My attempt at design

[demonstar]

I'm getting myself a breadboard to teach myself basic design but until it comes can someone give me a rough indication if I'm along the right lines here for a distortion pedal please? If I'm barking up the wrong tree please tell me. 

Thanks.

[Cardboard Tube Samurai]

Want to edit that image so that that we can view it without a magnifying glass!?

[demonstar]

Sorry folks! Is this better...

"http://aronnelson.com/gallery/Demonstars-Gallery/My_dist_2"

Thanks

[oskar]

Sorry folks! Is this better...

"http://aronnelson.com/gallery/Demonstars-Gallery/My_dist_2"

Thanks

Well the looks off it is a Distortion+ with a tonecontrol.
C2, R3 forms a highpassfilter... The formula for a simple RC-filter is:

f_c=1/ 2 * pi * C * R

where C is capacitance in Fahrad and R is resistance in ohm
f_c=1/ 2 *3,14... * 0,000001 * 56000 = 2842Hz...  Boy, you really don't like base!
change R1 to match a better cutoff frequency. 50 -100 Hz ?

Ok. This formula works if you have the end of the resistor going to ground. It isn't. It's going to a Low impedance DC-source. 4,5Volts provided by the divider and C1. In this case C1 looks rather small for the job. I'ld go a couple of sizes larger.

The gainpot is referenced to ground. It needs a cap to ground to make it work properly. Right now the OP is ordered to amplify what you've got on the other
input which is... Your sound + 4.5 DC. You need to stick a cap to the other end because it will block the DC. Now the DC will see a big resistance to
ground and therefore it isn't amplified. If you take a big cap like 10uF you don't need to care about it filtering audio but you can make it smaller to also help you cut lower frequencies. ( see the rat schemos ). Well that's a start...   

oskar

[demonstar]

Thanks Oskar! I'll go away and do some work on it.

Yes, it does look a bit like a dist+. That doesn't bother me too much because I'm just using it to learn to design and use opamps. The big diffference with this one is I came to it through my maths and formulae not some schematic that someone else had already done. I want to start making very small steps towards self sufficiency. I was kind of trying to design a simple opamp with hard diode clipping pedal with a tonestack so I guess by nature it's going to look like the dist+ with a tonestack.

Anyway thankyou. I'll report back when I've made adjustments.   

[demonstar]

Here we go again....

"http://aronnelson.com/gallery/Demonstars-Gallery/My_dist_2?full=1"

First off I increased C1 to 4.7uF and hopefully that should do the trick. [I wasn't quite sure how to calculate this one.]

I took a look at the high pass filter created by C2 and R3 and I put the numbers in to the equation and got different answer to you  which would make it o.k. Have I done something wrong? Here goes...

1/(2*pi*56000*0.000001)

=2.842052555

Eg. 3Hz which shouldn't be a problem. for a high pass filter if I want it to have no effect?

Th reason I chose 56k was in order to supply the input bias current of 0.08A I gathered I'd need...

R=V/I      R=4.5/0.08    R=56.25   R=56 (nearest whole number)


I understand the reason for putting the cap under the gain pot. I decided I would put a 0.1uF cap there to cut everything below around 20Hz as thats unnecessary to be there as I probably can't here it.

I think thats everything you said and I hope I don't sound silly with some of the questions I ask I'm just very eager to learn. I'm still new to this. I completed my first stompbox 8 weeks ago.

Hows it lookin' now? Thankyou

[oskar]

First off I increased C1 to 4.7uF and hopefully that should do the trick. [I wasn't quite sure how to calculate this one.]

I didn't know how to calculate it either, that's why I suggested something bigger as a safety measure... ( Actually I do know how to caluculate it
but I'm way too lazy. )  Just stick a 47u in there...   Actually I use to generate 4.5V with a spare OP.

=2.842052555

Eg. 3Hz which shouldn't be a problem. for a high pass filter if I want it to have no effect?

Oh, lordy... I'm allways a bit nervous when I hit the "Post"-button, more so when there is calculations involved...
We're both wrong ( At least obviously I am. But I have this settled the first two minutes when I wake up so I don't
need to worry about for the rest of the day  )
It's one zero more in the C...  which makes the cutoff ~30Hz ...really very OK... ( ? right ? )

Th reason I chose 56k was in order to supply the input bias current of 0.08A I gathered I'd need...

R=V/I      R=4.5/0.08    R=56.25   R=56 (nearest whole number)

Here goes my favourite... This is completely correct! If you have a low impedance source...
C1 can store up an electric potential but any DC would not see the cap at all.
So the bias current (because it is DC... ) would actually see first a 56k resistor and then the substitute for R1//R2.
// <--- This means paralell to...   so in the end you've got 56k + 50k ~110k to 4.5V  ( I'm like 86% sure of this )


The gain pot will only allow you to sweep over a rather limited range... make it bigger...

[soulsonic]

You might want to consider using a dual opamp and having a second stage after the tone controls to boost the signal back up. Once the signal has gone through the clipper and then the tone controls, there won't be much left and it won't be very loud. Putting another stage at the end will take care of this and allow you to set a nice low impedance for the output.

[demonstar]

Thanks oscar and good idea soulsonic!

I shall research on the last bit about the 56k resistor as I think I know where your comming from but as for the rest I'm feeling confident. I'm going to swap the order R5 and C8 so I've got a high pass not a low pass.

It's one zero more in the C...  which makes the cutoff ~30Hz ...really very OK... ( ? right ? )

The 30Hz cutoff is great.

I'll consider the gain pot alteration as well. That pot actually allows me to alter the gain from between around 100 to just over 200 I believe which if I'm right is still a canny bit.

I think soulsonics idea of using a dual op amp is great too. Once I've got it on breadboard and know the skeleton of it is there I'll really try tweaking things. I got loads of things I wana' learn how to design too but I guess you've got to learn to crawl before you can walk.

[PerroGrande]

I'm not a big fan of running Op-Amps on a single supply, as there are any number of issues inherent to the procedure.  By decoupling the biasing supply (C1), you've addressed a fair number of these issues. 

The 56K resistor in your circuit (R3) is going to dominate everything else at signal frequencies and will effectively be in the input impedance of this circuit.  To me, unless your goal is to suck some tone from the guitar, this value seems low.

Secondly, the 30Hz 3dB point for the bypass mentioned in a previous message is NOT good.  You want the 3dB point of the bypass circuit to be well below the lowest frequency of the amp.  In other words, I agree with Oskar -- raise C1 to get the frequency lower.   In an article from Analog Devices that talks about various instabilities in Single Supply op amp circuits, they recommended setting the filter created by R1//R2 and C1 to be set with its 3dB pole at 1/10th the lowest frequency allowed by the incoming filter (in this case C2 and R3). I think this is a very good idea.  So if you raise R3 substantially and have a large C1, you'll probably be good to go on both fronts.

Lastly, the addition of a stage that follows the tone stack is a really good idea.  Your first stage has a fair amount of gain than is then trounced by the limiting diodes on the output.  You follow that with a tone stack, which is an attenuate-only proposition.  A gain stage at the end will bring this back up to a more usable signal level and, as was correctly pointed out, make for a nice low-impedance output. 

[demonstar]

I took your advice PerroGrande and I hope I've implemented it well. I've checked the filters so hopefully now I've got the frequencies I want and not the ones I don't. I've chose to use a KA1458 opamp which is two 741 in a dual opamp package. I read up at geofex and found that to bias each opamp with a value which made R3 a suitable value according to Perrogrande (to set the impedance right, yeah?) I had to supply the opamps with a bias current near the max allowed otherwise the bias resistors going to each opamp were colossal in size therefore hard to source. The max bias current for a KA1458 is 500nF. I went for a 450nF bias current. I read the "Designing bias supply networks for effects" at geofex and calculated it using that.

I stuck another section on the end to provide a nice low output impedance and to boost the level up after the passive tonestack and the clipping diodes. The values of R9 and R10 are subject to change once on the breadboard I've just purchased in order to balance the gain stages nicely.

I also swapped the position of C8 and R5 to form a highpass instead of lowpass. Thats the lot I think.

Once on the breadboard I'm going to experiment with different diodes and clipping methods but also with the resistor and pot in the first gain-stage to provide a desired range of gain. I've calculated it and I think it's where I would like it but We'll see!

Hope it's looking better. Here it is....

"http://aronnelson.com/gallery/Demonstars-Gallery/My_dist_2?full=1"

Thankyou once again!

[Johan]

two things catch my eye..the first is the biasing network value..R1 R2 R3 R8 are all quite big..please consider that the opamp needs a little bit of current to remain at the wanted voltage, with this big vaules you might not be able to get a 4.5volt at the opamp outputs..

the other thing is just a suggestion and not at all a problem..
if you put the clipping diods on the treble viper ( just after the tonestack) instead of before that tonestack, not only do you get a slightly stronger signal into the next stage, you will also to some extent be able to tune the texture of the dist by cutting the frequencys before the diods ( and you can then also remove C3..)

johan

[demonstar]

Thanks Johan. Once it's on the breadboard I'll try moving the diodes.

As for the bias resistors, I think they should be O.k. If the stuff I used from geofex is right I should be supplying 450nA bias current into each opamp which is within the spec of the opamp for bias current according to the datasheets. I'll see how it works. I'll pay special attention to that.

Thanks!

[PerroGrande]

I'd still be inclined to raise the value of C1 -- 10 mfd or 22 mfd would be good.   With that value, it would be safe to drop the resistors in the Vdivider to a lower value (250K) and the value of R3 and R8 to something a little lower -- like 2.2Meg. 

Another thing to consider -- the 1458 isn't an FET-input op-amp so it is only going to present an input impedance of around 1Meg.  With the 2.2Meg replacement above, and the existing circuit, you're looking at an input resistance of around 407K which should be alright in most cases.  In other words, the parallel 1M from the pull-down and the op-amp are going to dominate anyway, so reducing the bias resistor to 2M2 isn't going to sacrifice much -- and you will gain some stability.

[demonstar]

Thanks. I've lost you a little bit though as the stuff I read at geofex said that in order to supply the correct bias current and keep the voltage divider stable I need to use resistors that big. I believe if I make them any smaller the current draw out of the the centre of the voltage divider will move Vref about and that I will provide too high a bias current to the opamp.

Please could you tell me how to calculate C1? I'm unsure how to. I know how to calculate the frequency cutoff as part of the filter.