Help save my face.

Started by makaze808, September 12, 2013, 08:47:35 AM

Previous topic - Next topic

makaze808

Hi, my son had started an arduino project and thinks I am an electronics expert after building a few fx and amps and so I hope you can help restore his confidence in me.

So he is building a tremolo effect using the arduino 5v on output pins, the software turns the 5 volt on and off at a rate set by a potentiometer.

I bought some optoisolators, which seemed to claim low ticking etc, and sadly, although the (abrupt) tremolo is there, it sounds like the start to time by pink floyd.

A search suggests ticking from chips is common, I understand the abrupt 5v on off from the arduino output pin is problematic, but I was convinced the opto-isolator would prevent anything from the arduino side entering the audio path. I hadn't thought of arduino as a CHIP.

He has to use the arduino in the project.

The signal is just a guitar, going through the breadboard, and onto an amp.

Help much appreciated (again).



darron

a few thoughts...

you're right. if you're using the opto-coupler in an entirely passive way then the tick shouldn't be 'bleeding' in. THOUGH, a passive square wave (which is what I think you're doing?) modulation can still make a clicking type noise. that shouldn't be happening though when the signal passing through is silent.


does the noise still happen when the signal to the optocoupler is disconnected?

breadboards can be a giant carrier for noise too.

if you did find the noise is still happening just from the module then isolating it with a resistor could help out, since i assume you've got it regulated down from 9v there could be plenty of voltage to drop off.



you didn't really say how you're using the optoisolator to modulate the volume though? is it in an active feedback path? or as a resistor in series or as a resistor to ground? don't forget you might need another resistor to help make up the other side of a voltage divider if you haven't used one.


kind of funny to use an a arduino when two semiconductors can do it too :P but i understand it must be a study project
Blood, Sweat & Flux. Pedals made with lasers and real wires!

tubegeek

#2
Also, just to be on the safe side, let me state the sort-of-obvious: a standard tremolo effect does not modulate the signal on/off/on/off like a switch, it ramps up and down like a volume pedal, even when the effect is extreme like a hard stuttering tremolo.

Clicks in the audio path coupled from another section are often the result of current pulses in the power supply that are large enough/sudden enough to disturb the supply to the audio path. Additional decoupling in the power supply between the two parts of the circuit (control part and audio part) might help.

The simplest form of decoupling is a resistor in series with the supply voltage, followed by a capacitor to ground (RC filter.) Schematically, it'd look like this:



shared supply-> -----+-----vvvvv-----+------> Audio circuit
                    |   about 1K    |
                    |               = about 100uF
                    |               |
                    |               V  <- audio ground, see note
                    |                    
                    +-----vvvvv-----+------> control circuit
                         about 1K   |
                                    = about 100uF
                                    |
                                    V  <- control ground, see note

Note: the node marked "audio ground" should meet the node marked "control ground" at one point only.
This split point should be far enough back towards the raw supply that all current loops that include
audio subcircuits should be closed on their separated side of the split, and all current loops that include control
subcircuits should be closed on their own separate side of the split.




Man, it's been a while since I made an ASCII schematic!

Oh, and good luck with maintaining your parental-omniscience credibility! From my experience, that's a hard task once they hit about 12, but your strategy of claiming specific, limited areas of expertise is (also in my experience) about the best you can hope for. Hopefully the missing "omni-" part of your omniscience will go unnoticed.
"The first four times, we figured it was an isolated incident." - Angry Pete

"(Chassis is not a magic garbage dump.)" - PRR

duck_arse

and to go with the supply bypass as above, you want to run 2 seperate earths, one only for the digital circuits, and the other ONLY for your audio section. they both meet at a common point, obviously, but you want the digital current thumps kept out of the audio grounds.
" I will say no more "

tubegeek

#4
Quote from: duck_arse on September 12, 2013, 10:43:41 AM
and to go with the supply bypass as above, you want to run 2 seperate earths, one only for the digital circuits, and the other ONLY for your audio section. they both meet at a common point, obviously, but you want the digital current thumps kept out of the audio grounds.

Yes!

Schematic slightly revised above to include this key info. Thanks DA!
"The first four times, we figured it was an isolated incident." - Angry Pete

"(Chassis is not a magic garbage dump.)" - PRR

makaze808

Thanks for the replies.

The arduino has it's own power supply. It produces (via the software code) a 5v then Ov state on the pin. This connects to the led side of the opto, and returns to a ground pin on the arduino board. So the digital and audio grounds never meet.

Guitar plugs straight into a jack on the bread board, the ground going straight to the output jack, the signal into the resistor side of the opto and out on the other pin to the output jack. The is no power involved in the audio path. The spec of the opto was described as best for series resistance attenuation.

The ticks were louder while playing, but I think they were still there when no audio input was present, I'll check that again.





PRR

> I bought some optoisolators

There's several types of optoisolators. What exactly have you done? Details!
  • SUPPORTER

makaze808

Last night I placed the filter as shown above onto the 5v from the arduino board and the ticking DISAPPEARED :icon_biggrin: There is a sort of hushing sound but it's quiet and when the guitar plays it is not a problem.

I removed the filter and left the answers posted here on screen for him to see this morning, he quickly placed the filter onto the ticking breadboard and he had that moment of joy when a build that seems to be going nowhere finally works.

While he has realised I am not the electronics genius he thought I was, he was taken aback by strangers helping him sort this out, and he has really grasped what an "online community" is. At a time when he has been having some bullying at school because of his "geekiness" this has lifted his spirits no end.

I hope he keeps up this interest as I have a large box of crap from all my failed effect projects which I might be able to pass on :icon_razz:

Thank you.

tubegeek

Quote from: makaze808 on September 13, 2013, 05:42:54 AMhe was taken aback by strangers helping him sort this out, and he has really grasped what an "online community" is. At a time when he has been having some bullying at school because of his "geekiness" this has lifted his spirits no end.

Boy, did you just make my day! Thank you.
"The first four times, we figured it was an isolated incident." - Angry Pete

"(Chassis is not a magic garbage dump.)" - PRR