First distortion

[KimJongFun]

Hello everyone,

I'm new to pedal building (and also new on this forum) and instead of modding an existing pedal I wanted to design my own, so I ended up making a two gain stages preamp (IMO pedals work a little bit like preamps and I will probably use this pedal as a preamp anyway) using op amps, I breadborded it and it sounded alright.

The breadborded version was only using 9V (I don't have the LT1054), powered by a battery, had no input and output buffers (are they even needed? I was using a 1MegOhm resistor instead of 10k just like the Tube Reamer from RunOffGroove), and I only used TL072 (only what I had available), and the active EQ wasn't included (because of lack of space... and also lazyness/massive headache)
I simulated both parts of the schematic using LTspice and everything seems to work fine, the only thing LTspice doesn't give me is the orientation of the electrolytic caps.

The schematic:
The first stage is obviously a slightly modded OD808 without gain and tone controls, the gain is fixed and the tone was simply removed, with a 0.033uf cap instead of the stock 0.047uf, the reason: less basses = tighter sound, but I also wanted to be able to choose to have a thicker tone when I wanted, so I decided to have a 3 way switch to choose between 0.082 0.056 and 0.033uf, which can provide thicker tones without turning the pedal into a fuzz (but maybe I will put a 5 way rotary switch because I kinda love fuzzy tones sometimes)
The second gain stage is a simple distortion with asymetric clipping.

The EQ stage is made of two parts, a "simple" active EQ with +/-12db of gain (8db would have been more than enough), with a mid scoop (maybe there are other and better ways of doing it BUT I couldn't find any website which explained how I could have a 450hz 0.7Q -5.5db scoop... I took inspiration of the pre&postEQ stages of the boss MT2)
The bass control is set at 100hz, Treble control at 2,7Khz and a Presence control (high shelf) at ~5khz
The second part is a Wien Bridge Mid control with selectable frequency between 220hz to 1,2khz.
the frequencies were chosen inside my DAW, they are the frequencies I usually boost or cut, I couldn't decide for the mid frequency, so I ended up choosing a sweepable mid control.
And now that I think about it I wonder if a simple passive EQ wouldn't have been enough... but I can't design any that would work the way I want to I ended up with an active EQ.
During my testing I noticed there was a lot of mids, so I decided to add that mid scoop switch, so I can still use the mid knob for more shaping.
I made a -6db attenuator before the EQ section because it was too loud... (yeah I know there is a volume control at the end, but I prefer to have proper volume output when the volume control is set at noon)

During the LTspice simulation I noticed the curves produced by the active EQ were quite ugly using 10kOhm pots so I ended up using 50kOhm which fixed the issue.
The Wien bridge also use a 50k pot, simply because I don't want to orded a single 10k pot :C

I chose 2N5089 for in and output buffers as they are supposed to be low noise and high gain (to have an output closer to unity gain)... I haven't breadborded the buffers as I only have 2N3904 (gain too low I suppose).
The OpAmp were chosen for their tonal quality (OPA2134 for the gain stages) and very low noise levels (LME49720 for the EQ stage buffers which is also very transparent, the OPA2134 also has lower noise levels than the TL072).

After testing I feel like it's missing something, it lacks some "oomph", a larger cap for the tube screamer stage didn't provide what I was looking for, so I guess I should probably add a clean boost.

(click to enlarge)


Any thoughts?

PS: I hope that post is easy to understand, english isn't my native language and I tried my best to order my thought in a logical order without talking about pointless things for too long.

[EBK]

Any thoughts?

PS: I hope that post is easy to understand, english isn't my native language and I tried my best to order my thought in a logical order without talking about pointless things for too long.

Your English was perfectly fine.  The problem is, it's really tough to describe a "sound" in any language, so I'm not 100% sure what you're after, but that's ok.   
Next time, lead with the schematic, then describe it.  I was trying to follow along, piecing it all together in my head before realizing about halfway through that you drew it out.  Perhaps I'll reread it all later when i have time.  No gain control struck me as an odd choice....

[thermionix]

No gain control struck me as an odd choice....

He's got a gain control, but not where I would have expected it.  I see the input buffer and clipping stage of a tube screamer (basically)...after that I'm lost.

[EBK]

I'm slowly taking another look at this.  It is quite remarkable in its complexity and quirkiness (interesting oddness) for a "beginner" pedal project.  I am having a hard time framing my thoughts into meaningful feedback based on your description of yourself as being new to pedal building while obviously knowing your way around breadboarding and simulating this stuff.  Usually, when we see someone new asking for feedback on a first project, we see something entirely different, as you can probably imagine.    I forgot to say this before: welcome to the forum! 

[KimJongFun]

No gain control struck me as an odd choice....

He's got a gain control, but not where I would have expected it.  I see the input buffer and clipping stage of a tube screamer (basically)...after that I'm lost.

Yes, the gain control is in the second gain stage, as I said in my previous message, I tried to mimic an amplifier drived by a tube screamer, the first stage is only here for tone shaping with a slight overdrive in it (maybe I should try without the diodes...) and the second stage is to control the gain, BUT now that you say it I never tried to control the gain in the first stage and see what it does... but from what I gathered arround, it would change the response slightly because of the EQing, am I right?

[EDIT: to remove any confusion, the response for the question is for you, because I know what the gain of the first opamp will do in the schematic, and I knew what I was looking for, increasing the gain in the first stage will mostly increase high frequencies (above 1k with the 0,033uf capacitor) and the diodes in the feedback loop will clip the signal, same for the diodes in the next stage resulting in a different kind of distortion]

It's:
transistor buffer > opamp overdrive (tube-screamer with fixed gain) > slight filtering > opamp distortion (with variable gain and asymetric clipping to ground)> attenuator > opamp buffer > active EQ > Wien Bridge Filter > passive volume > transistor buffer without the tube screamer filtering.

To make things short:
it's an overdrive section then a "tube-like" asymetric distortion with a 4 band EQ with variable mids and a mid scoop switch.

I'm slowly taking another look at this.  It is quite remarkable in its complexity and quirkiness (interesting oddness) for a "beginner" pedal project.  I am having a hard time framing my thoughts into meaningful feedback based on your description of yourself as being new to pedal building while obviously knowing your way around breadboarding and simulating this stuff.  Usually, when we see someone new asking for feedback on a first project, we see something entirely different, as you can probably imagine.    I forgot to say this before: welcome to the forum! 

Thank you, it probably started when I was 17 and wanted to mod every pedals I owned (but I didn't)
No, but seriously, it was really difficult to gather all the informations and to understand how everything worked, simulating various circuits with LTspice helped me a lot, because I didn't have all the components at home, and being able to switch from theory to (simulated) practice answered many question I had while also clarifying some electronic websites I was reading.
Breadbording was confusing at first, but I learnt how to use a breadboard with internet, like everything else in electronic, and I had a basic version of that pedal, with no EQ and no buffers, and I don't think I could breadboard it entierly.

it's not really my first "pedal", I abandonned my first pedal project, which was inspired by a sunn beta bass amp but change the response of the gain controls of each channels and adding another band to the EQ section (I bought the sunnbather from abominable electronic but I never opened it... so the two channels were remade from scratch using heavily modded big muff for each channels instead of using CMOS...)

Same with my second project which was an advanced blender with high and low pass filters, it could have been made easier because I didn't properly know how opamp worked back then, but even if I did, it would have been too dificult to make, because I needed Quad unit potentiometer to have the response I was looking for (to make variable 12db/octave high and low pass filters instead of 6db/octave) and quad unit potentiometers are very hard to find.

And before this distortion I modded a boss HM2 (I only had the schematics), this time it's not an original circuit, so I don't take much pride, I changed a lot of values to add more basses or make the distortion less harsh, I also saw that there were really weird design choices in that schematic, I changed those parts to make them easier and it still worked fine, I couldn't find any pnp transistor so I converted the second boost&filter stage to npn with LTpice I saw that it made no difference at all...
Then I changed the soft clipping to Mosfet clipping.
I also made my first 4 band active EQ on that pedal
It sounds nothing like a HM2 but now it's a very cool bass fuzz (and since it's finished I will probably make a post about it here later).

I saw plenty of forums were beginners wanted to make a specific pedals for their first project but were advised to make something easier (a DIY booster kit in most cases), so I simply googled generic terms related to electronic or features in guitar pedals and sent emails to some pedal makers asking them about stuff they wrote on their website and some of them were kind enough to talk a little bit about it.
I also saw some videos because I was confused about the choices for opamp in guitar pedals, so at the end I only chose part that would make the less noise and color the tone the less (especially for the EQ section, the BurrBrown Opamp for the two gain stages it's because I've heard some comparison, and even if the differences were minors in tone with other opamps it's supposed to be less noisy than a TL072...).

Anyway... Thanks again.

[stringsthings]

Interesting circuit!  Very advanced, IMO.  Would love to hear some soundclips.

[KimJongFun]

Interesting circuit!  Very advanced, IMO.  Would love to hear some soundclips.

Thanks for the interest !
Well from the responses I got, I guess I will have to record some soundclips, I also have to buy some stuff before I can make a proper demo with the finals values and with the whole schematic (with the EQ section), so it will take some time because I also I don't have the breadboarded pedal anymore... I had another project and other stuff going on in my life.
But it worked and the simulation is working too, so I'm confident with the EQ section, I will probably make a prototype PCB to test the whole thing.

EDIT:
I changed the schematic a bit, no need to post the new version, the modifications were minors:
I can't remember when I started working on the schematic, so there are some design choices I don't remember why I chose doing them that way instead of another (plus some values were chosen by ear after testing on the breadboard).
Anyway:
Swapped R3 and C5 I gained 0.1dB in LTspice ... (inaudible but hey, it's more efficient)
Removed R9 R10 and C12, it was a 1dB attenuation with a low pass filter (to remove harshness), I have the same results with a bigger cap in C10 like roughly 94pf (which I found using LTspice) ... I could use two 47pf caps, but is it critical? not at all, 100p is fine.

If anyone wants I can explain every parts of the schematic.