Hiss from 555 timer cleaning

[willbloody]

Hello World.

I am building fuzz guitar pedal and want to include simple fading led(just a visual design idea) to the circuit. Well it works, but there is a lot of unwanted hiss in the signal. So I am struggling with method to reduce it. There were loud pops at first which matched frequency of a 555 timer, but simple ceramic cap between +and- of a power supply made theme gone. I believe that I must somehow isolate power supply, but don`t know how(Just learning).

 

[rjenkinsgb]

The fuzz part is just an asymmetric overdrive, which has high gain. Any input noise or transistor noise will added to the output.
An electrolytic cap across power would not hurt.

As you have gone for PNP transistors, does that mean you are using germanium ones?

You could try adding a small capacitor, possibly 100pF or something in that range, either between base and collector of the first transistor, or across the input jack - see which works best.

If you have a selection of caps, try different values until it starts to affect the sound at high frequencies, then use eg. a half or quarter of that value, to balance noise reduction vs. tone change.

You could also try different transistors? If you are using silicon types, then you could try ZTX951, which are very low noise and readily available.

 

[AnalogKid]

1. Almost all integrated circuits, especially analog ones, *require* power supply "decoupling". This is done by piling up capacitors as close as possible to the device's power pins. The leads should be as short as possible. This is very easy with the 555 because the power pins are 1 and 8. Start with a 0.1 uF ceramic and 10 uF-to-100 uF electrolytic in parallel. Put the ceramic closest to the pins.

2. There is nothing about the 555 circuit that could inject hiss into a nearby audio signal. Does the hiss go away when you remove power from the 555?

ak

 

[rjenkinsgb]

There is nothing about the 555 circuit that could inject hiss into a nearby audio signal. Does the hiss go away when you remove power from the 555?

I had a thought during the day that it could possibly be the LED acting as a noise source??
I know some semiconductor junctions do, but I have no idea about LEDs..

It may be worth trying a ceramic cap (eg. 0.1uF) across the LED and its resistor, or across the LED alone?

 

[Nigel Goodwin]

I had a thought during the day that it could possibly be the LED acting as a noise source??
I know some semiconductor junctions do, but I have no idea about LEDs..

It may be worth trying a ceramic cap (eg. 0.1uF) across the LED and its resistor, or across the LED alone?

You get plenty of LED's in preamps etc. and they never capacitors across them.

I would suggest the first thing to do is stick a decent size electrolytic across the supply rails, absolutely essential on battery powered equipment.

 

[willbloody]

Thanks for reply. Using decoupling caps does not change the situation. Let me explain the hiss...it is not even the hiss...it is like synthy swirling birds that sing in beat of fading leds)) if I increase the fading rate to max, I receive a good oscillator. Noise goes away after removing supply from 555. The transistors are germanium ac128. maybe they are just incompatible with 555 timer

 

[rjenkinsgb]

Can you post a photo of the unit, as you have built it?

 

[AnalogKid]

I know some semiconductor junctions do, but I have no idea about LEDs..

Yes, but in any circuit that uses a semiconductor as a noise source, you will see lotsa gain after the device. That's because while some devices are noisier than others, none of them make enough noise to overpower a 1 V signal.

If the LED is the noise source, that noise has to modulate the amount of current through the 555 to the point that the Vcc source starts to fluctuate despite locas power supply decoupling. Then, those fluctuations have to get through the signal amplifier's common-mode noise reduction to appear on the actual signal. I've been through this with TEMPEST designs. Whatever you think, it's harder than that.

ak

 

[willbloody]

Unit is currently in breadboard mode) tried to rebuild it to negative power supply and vise versa, no result. The timer scheme is presoldered on pcb

 

[Nigel Goodwin]

Been on breadboard isn't going to help, it's quite possibly unstable and oscillating?.

 

[rjenkinsgb]

.it is like synthy swirling birds that sing in beat of fading leds

That sounds like ultrasonic or RF oscillation / squegging, due to the layout and/or lack of sufficient decoupling.

It could also be affected by the proximity of input and output connecting wires to other parts of the circuit, hence the request for photos.


I've built diode noise sources so I'm aware of the gain needed, and also I've never experienced noise due to LEDs.

But I've never used one in a high gain audio circuit with no decoupling, either, hence the question of if that could have any effect.

Edit - post from eight hours ago.. I failed to hit the button

 

[Nigel Goodwin]

That sounds like ultrasonic or RF oscillation / squegging, due to the layout and/or lack of sufficient decoupling.

It could also be affected by the proximity of input and output connecting wires to other parts of the circuit, hence the request for photos.


I've built diode noise sources so I'm aware of the gain needed, and also I've never experienced noise due to LEDs.

But I've never used one in a high gain audio circuit with no decoupling, either, hence the question of if that could have any effect.

It's chances of been stable seen pretty remote

Perhaps the OP would like a post a picture so we can see what he's actually doing?.

Edit - post from eight hours ago.. I failed to hit the button

Been there, done that

 

[willbloody]

I am experiencing this kind of problem not only once.. few month ago i have abandoned project with an already purchased completed pcb of a fuzz, and have added a soldered fading led scheme....the result is - same weirdness) Everything shielded, grounded etc...so I build from scratch everything. It works perfectly and is extremely quite, like any fuzz should be....it looks like breadboard full of wires....chaos as it is))) but only if I use ne555, fuzz transforms into random synth)) I should mention that reducing gain or dealing with Bias pot(33k resistor on scheme) does reduce the noise, but the sound suffers.