First Attempt At Distortion

[TheShizzle]

Hi there,

Im currently doing a GCSE course in Electronic Products, and for my project i decided to go against the normal bike lights or room alarm and make a distortion pedal, you may call me crazy but hey

ive built the circuit and am having a couple of problems, first one i can solve by rewiring one of the switches to kill the battery aswell as switching the op amp in and out, but i have another i dont know why it happens

when i plug my guitar into it, the distortion effect if loud when the volume pot is either at 2 or about 7, anywhere else on the volume pot and the volume is distinctly quieter, why might this be?

i dont have a circuit diagram handy, but will be able to post one in a day or so

hope you can help somehow,
thanks a lot,

alex

[Joecool85]

What is the name of the circuit you are making?  Sounds like possible poor grounding.

[TheShizzle]

its one designed by me and my ep teacher a bit, just uses an op amp with a capacitor feeding to ground in the negative loop i think, it has a true bypass switching system too, sorry if im talking a load of rubbish, im relatively new to effects n stuff.

how would i rectify poor grounding, what things should i check?

[Melanhead]

can you post a schematic ?

[Ge_Whiz]

Sounds to me like the two outer lugs of the pot are tied together somehow, or both earthed.

[TheShizzle]

heres a circuit diagram, pretty much as it stands, please criticise away!



ive corrected a few of the problems since that design, but havent implemented them, im going to put a resistor between the potmeter and capacitor in the negative loop, any suggestions of value?

also i am going to wire the switch on the 3PDT switch to cut power to the opamp aswell as turn the LED on and off, to kill the hum i get when the battery is still connected and the bypass is being used

my capacitor may seem a totally random value, but i found a nice bag of colourful old capacitors at school, and it seems to work ok, will the value cause particular problems? or the age or something?

the opamp i am using is a TL072 BI-FET Dual opamp

thanks for the help, hope you can help more!

thanks

alex

EDIT: all of the little triangles with no 0v next to them are also connected to the 0v rail, just i forgot to put the 0v on

[Melanhead]

hmmm ... okay, first off I'm more of a tweaker then a theory guy but here goes a few things:

They way the switching is setup up you are switching the parallel diodes in the ground ( 0V ) path and I'm not sure what this does to your ground reference but I'm assuming it's not good, floating ground perhaps ?  ... It's best not to switch ground ( 0v )  in and out at all .. I can only assume that you wanted the diodes to ground on the input of the op-amp but I'm not sure if you'd get much clipping at this point in the circuit because of the low signal level. Diode clippers work well from the op-amp output to ground ( hard clipping, more like distortion ) or in the feedback loop ( soft clipping, more like overdrive ). I'm not sure but shouldn't this have a vref ( 4.5v ) at the opamp input ?

I'm sure that someone with a lot more theory will come along with more answers ... anyone ... ?

they way it's set up now I'm assuming you may be getting level boost but no clipping/distortion ?

Go here: http://www.geofex.com/ and check out "The Technology of the Tube Screamer" article. It's a good read and may give you some ideas. 

Good luck ...

[TheShizzle]

at the minute i tried to get two different switchable bits in the circuit, as i have two switches, but i jsut want somethign to slightly modify the soundas the second option, my teacher will seriously not like it if i try to include two op-amps, so i put the diodes in to get some sort of very soft modification, and it is a very soft change, but there is a change and thats all that will matter for my marks

at the moment i am getting a kind of distortion, not particularly extreme, but soft, due to the capacitor in the loop

basically im not overly bothered about having a pro quality sound, its my first go and its aim is to get marks for working electronics not nice sound, if it doesnt do much i guess i could switch the diodes in and out of the negative loop on the opamp instead of in the 0V path, would that work?

what things ought i watch out for in the circuit to prevent hum etc? any changes need making, other than switching the diodes in and out of the loop instead, and would there be any reason to explain my previouslydescribed problem whereby at two levels of input, one low and one high, it had much more gain than the inbetween levels, and should i have a smaller or larger resistor on the input of the opamp to prevent the total loss of noise when the input gets to full on my guitar?

[grapefruit]

Don't take this the wrong way, but there's so many things wrong with your circuit.

Yes, you need half rail bias on the + input of op amp.
Output volume is wired incorrectly.
Diodes wired incorrectly.
Feedback loop wired incorrectly.
Feedback capacitor too small.
Don't switch the GND.

Rather than try to explain, take a look at the MXR Distortion Plus schematic.

[scaesic]

 you need to have the positive of the op-amp biased to 4.5v. and you need a cap and resistor to ground on the output and input to isolate the 4.5v dc from any circuits connected to it.

you need to bias the op-amp to 4.5v, which you can make with a simple op-amp configurationf rom the 9v. if you think about it, the a.c signal is oscillating around ground, if the bottom rail of the op-amp is ground, then the negative part of the wave will be completely clipped, biassing the op-amp to 4.5v will resolve this. The capacitors/resistors you need to put in are to isolate the 4.5v from anything connected to the pedal.

Might also want to consider putting a unity input/output buffer, which is just an op-amp wired as a buffer, this just helps to pass the signal along the line more completely, otherwise you end up loading things, and it'll all go wrong.

putting the back to back diodes over the feedback circuit of the opamp will make the amplified signal clip as it reaches the clippinf threshold, im not sure why they are where they are, or what any of that switching is doing, to be honest, but you don't need nearly so much. you need a 3pdt at most, one pole for led, and 2 for bypass if you're going true bypass.

take a look at the technology of the tube screamer here http://www.geofex.com/Article_Folders/TStech/tsxfram.htm

take a looke at "What are all those parts for?" at GEO.

i suggest once you understand these two articles (print them off and go through them with your teacher). you redesign the whole thing.

[scaesic]

at the minute i tried to get two different switchable bits in the circuit, as i have two switches, but i jsut want somethign to slightly modify the soundas the second option, my teacher will seriously not like it if i try to include two op-amps

you already have 2 op-amps, the lf412 is a dual op-amp, theres 2 of those babies in there. thats why theres 8 legs.

[scaesic]

and would there be any reason to explain my previouslydescribed problem whereby at two levels of input, one low and one high, it had much more gain than the inbetween levels

a broken potentiometer.
or a broken op-amp.
or a dodgy guitar volume pot.

[TheShizzle]

thanks for the help, im aware now that my circuit is total bollocks as it stands, no thanks to my ep teachers cursory glance

i think i can modify most things, tho i still dont get the 4.5v thing, i looked at the MXR Distortion pro schematic

is there a way i can make the circuit cope without an input buffer? if i make the distortion just based around one opamp, and using the biasing, would this make the circuit more, er, stable for want of a better word?

thanks a lot for the help, i really appreciate it, tho i really would like it if someone could explain why the 4.5v bias is needed and what it does, and how to get it

heres a new circuit diagram of it, i dont know how to get the 4.5v tho, would this work as a minimal solution?

[aron]

Take a look at the Shaka Braddah:

http://diystompboxes.com/pedals/shaka.JPG

Take the circuit up to the 2.2uF cap (ignore the resistor with the ? before it). Wire up a volume pot after the 2.2uF and you are done.

4.5V is the snippet of the circuit above the mail circuit on the left.

Of course read this:

http://www.diystompboxes.com/cnews/FAQ.html

[scaesic]

that circuits a lot better, first shift the diodes(in parralel) in parrallel over the feedback resistors.(see the shaka) although your missing the caps/resistors to remove the d.c on the input and output lines, compare your circuit to the link aron put up and you can see where the caps/resistors ont he input and output are needed - the 0.01uF on the input, the 4.5v is brought into the line via a resistor which you need to add. also the 2.2uF cap which aron mentioned, that removes the dc after the clipping op-amp (it is to stop pops while switching, and also adding a 4.5v dc from wondering into your guitar/pedals before it  or other amp/pedals after it.)

now to why you need a 4.5v line.

you have your guitar signal bouncing around ground, it's an ac signal.

it goes into the op-amp, which has power rails at 0 and 9, ok?
right im sure you must know what happens if you now put in a signal above 9v, it clips, similarly, if you put a signal below 0 in, itl clip.

so as it stands your guitar signal (which is a.c about 0v) will only be amplified for signals greater than 0, half your guitar signal is lost.
if you add a 4.5v d.c to your guitar signal, the a.c signal is now centred around 4.5V. Now the signal sits in the centre of the power rails, so both the positive and negative sides of the a.c wave are amplified.

ok, thats that sorted, but now you have your a.c signal offset by 4.5v. you can remove this by putting the signal through a capacitor, as dc cant get through caps, also after the cap you should put a resistor joined to ground, so the signal know's where ground is.

you should put such capacitors on the input and output line of the whole circuit.

now on to clipping, as it stands im pretty sure your diodes are doing nothing, put them in parrallel over the op-amp feedback circuit, and as it amplifies the signal, it also clips the top and bottom of the signal (clipping using only an op-amp is possible using the method i described above, but diodes in the feedback circuit is the most common/pleasing way)

to get 4.5V, do this     
where R1=R2, you will get a voltage divider that drops half the voltage in between the resistors, shove that 4.5v through the + of an op-amp and you have a usable 4.5V line.

-----------------|-------------------9V
                   R1
                     |____4.5v      you can put an op amp here, or just copy the shaka can one, putting the 4.5v into a + input of an op amp which has the
                     |                   negative tied to the ouput will stop you                                                   
                     |                   loading the circuit, but the shaka one will be fine too probably.
                    R2
                     |
                    ground

-capacitors are also seen sometimes before r1 to ground, just look at some schems.
this will created a 4.5 line that you can use anywhere in the circuit without loading. If you ommit the op-amp, you'll load the circuit and it wont be 4.5v anymore (you can view an op-amp as a box which means that nothing before it effects anythign after it, that is partly why it was designed). This is pretty essential, the clipping you're getting right now is probably on the acount that the negative half of the guitar signal is being completely lost. The diodes arn't doing anything. If you don't add the 4.5 line and bias your guitar signal, the circuit isn't doing anything that it's designed to do (even amplifiy the signal like an op-amp should) and you'll likely fail completely if you are assessed by someone who know whats going on.

you can do without an output buffer, but considering how simple it is to put in, and the magnificent difference it'l make to add an input buffer, you might be worth just sticking one in. I'm sure you're teacher won't mind you using more op-amps, you need at least 2, one for the clipping part and one for the 4.5v line, and preferably 3/4 if you add an input/output buffers.

for switching, theres a nice article on off-board switching somewhere, i think its on general guitar gadgets.

[scaesic]

p.s if you add an input buffer, it'l be a perfectly useable distortion box.

[Steben]

Bwaaaaah, an opamp circuit shouldn't need an input buffer, since the input impedance is set by bias resistors (much more different than with singe-discrete stages...)

[scaesic]

Bwaaaaah, an opamp circuit shouldn't need an input buffer, since the input impedance is set by bias resistors (much more different than with singe-discrete stages...)

bias resistors are pretty small, compared with the inpute inpedance of an op-amp.

[TheShizzle]

THanks a lot for all the help guys, will this version work as a basic distortion, even if the effect wont be really good quality, cos i really ought to start getting it built again, having only 3 weeks next monday to the deadline lol



i dont really want to go into input and output buffers using op-amps, but as i have a dual op-amp i might aswell use it to make my 4.5v input (thanks for the explaination, i get it now, my teacher really knows nothing about audio circuits obv, and the eoftware we use to sim it at school sucks)

in the shaka it has a cap parallel to the diodes, is this needed to get an effect? and what does it do if i put it in?

[scaesic]

it forms a low pass filter i think, cuts out all those high frequencies.
http://images.google.co.uk/imgres?imgurl=http://www.radio-electronics.com/info/circuits/opamp_low_pass_filter/op-amp_non_inv_withrolloff.gif&imgrefurl=http://www.radio-electronics.com/info/circuits/opamp_low_pass_filter/op_amp_lowpassfilter.php&h=890&w=500&sz=8&tbnid=E_WuxJxnPJgezM:&tbnh=145&tbnw=81&hl=en&start=2&prev=/images%3Fq%3Dhigh%2Bpass%2Bfilter%2Bop%2Bamp%26svnum%3D10%26hl%3Den%26lr%3D%26client%3Dfirefox-a%26rls%3Dorg.mozilla:en-GB:official%26sa%3DN
not sure if this is being used for tone, or to cut out radio frequencies.

p.s you'll want to bias the + input via a resistor, just to limit the current going in, at the moment you have it going straight in.you can see this again on the shakka.