Creating own circuits: How?

[Colossus]

Hey there,

I've already built some pedals according to existing circuits. Now I wanna build my own Fuzz-Stomp, but I don't know which component has which effect to the sound. Is there any software-emulation of circuits so I can hear the outcome of it? Where should I begin? Any tips? I already tried to search, but found nothing. Thanks in advance!

[darron]

hit search. there's an abundance of info (:

personally i try to read as much of what goes through this forum as i can as there are a lot of pedal gurus here!

http://www.diystompboxes.com/smfforum/index.php?topic=47235.0

[DougH]

Easiest way to get started is to breadboard an existing circuit, tweak values of components, make mods etc and learn how each section affects the sound. If you are still curious after that, starting learning some electronics- lots of tutorial links in the link section and wiki.

[earthtonesaudio]

Is there any software-emulation of circuits so I can hear the outcome of it?

That would be some sweeeet software.  But sadly, I don't think it exists.

[frequencycentral]

Get a breadboard and some jumpers, build a circuit you know you like, change the component values and see what difference each one makes.

Then do another circuit the same way.....

.....and another.....

.....and another.....

.....and another.....

.....and another.....

.....and another.....

Pretty soon you'll have a good understanding.



Hey, I just noticed that this was my 1000th post. Nice round number, think I'll stop posting now.......!

[petemoore]

  Happy 1,000th FC !
  Doug and FC hit it pretty good, build and read, then repeat.
  Finding the 'sub-info' [the kind that everyone else seems to have swum through, and doesn't seem to be on the surface so much].
  Get a little [say 12 parts or so] active stage goin', like LPB.
  Then diddle with it and see what the DMM and speaker tell you about:
  The first capacitor value [changes the voicing if small enough].
  The bias on the transistor [what happens when the bias resistor values change?]
  Every part in the circuit does something, swap the transistor..etc.
  Try it at lower voltage [nah].
  Build you a Jfet and read GEO a buncha times.
  Get your reference data to set next to your other data and schematic [when necessary, dig up 'the rest'...whatever it takes to try hacking through some of the theory.
  Attack the math head on, and you may develop better 'actual math' skills than myself.
    You will tend to find [if history serves], that treading across extremely well worn paths is the norm, re-inventing the 'gear' or 'wheel' is an abstract brick wall many have run headlong into, and tried to find ways around [thinking that a whole new topology can be found], not sayin' it can't, just sayin' it's probably been tried.

[bipedal]

Check out Cook Your Own Distortion.  I found it to be a useful overview...

- Jay

[shadowmaster]

My suggestion. The best way to start is by reading this...

http://www.diystompboxes.com/wiki/index.php?title=Simplemods#Simple.2C_Easy_Mods.2FTips_.26_Techniques

Check gain stages of different types of distortions (op-amp-based, transistor-based, cmos inverter-based, jfet-based, 386-based etc...).

Check different types of tone controls.

Check different clipping methods.

Maybe check out some buffer circuits too.

Mix and match to your own liking.

[Gus]

Read books
Go to school
Keep reading books and app note and.....
Build stuff
Keep reading
Check/test what you read
Buy test equipment
Keep reading
Check/test what you read
Why would it be easy to design well without a good understanding?

[dschwartz]

Read books
Go to school
Keep reading books and app note and.....
Build stuff
Keep reading
Check/test what you read
Buy test equipment
Keep reading
Check/test what you read
Why would it be easy to design well without a good understanding?

i agree with that..reading and understanding something one day at a time..i keep a pile of articles, books extracts, educational posts and schematics in the bathroom, and i read them over and over everytime i go the bathroom ..i have learned a lot that way!!

my process to create a circuit is this (assuming you understand electronics)
step one: Have an Idea or need.
step two: search the web different ways to implement the idea..
step three: make the circuit and simulate it with LTspice..it lets you input a .wav file and listen to the output, but it´s never like reality, trust in transient and frequency response simulations..
step four: breadboard the circuit and tinker with values..
step five: make the layout and PCB
step six: build the definitive version..

[mojotron]

Read books
Go to school
Keep reading books and app note and.....
Build stuff
Keep reading
Check/test what you read
Buy test equipment
Keep reading
Check/test what you read
Why would it be easy to design well without a good understanding?

I agree: There is a lot going on in all but the most basic circuits. IMO, You only get out of it what you put into it. The problem with something like the software that you asked about is that it would limit a person to only what the software would be able to do - and it would be obsoleted in days. Circuit design is really an engineering activity - there's a big difference in what it takes to build effects and what it takes to have a sound in mind and then design something to make that sound. There is enough to learn that it will take years to really get a good understanding of electronics - there is not shortcut because "it's not about the destination - its about the journey" (as Mark Hammer mentioned in a thread some years back - which hit home with me...) as you learn you will realize that it takes input form others and working with other people's ideas to really get your own ideas to work out.

I don't think designing circuits really takes a specific academic background - just a lot of interest on the designers part.  What makes this much easier than ever is this forum; keep building, reading, tweaking, searching and figuring things out and you will find that if you really love electronics you will get good at designing things.

[Colossus]

First BIG thanks to all of you!!! Haven't thought there be such a respone - really great. So I think I  go for some more pedals and make maybe a dummy circuit, where I can just change some parts.

Is there any TOTAL basic scheme for an overdrive or a fuzz? Where really just the ESSENTIAL components are on?
greetings,

[jacobyjd]

I'm not sure if you have a breadboard, but that would be the best place to start if you want to try development on your own.

Here's a link to a breadboard project from BeavisAudio (it's a PDF, just a heads-up): http://www.beavisaudio.com/bboard/projects/bbp_BoutiqueTubeScreamer_Rev1_1.pdf

This is a tube screamer without buffers. It's about as simple as you can get when you think of a traditional overdrive.

[petemoore]

Is there any TOTAL basic scheme for an overdrive?
  Hookup a teeny little speaker, say 2w thing to your 20w amplifier, try to eak as much overdriven speaker tone as you can before the speaker coil bakes, the speaker should exhibit distortion characteristics when driven past it's...it's capability to produce sound that isn't distorted. The amplifier with ability to push the coil to where it's pinned to the limits of the cone suspension during excursions will produce a 'distorted' or 'overdriven' tone, two descriptive words that can be used to describe any electronically or other sound, anything from 'Guitar with a severe Fuzzbox Distortion' or 'ultra-clean stereo' componentry which almost perfectly replicates whatever it is amplifying, all are capable of fitting the extremely generalized terms 'overdrive' or 'distortion'...'accoustic' is what is outside the meaning of the super-vague-description terminologies.
  The amp should have sufficient drive to blow the speaker, the speaker won't blow unless it's 'over-driven', somewhere between there the speaker will produce sounds as it is overdriven, producing 'distortion'.
  A 20watt speaker driven by a 2watt tube amp...if HB pickups don't drive the amplifier to distort, boost the input voltage swings so it does, the speaker should sound as clean as a speaker can sound in these conditions, but now the amplifier tubes can be pressed into distorting the electronic, AC input/source waveform. Intentional terminology usages here...
  3. Make the amp big enough headroom to be 'clean' [another general and vague term], also attach enough speakers to be 'clean' [no such thing but for discussion with only silly words avaialable] so that the sound produced from the speakers fairly replicates the waveform of the input signal [clean sound reproduction, relatively clean enough to call it 'clean'], no overdriving, no distortion as a result of overdriven anything...pure clean with extra headroom the whole way...
  Now distort the electronic signal before the amp input [such as a fuzzbox between guitar and amplifier].
  A Fuzzbox needs at least these two things: Boost and limited headroom.
  Boost can be 1 transistor or opamp stage, limiting headroom can be done in a large number of ways.
  Put diodes which shunt signal levels over X amount [ie the AC signal swing supercedes the diode clipping threshold voltage]...wave in channel analogy:
  Make a wave large enough to hit the ceiling go down a channel, before the wave hits the ceiling it looks like...itself, while the ceiling is clopping the top off of it, it's shape is altered...distorted, same thing as boosing an AC waveform and clipping it with diodes...smaller threshold voltage diodes 'move the clipping wall down', making the wave even more distorted looking, or, boosting the wave amplitude [making it bigger] has similar or same result in making the wave look more distorted. A small enough wave can able to pass under the clipping ceiling wall, to emerge undistorted after the clipping diodes.
  Another way to introduce a ceiling which the boosted waveform can run into [causing distortion of it's original form] is to have headroom limitation which causes the boosting activity [say...transistor, because commonly used this way], if the AC signal input wave is boosted to large enough voltage swings, and the headroom is less than the voltage at the peaks [- and +] of the waveform  [because supply voltage is small], the transistor cannot 'produce' voltage so the signal swings as far as it can toward a peak, but can't reach it because the input signal is greater than the power supply. [transistors basically just switch supply voltage and try to make a larger AC voltage output look like the AC voltage input, using DCvoltage as power supply. 
  So take your basic waveform [AC], boost it [larger voltage AC swings], clip it [put diodes with low threshold voltage, low enough so that they conduct some of the input to ground...in the case of a Distortion +]. Dist+ has an opamp boosting AC level  which is then 'hit' with clipping diodes, this 'clips' off the peaks of the AC signal waveform swings/distorts the waves form.
 

[MikeH]

Start by reading any article at geofex.com with the word "technology" in the title.

[DougH]

Is there any TOTAL basic scheme for an overdrive?
  Hookup a teeny little speaker, say 2w thing to your 20w amplifier, try to eak as much overdriven speaker tone as you can before the speaker coil bakes,

Drill holes and/or install pop rivets in your speaker cone. That's totally basic.

[bipedal]

Drill holes and/or install pop rivets in your speaker cone. That's totally basic.

Nice!  A variation of The Kinks' original approach to distortion, if I'm not mistaken...

[asfastasdark]

What I have kinda seen as a pattern is, for one type of distortion/OD is one or more (usually 1-3) BJT's that amplify each other with V+ to collector, emitter>resistor>ground, and basically using resistors to bias transistors and capacitors in conjunction with those resistors to form LP and HP networks.

[newfish]

+1 on the 'Breadboard and existing diagram' path.

You could do a lot worse than going through the 'Technology of the Fuzz Face' article and building each circuit in turn.
This isn't as laborious as it seems - once you've got you first creation working, the sense of satisfaction is almost as great as the 'now what can I do to it?' feeling.

Why not build the simple Fuzz Face (again in the article...) and use trimpots for the 100K feedback resistor, the 33K 'load' resistor etc?

You get to understand why these values were chosen - and what happens when you monkey with them.

I'd propose renaming the article to 'Fun ways to spend your Jetlag' after this weeks' performance, but I know I'll still be tweaking (in a fun, wholesome way) well into next week.

A little (or as much as you feel the need to do) bit of background reading never hurt either.

The sources are out there - Public Libraries are great - not just the Internet...

[zyxwyvu]

Is there any software-emulation of circuits so I can hear the outcome of it?

That would be some sweeeet software.  But sadly, I don't think it exists.

This is a good idea. I'm on it.