ilovefuzz.com

Im new to the group. Thanks for having me. Before I got my first axe in 92 (i was too poor) i built small electronic gadgets for kicks and giggles. I taught myself to read a schematic / diagram. My first effect was a "mid range" meant for a stereo system. There were 3 toggles: on - off - on. So there was high bass, no bass, low bass, high mid, no mid, low mid, and high treble, no treble, low treble. I didn't have a guitar or amp to test it so my friend from school tried it out on his bass. He said it acted as a booster more than anything with a little fuzz - no equalizer effect at all just this unintentional result. Having said that, all fuzz has capacitors, resistors, transistors, and some op amps. What is the most important component to distortion? In other words, when a circuit is complete on the bread board can be swapped out to find the desired sound or distortion you like the most? All answers and opinions are welcome.

el badger wrote: What is the most important component to distortion?

Your ears?

You plug your ears into your breadboard? Do your eyes become oscilloscope screens?

Here's my 'intermediate tinkerer's' stab at an answer: A pedal clips voltage levels so anything that can alter voltage in the circuit can potentially have a profound effect on the sound - I guess that's transistors, diodes and resistors particularly bias resistors.....does that sound right?

The short of it is, every individual component is important and plays at least some role and if it doesn't, that component shouldn't be there.

If you are looking at altering purely the "distorted/fuzz" section of the sound, you need to look at the clipping section like mentioned above.

For just one example, a common clipping method is by using diodes. The choice of diodes determines things like the clipping ceiling and the "threshold", which do a lot to deform the waveform. Of course, you can force more by changing the gain going into it to throw more signal above the threshold. Diodes like an LED for example, tend not to clip until you push the signal hard whilst the very common 1N914/1N4148 tends to clip easier, so by just changing the diodes and their configuration, you get different levels of clipping when you have the same input, therefore a different waveform.

I can use a popular DIY pedal to demonstrate this since that choice makes an astounding effect on the sound if you don't have an oscilloscope, the Bazz Fuss which is pretty much as simple as fuzzes get and the differences can be heard quite easily. If you play one signal into a LED, you may only get clipping on the attack going into borderline overdrive/distortion territory whilst the same clip with 1N914's will have clipping potentially throughout the whole signal generating a more square wave like form, creating something closer to a fuzz like tone. Even with the diodes being changed in this example, the caps and transistor you choose all play a massive role in determining the signal going in, and out of it. This is all before even taking on any tone shaping to alter the frequencies after they have been clipped. With other examples, how you bias the transistor can affect it, so can the signal flow, feedback, octaving, etc. So, so many ways.

I like to list the holy grail of DIY and effects discussion, Geofex as it goes into length about the ideas of distortion and lists many, MANY ways of going about distorting your input.
http://www.geofex.com/effxfaq/distn101.htm

Genuinely recommend having a good read into it and experimenting with how your clipping/gain section shapes the waveform. If you clip it hard so it's a square wave, you start getting very synth-like tones which may sound huge (and yes, they will) but try playing a (non-power) chord with it and you'll notice the harmonic structure starts to become a bit cruel and it becomes hard to distinguish the chord structure in the output so you may want to consider ways to find the distortion musically pleasing. This is where your ears come in handy the most.

Once you have a distortion sound that you like, you can start shaping it with tone controls. There are again, many ways to go about it, from already existing ways such as the standard Big Muff tone control to Bandaxall to whatever you like. I would recommend reading (see, theory) on how tone controls work and understanding some of the mathematics behind it so you can learn how to do low-pass, high-pass, notch filters to shape the sound and understand how all the components come to play. You can add them wherever you like but just remember as a rule of thumb, the more you do, the more signal gain gets lost so you may need to add a gain recovery stage to pickup the slack.

If you can't afford an oscilloscope, I would recommend picking up a copy of LTSpice for your computer. You can use it to simulate tone controls and the effects of your distortions. It's a very powerful tool if you want to tinker with effects and see what each component does in visual form. A bit of a learning curve and it is theoretical but worth it as an analysis tool.

tl/dr. Everything is important. Read up on how distortion/tone circuits work, then sim/breadboard it and swap to view/listen to the effect of the change. Theory and practical come hand in hand.

Hype.

N.O.S mojo components

Goop
Badassness

Tropical fish capacitors and resistors with the worst tolerances possible.

I love this. So it is everything. Electronic harmony. Another question: tolerances - what indicates high and low?

Honestly the most important part of distortion and fuzz is the filtering. The tonestack alone will change the sound WAY WAY more than changing diodes from symmetrical to asymmetrical clipping for example. So many things are circuit specific that it's a very hard, if not impossible, question to answer in a general way.

When I referred to tolerances, I was referring to resistors in my case and it's an easy example to work with. I use 1/4W Metal Film resistors with +-1% tolerance, meaning if you have a 100 kiloohm resistor, it could vary anywhere between 99 to 101 kiloohms at manufactured, unloaded, measured at 25 degrees Celsius. It can vary more after being soldered into a circuit but that's extremely unlikely and requires a whole other load of discussion on the physics of it. +-5% would make it 95 to 105 which is a tolerance of Carbon film resistors. Usually don't want to go worse than that now and for the record it's not fully cumulative, otherwise you would be very unlucky if all resistors were off by -5%.

If you wanted to go horribly bad, you at least used to be able to get carbon film resistors with 20% tolerance which is a huge margin to be off, especially if you've got something like a transistor emitter resistor, which in some circuits, may call for say, a 100ohm resistor and a 120ohm resistor for each transistor. There's a chance that their values could be swapped by tolerance alone but again, another discussion that if you're reasonable and using 1% or even 5% tolerance resistors, you should be okay.

Barracuda is right. The circuit can really determine the priority of each component so it's not a clear cut question. I tried to mention quickly how stages work. If you're looking at defining what you hear as far as highs, mids and lows go, the tone stack is majorly where it is at. You can tinker with each component in there to change things or you can swap the tone stack out entirely for something else to see how that goes. It all depends on what you want to achieve.

I don't know, I'm not a tech guy at all but from my brief experience it seems that the measured values of the used transistors might actually be more significant than which exact transistor is used in the circuit, especially concerning fuzz pedals.

That will do a lot too. The HFe value or, forward current gain.

For first builds, the HFe values is why I would always recommend silicon as the engineer in me prefers the consistency and they are more build friendly. Even if the mojo isn't there, undeniably, many great pedals have been built from them. You can buy a good multimeter that has a HFe reader on it for PNP and NPN transistors and it'll be really useful. It's worth noting that it can vary immensely. Normally, they will specify a minimum on the datasheet, sometimes an average and sometimes a maximum. Most times, a batch of transistors will be really close. Higher HFe is not always better. It is all application specific.

There is another thread here where a user mentioned why a Fuzz Face isn't a good starting point for builders because of inconsistencies and that eternal fight to get the best sounding one. I can completely empathise with that. Of which, it normally involves using Germanium transistors which are expensive comparatively and have inconsistent values, not to mention, current leakage which could mean throwing out a transistor, biasing issues which involve selecting the right resistors based on that. If you want to do it perfectly, it gets complicated and may be a worse result in the end anyway depending on the rest of the circuit.

Best way to discover what everything does before having to test. Get into reading some Electrical Engineering books. You don't need to get ultimately advanced to understand what everything does, at least theoretically. Don't let the maths scare you.

Also, have I mentioned LTSpice? Best way for people that are new to this to understand what component swaps do graphically. Can't help plugging it.

first let me correct OP in the statement that all pedals have opamps. some pedals have tubes or transistors. some pedals have arduinos with no opamps or transistors. there will usually be at least one active device. that is to say there will be some semiconductor performing amplification.

it is possible to discuss this stuff with no technical knowledge of electrical components if you consider three black boxes that perform basic functions. these are

1) volume changes such as attenuation and gain
2) EQ aka filtering
3) clipping. exceeding the loudest volume you can pass through this magic box

pedals that rely on phase differences in signals being mixed together, pedals that use waveshaping techniques that are more elaborate than just clipping the tops and bottoms off, are a bit more rare.

we can sum up the entire guitar pedal industry with a big muff and a tube screamer being %80 of whats out there for distortion and fuzz.

you could add fuzz face, electra, rat, superfuzz, distortion+ to the short list and thats about %95 of all distortion pedals.

this Hfe gets discussed a lot on forums and it is important in some circuits but it is also good to remember that it needs to be in context of what happens next. in my example 1) and 3) are related. more volume (more Hfe) is the same as having less headroom. you can get away just copying someones Hfe because you are both using the same circuit with about the same headroom. if you do not have enough volume, then by default 3) does not do any clipping at all.

tech stuff
Hfe can be altered with resistors and capacitors. Hfe changes from temperature can result in undesirable affects to a poorly designed and/or cheap circuit. you should not shy away from a transistor who's Hfe is too high for what someone told you you need. instead you can add an emitter resistor, feedback resistor or feedback cap to decrease Hfe with the added bonus of increased immunity to temperature changes. you can not increase Hfe without using two transistors so in some cases you actually do need to get the right Hfe.

clipping is mostly done with a pair of shunt diodes in either direction. you can double them up to increase headroom. you can select diodes with very low Vf and If to decrease headroom.