I'm in the middle of building my own pedal: ferrite cores?

Thecomedian · #1 11-11-2012, 05:53 PM

Can ferrite cores or rings be used inside a pedal to reduce noise? I believe I've seen these things in TV circuits, but they mostly come on the ends of laptop cables. Can they be used inside a pedal to filter noise out? Would there be any audible benefit to doing this?

Bassamatic · #2 11-11-2012, 06:43 PM

Ferrite cores are only used to filter out very high RFI type noise and noise from switching power supplies, etc. They will not remove normal audio hiss and hum unless it is caused by something like a 60 kHz digital clock on the power bus, radio or TV signals leaking in, etc. Just use design, grounding and shielding to minimize noise.

icecycle66 · #3 11-11-2012, 06:49 PM

What sort of pedal are you building?
If you are picking up faint radio of TV stations, the ferrite could be useful.

Thecomedian · #4 11-11-2012, 07:38 PM

Quote:

Originally Posted by Bassamatic

Ferrite cores are only used to filter out very high RFI type noise and noise from switching power supplies, etc. They will not remove normal audio hiss and hum unless it is caused by something like a 60 kHz digital clock on the power bus, radio or TV signals leaking in, etc. Just use design, grounding and shielding to minimize noise.

alright, thanks.

Quote:

Originally Posted by icecycle66

What sort of pedal are you building?
If you are picking up faint radio of TV stations, the ferrite could be useful.

https://www.smallbearelec.com/HowTos...dboardSiFF.htm

I have almost enough knowledge to understand how to use op-amp circuits and diodes, but this is my first pedal, its pretty simple.

play4sanity · #5 11-11-2012, 07:41 PM

I've designed lots of electronics over the years for many different applications. If I were doing a full custom pedal, I would put an LC type filter on the incoming power to keep any noise from the power supply from getting into your pedal. It also keeps any noise in your pedal from getting back to the power supply. Just a good practice kind of thing.

A ferrite core for the inductor would be common just make sure it's rated for the current your drawing. I would also use a ceramic cap for the C in the LC.

I would also recommend using a schottky diode in series with the power line to protect against power reversal. Might keep a woops from becoming an uh-oh.

As mentioned earlier, the noise here isn't audible frequencies but electrical noise which can cause problems with other circuits not operating properly. Ferrites don't make good audible frequency inductors because the inductance changes with temperature and frequency and would produce results that would sound poor. The best audio frequency inductors (like the ones used in crossovers) are air core which are very stable over both temperature and frequency. They're just really big, way too big for use in pedals.

My 2 cents ...

gumtownbassman · #6 11-11-2012, 07:48 PM

If you do want to use the ferrite bead, use it on the input to keep the noise from getting in the first place.
They are only used on outputs if the pedal contains digital (DSP) or switch mode supplies, to filter artifacts generated by such circuits.
Quite likely no audible benefit from using one.

Thecomedian · #7 11-12-2012, 12:48 PM

Quote:

Originally Posted by play4sanity

I've designed lots of electronics over the years for many different applications. If I were doing a full custom pedal, I would put an LC type filter on the incoming power to keep any noise from the power supply from getting into your pedal. It also keeps any noise in your pedal from getting back to the power supply. Just a good practice kind of thing.

A ferrite core for the inductor would be common just make sure it's rated for the current your drawing. I would also use a ceramic cap for the C in the LC.

I would also recommend using a schottky diode in series with the power line to protect against power reversal. Might keep a woops from becoming an uh-oh.

As mentioned earlier, the noise here isn't audible frequencies but electrical noise which can cause problems with other circuits not operating properly. Ferrites don't make good audible frequency inductors because the inductance changes with temperature and frequency and would produce results that would sound poor. The best audio frequency inductors (like the ones used in crossovers) are air core which are very stable over both temperature and frequency. They're just really big, way too big for use in pedals.

My 2 cents ...

So, if I read that right, the rest of the circuits in the board could function improperly, leading to poor performance of the pedal and its intended use, if power noise is allowed to get into the system?

play4sanity · #8 11-12-2012, 02:10 PM

Quote:

Originally Posted by Thecomedian

So, if I read that right, the rest of the circuits in the board could function improperly, leading to poor performance of the pedal and its intended use, if power noise is allowed to get into the system?

Yes. And integrated circuits like opamps and such are more vulnerable than discrete components like transistors or even vacuum tubes. They almost always have a small (0.01uF-0.1uF) ceramic capacitor placed as close to their power pins as possible and an LC filter on the power if it's coming from some distance away. Like over a cable or from another PCB if it's in a large system.

Amplifiers in general are considered "linear" meaning what you get out is always proportional by some set amount to the input. Distortion happens when that "set amount" changes. Sometimes we like it, say in a tube amp or a effects pedal. And sometimes we don't. Like when hums, pops, cracks, or other unpleasantness creeps in.

At low levels it will usually just show up as an increase in the background noise. As the levels get higher you'll start to hear it in the quality of the sound. The higher frequencies always get affected first thus the reason those get cut first to decrease the problem.

You can't eliminate noise but you can reduce it's impact through some tried and true steps. And filtering your power is an easy thing to do that won't cause any corresponding audio signal loss.

And don't forget proper grounding! The best is a star type configuration where everything gets tied back to a single point. The worst is a daisy chain with long links in between. Reality usually falls somewhere in between. Just try to make it fall closer to the best instead of the worst.

Edit: Usually the LC filter on the power is only used once for the whole board (or in your case, the whole pedal) where it enters the board. So probably wire from the incoming connector to the LC filter and then to the rest of the board. Going back to the "best" star configuration ground, the ground lead of the "C" in the LC is typically the best place to become your "single point". That works well for the power too if you can do it.

gumtownbassman · #9 11-12-2012, 02:14 PM

There are an infinite number of factors that can influence project performance,
but i would not worry about those unless they present a problem after completion,
just build the project and test it first.
Then find any issues after,
the fun of D.I.Y. projects. :-)

play4sanity · #10 11-12-2012, 02:32 PM

A perfectly acceptable strategy but just not one that my boss allows me the luxury of ...

boomertech · #11 11-12-2012, 03:29 PM

One thing that is nice about op-amps is that they have a large PSRR (power supply rejection ratio) sometimes listed as kVSR on data sheets. The TI TL074 has a kVSR of 86dB… that means that any ripple or noise on the supply pins will be 86dB less at the output. 86dB is a voltage ratio of approx 20,000 to 1.... very nice!

-Frank

Thecomedian · #12 11-13-2012, 11:48 AM

And here it is almost finished. Need to replace the 2.2. mfd cap. I did as much ground "star" as possible on the backside. I got all these parts for free, so..

gumtownbassman · #13 11-13-2012, 01:00 PM

So... what does it do?
looks like a distortion pedal maybe?

Thecomedian · #14 11-13-2012, 04:59 PM

https://www.smallbearelec.com/HowTos...dboardSiFF.htm

supposed to be this one when it's finished. "fuzz". It has a lot less parts on the link because I had to do some series resistor groups to add up to the value I needed.

Thecomedian · #15 11-25-2012, 12:37 AM

It works. the circuit is in nominal function, the only problem is.. the distortion is not so noise friendly (unless you like electronica sounding music) at the high end of the "tone" 1k pot. Turned all the way down, it sounds great for the low ends, but the high frequency fundamental notes need some heavier string plucking to get the distortion effect to occur, otherwise they will ring out distorted very nicely if you hold frets on both D and G high strings. On the other hand, any more than one string ringing when a low string is going off equals a huge delay/distortion effect between the notes, meaning that the number of milliseconds the transistors are in saturation and shut off mode is being elongated by lower frequencies, and higher frequencies are not posing this same problem, so I'll look into changing capacitors first in simulations, then tweaking resistors if capacitors don't have any useful effect.

Also, I'm getting a lot of "hiss", which increases exponentially when I turn the volume up. I'm betting it's due to being a 15 watt "practice amp".