555 timer noise on tremolo circuit

[Flynn]

Hello all,

I built the 555 Tremolo Circuit by Genxyde.    The noise from the timer is very audible. 
I then isolated the two sides, powered them individually, and zero noise of course.

At present, I am running one 9v supply into the 555 side, splitting it off, and trying different filtering options on the breadboard before sending that rail to the J201 side.

A 100uf to ground didn't make much of an improvement. 

Would love to hear some suggestions in hopes of avoiding having a separate supply.

As a side note, it would also be cool to make the LED more 'sine' and less 'square'; I tried putting a cap in series with the LED with no success.

Schematic:

[GibsonGM]

Try a 330R followed by a huge cap to ground where power goes to the 555.  Like, 470u to 1,000u)....might work, might not.   

If not, you could try placing them after the "Vcc" on your schematic, where power feeds the drain.  The idea is to try to decouple the 555.   Alternatively, come up with another source to tweak the LED rather than a (current hog) 555.  CMOS version might work better....

[duck_arse]

yeah, like gibson says. start with a cmos 7555 or LMC555 or whatever. then move that "+9V" point on your dia to the line between the "Vcc" pin and the 1k (fet) resistor. now put a resistor between "+9V" and the 7555 supply - 100R should be fine enuff - and a cap to ground - 22uF, 47uF, 100uF etc, whatever. then add another resistor and another cap between the "+9V" and the 1k (fet) resistor, so you have now isolated both circuits from the supply.

and - build the 7555 and all its grounds on one side of your (-) supply to the board connection, and all your audio grounds connect to the other side of the (-) to board point.

there is a way to add a transistor [emitter follower?] to buffer the Discharge pin (from memory) that could then drive the led instead, the output will be more sawtooth then squarewave.

[that diagram shows a mosfet, without bias, which I don't think will work. I'm not sure a jfet will trem right either.]

[Flynn]

Thank you, gents!

I've got a 1000UF across the 555 rails and added a 330r in series from the 555 supply to the Jfet side with a 2200UF to ground - the noise is 100% gone!

Rockin!!!   

It sounds good, even if its more squarish than sine-ish.  This is a good little circuit that just needs a few extra components.

Thanks again to you both!!

[GibsonGM]

Oh wow, you mean I diagnosed the problem correctly??   Ha ha     Glad it worked Flynn!

[blackieNYC]

Will a cap across the LED soften the waveform?

[Flynn]

GibsonGM:  Yes you did!  I actually already had that cap in place.  I needed only to add that 330r in series and whamo!   One single resistor!!!  Thanks for your expertise.

Hey blackieNYC, thanks for the suggestion.  I did try a few different cap values in parallel with the LED.  did i do that right?  it had no effect, but willing to try again.  Hey by the way, weren't you in LA for a while doing tube amp repair, etc?  If so, when i was in LA we spoke on the phone about tube amps and punching face plates.  Still don't have the greenlees you suggested but my harborfreight 'christmas tree' bits have been doing the trick so far!  :-)

[blackieNYC]

Not me.  And not sure about the cap. Big electrolytic? If you tried it, you tried it. 

[Flynn]

yeah i tried it.  no luck.  but not sure i did it right!

btw, it was Blackie Pagano i was talking about...  a coincidence i guess!  errks, sorry...

[GibsonGM]

Read up on R-C time constants, and RC filters, Flynn...there is a lot to the story of why just the cap didn't do it.  Needed the "R" to go along with it.   

All we did there was provide a big reservoir so that when the hoggy 555 pulled a lot of current to do its thing,  the following FET didn't have an interruption in its power supply.   The resistor could be thought of as a 'valve' (umm kinda) that isolates 555 from FET.   It's a very common trick to do to a power supply when a low-speed oscillator is causing ticking, so keep that one in your parts drawer for later use!  

[ensi_genxyde]

Well howdy, you wonderful individuals.
I came across this page a few days ago while googling myself, and I wanted to thank you for tinkering with the circuit.
I had tried to remove the ticking with a voltage regulator to separate the two sides and a large valued cap, but that didnt remove the issue and it made the LED's power weaker, so I had to switch from a transistor to an IC to reach unity signal compared to bypassed. That was version 3 that I never showed off, lol.
Hopefully this winter I can get around to trying your suggestions.
I left a link to this page, on the tremolo's webpage so that others can find there way to this modification.
Once again, thank you very much for tinkering with it and offering some suggestions.

[Flynn]

@GibsonGM:  Thanks for taking the time to explain that, and yeah this is one i won't be forgetting about!!!

@ensi_genxyde:  Cool to have you post here!  Your circuit is great and these two additional parts did the trick completely.  The only thing i have been playing with is 'softening' the edges of the trem.  I tried a cap across the LED but it's still pretty 'square'.  After shrink-tubing the LED and LDR, it was super square, so i took the tubing off, put them at an angle to each other, and hot-glued them together.  That helps a bit, but would be cool to work the circuit in the future to be more of a sine wave driving the LED.  By the way, i tried to contact you through your site when i got this working to let you know about it.  Glad you found it anyway!