Voltage Doubler inducing annoying noise. Help?

[masinyourface]

After reading and re-reading Jack Orman's article on getting a cleaner sound from boost pedals http://www.muzique.com/lab/boost.htm, I decided to attempt an opamp based clean boost with a voltage doubler circuit to clean up my boost a bit more. I had a few 555s on hand so I decided to design the voltage doubling part of the boost around one.

Here's the voltage doubling circuit I used.


I then used two 10K resistors going from +18V to ground in order to get a bias voltage for the guitar signal going into my opamp.

Breadboarded it out, and the boost itself works great, but I'm having problems with an annoying high pitched tone that remains constant no matter how the variable gain knob is set. I'm almost certain it's because of the 555's oscillator.

I tried to add Dano's Humminator at the 18V output from the 555 doubling circuit out of curiosity. (minus the LED) http://beavisaudio.com/projects/Huminator/index.htm Still no luck.

Should I invest in a voltage regulator? Would that eliminate the noise? Maybe even a zener based voltage regulation circuit? I'm open for suggestions!

[crane]

My first guess would be - you need some ceramics (I'd say 1nf +100nF) in paralell with thos big ass electrolitics.
My second guess would be - how high is the switching frequency? May be it would be good to make it higher than 20k?

[artifus]

My second guess would be - how high is the switching frequency? May be it would be good to make it higher than 20k?

that was my first thought. according to http://www.royalrife.com/555_calculator.html and http://www.electronicdesignworks.com/utilities/555_frequency_calculator/555_frequency_calculator.htm the frequency is currently 4.47k. i would try much higher too.

*punk-chew-ation edit*

[ubersam]

My second guess would be - how high is the switching frequency? May be it would be good to make it higher than 20k?

That'd be my guess as well... try lowering C1 to 1nF, or lower R1 to 1K

[R.G.]

The 555 is known to cause big problems with noise passed through the power supply and ground. At the instant of switching, it pulls short pulses of very high current.

There are two solutions: (1) be very, very paranoid and thorough about wiring, decoupling, and so on or (2) use something other than the 555.

You might really like the LT1054. Just a thought.

[masinyourface]

Thank you everyone for your input! I think I'm going to tinker with this a couple more days and see if I can get it to work with the 555. If not, I'll take a closer look into the LT1054.

[Toney]

Yeah, if its up and running and oscillating above audio frequency, I guess chucking all you got at it with filtering is the go.

If you have a few BJTs and a scrap of Vero lying around you can make this:
http://www.diystompboxes.com/smfforum/index.php?topic=97141.0

Pretty quick and easy.

[PRR]

> used two 10K resistors going from +18V to ground in order to get a bias voltage

Any filtering (capacitors) on that??

There's quite a racket coming off the 555 and diodes. Cut in half and feed to the INput of an amplifier, that's a lot of racket right into your signal. 10uFd off the 10K+10K junction should make a difference. (You 'could' have just took your +9V direct as your half-voltage, though there could be tricky pitfalls.)

Also: what diodes? 1N400x series are slow for 4KHz work, hopeless if you move the buzzer up above the audio band. This could give a very ugly waveform. 1N914-like devices will take the speed but will be very near their current limit working between a 555 and a 220u cap. There's several types of fast-switching power diodes.

[Toney]

 You can get 1n400x and a few other ultra fast types at Futurlec

http://www.futurlec.com/DiodesGen.shtml#Fast