp a la mutron phasor II) all I get is noise. Anyone have any ideas? Any help would be much appreciated, thanks.

PS: The phaser is running off of +12V with a buffered virtual ground of +6V to use with the opamps. The reason for +12V is that the lfo is driving a lamp similar to a univibe and it was a lot easier to design with the extra voltage on tap.

[ElectricDruid]

What happens if you take the "U?A" inverting op-amp stage out of your top line schematic?

If that works, then the bottom line schematic is simply sticking that stage in at the beginning of the chain. When you say it's "noisy" do you mean it makes a lot of white noise, or do you mean it oscillates and howls?

[RLawlor]

I wouldn't quite say howling, you can hear the current from the lfo almost. It's more power noise than a sort of feedback noise. I'll try removing that opamp and see how that goes, thanks.

[Kipper4]

Feedback in phasers is usually low gain so it won't get out of control.
You don't say what lfo you are using or what kind of wave shape you utilised.
Is the lfo power supply buffered and or isolated from the main power supply and do you use smoothing caps in the supply.
I hope you find the problem or can isolate the noise so it might be diminished.
Rich

[RLawlor]

You don't say what lfo you are using or what kind of wave shape you utilised.
Is the lfo power supply buffered and or isolated from the main power supply and do you use smoothing caps in the supply.

I didn't include the lfo because its something I've spent ages on designing myself. Depending on switch settings it can be triangle, hypertriangle, inverted hypertriangle or sine. It uses transistors to put out 60mA (only at the highest point in the wave) through the 6V/60mA E10 bulb I'm using so I think the high current could cause an issue.

You said about isolating the supply, currently I'm sharing the virtual ground between the LFO and audio sections of the circuit so would you suggest having one for each? I've used a 100uF electrolytic and 100nF to smooth the supply, should I use a 100R resistor in series to the audio supply and one to the LFO supply to help isolate it as much I can? Many thanks for the help.

[11-90-an]

Sharing the lfo schem could help... (but if you don't want to, we can't do anything about it, can we? )

So this circuit of yours, is it on breadboard or on pcb already?

[RLawlor]

Sharing the lfo schem could help... (but if you don't want to, we can't do anything about it, can we? )

So this circuit of yours, is it on breadboard or on pcb already?

If it doesn't start working I'm not going to have much choice but to share the LFO  . I've currently got it set up on a breadboard, LFO set up on one and the audio on another. I'm in the process of designing PCBs but I'm a bit wary to order anything as I feel feedback of some sort would be considered a must in a modern phaser. I do like the sound but for slow sweeps a bit more oomph is in order  .

[11-90-an]

If your LFO uses op-amps, did you put some decoupling caps near them?
I'm not sure, but try placing a buffer just on the output of your LFO...

If you want to try some feedback, try looking at this phaser by Freppo for reference...
http://parasitstudio.weebly.com/uploads/2/4/4/9/2449159/parasite-phaser-final.pdf

[Kipper4]

You do have a choice. If you want to keep it to yourself that's not an issue.
The wave shape might be effecting the phased sound.
If you have a cro, might be worth a poke.

I did a phaser with and 8 waveform lfo from Electric Druid and not all where appropriate for all music genres.

Using a lamp might be making a difference too.

Breadboards are notorious for being noisy too without a faraday cage. Could be a contributory factor.

Hope this helps.

[Scruffie]

Sharing a passive v.ref between audio and LFO usually ends in tears and assuming all those 6V points on your schematic are actually V.Ref, you've got a lot of places for trouble to enter.

Either buffer it (which comes with its own gotchas someone will point out but is usually fine for anything not high gain) or use separate ones.