Play yer guitar and shut up...

[Paul Marossy]

OK, now that I have your attention...  :wink:

I was looking at the waveforms on my scope from my resurrected BSIAB the other day. (I had problems getting it to work properly and put it away for a while). My circuit is basically per the schematic, except that I used 1M for the drive pot and a .0033 cap for the one that "changes the tone".

I fed it a 1kHz sine wave w/ 10dB attenuation from my function generator and looked at just the output waveform at the output jack. (I will do more poking around later). I was expecting to see a clipped waveform , but instead I just saw really tall, more or less sine waves when I turned up the gain to max. When I turned it down low and got more of a SRV tone, it looked like the classical tube overdrive with gently rounded corners. Turning the gain control also changes the amplitude of the signal.
I tried other frequencies between 80Hz and 3kHz, but they all looked more or less the same, or at least I saw what I expected to see.

The tone control was the most interesting to watch. It really manipulates the waveform, from a shark fin looking affair to an almost perfect triangle wave with a slightly rounded peak. The overall amplitude is also affected.

So, aside from it being generally interesting findings, it made me think a little bit. It seems like what I was seeing was more compression than distortion, and at low gain levels, it looks more like an overdriven tube amp. So, where does one make the distinction between overdrive and distortion? Or fuzz? Those terms seem to all be used interchangeably, but it seems to me that one could see things clearly with a scope. Is distortion more of a hard clipping thing, and overdrive like what the BSIAB does?
And on the tone control - when the highs are cut, I get a triangle wave. Does that mean that a triangle wave is devoid of harmonics, so it just ramps up sort of linearly? And I have heard that square waves are rich in harmonics... So does that mean that fuzz type circuits have a lot of odd order and even order harmonics? It would seem so from listening to them.

All of these things seem to hold true for me so far, but I'm just trying to understand what I have been observing on my scope a little better. My ears don't always correspond with what I see on a scope, except for I can hear odd order harmonics pretty well...

[RobB]

Does that mean that a triangle wave is devoid of harmonics, so it just ramps up sort of linearly?

The only type of wave which is devoid of harmonics is a simple sine wave.  Sine waves are in fact the building blocks of all other waveforms.  If it isn’t a simple sine wave then it is referred to as a complex wave.  The branch of mathematics, which demonstrates this, is called Fourier Analysis.  

http://psych.hanover.edu/Krantz/fourier/square.html


As a general rule, complex waves which have pointy bits to their shape have higher degree of odd (dissonant sounding) harmonics.  Complex waves which have a smoother outline have more even than odd harmonics (pleasant sounding).

As for where overdrive stops and distortion begins, I think it’s a matter of terminology rather than one or the other.  Technically, anything that changes a signal is a distortion.  Even tone controls such as high cut, mid boost and low cut are changing therefore distorting the signal.  Overdrive is a distortion associated with putting a too big signal through an amp/opamp/preamp/speaker etc.

[WGTP]

Interesting observations.  I have a Sweet 16 that is sort of a higher gain BSIAB w/ a different tone control and it definitely sounds like it is distortion as well as compressing.

To me a Fuzz is a definite box on the floor sound that I associate with the late '60's.  Wooly, Nasty, typified by the Fuzz Face.  At the same time I have modded the FF/Tone Bender circuit to sound more like a distortion and so have many others.

IMHO distortion is supposed to be more Tube Amp Like, like a Raging Marshal Stack, to use Aron's phrase, but some sound pretty fuzzy.

Overdrive I see as being more smooth, less hard edged, or maybe even clean, but used to OD a tube amp.

That's how I conceptualize it anyway.  

[WGTP]

Interesting observations.  I have a Sweet 16 that is sort of a higher gain BSIAB w/ a different tone control and it definitely sounds like it is distortion as well as compressing.

To me a Fuzz is a definite box on the floor sound that I associate with the late '60's.  Wooly, Nasty, typified by the Fuzz Face.  At the same time I have modded the FF/Tone Bender circuit to sound more like a distortion and so have many others.

IMHO distortion is supposed to be more Tube Amp Like, like a Raging Marshal Stack, to use Aron's phrase, but some sound pretty fuzzy.

Overdrive I see as being more smooth, less hard edged, or maybe even clean, but used to OD a tube amp.

That's how I conceptualize it anyway.  

[Paul Marossy]

I guess it's kind of hard to see what is really going on without a spectrum analyzer (that's what they are called, right?)

I know that a sine wave is the building block for all the other wave types, and I remember reading that it is the simplest of waveforms. That would imply that it has the least amount of harmonics.

So, is it possible to see what harmonics are present in a waveform without a spectrum analyzer? If so, how do I do that with a scope? It kinda looks impossible from what I can tell...

[RobB]

That would imply that it has the least amount of harmonics.

A perfect sine wave contains no harmonics at all.

There may be some free scope type software out there, which will also do a spectral analysis of a sampled waveform.  Haven’t looked but if something like that exists it would be your easiest option.

In a practical laboratory session at college (many years ago) we did it like this.  
-Square wave produced by clipping a sine wave from a signal generator through a 741 opamp  (clipped it to the rail voltages).
-Fed this into a bread boarded narrow band pass filter of known Q and peak frequency.
-Measured amplitude of signals in and out.
-Leaving the bread boarded filter alone we then decreased the frequency of the square wave with the dial on the sig gen and looked for harmonics measuring amplitude in/out and frequency at which it peaked, when we found them.
-By working backwards in relation to the frequency of the original signal we were able to construct the harmonic series of a square wave.

This approach only works if you can faithfully reproduce your waveform over many octaves.  Your tone will change with frequency so try finding some software instead.  

Why does all this interest me? Because the inner ear is a spectral analyser.  It deconstructs tones into their sinusoidal components.  Your brain is not presented with the image you see on your scope.  Your brain is presented with a harmonic series such as a spectral analyser would display.
When you consider that western diatonic scales are derived from the natural harmonic series you realise that a tone is in fact an extended chord of sine waves (played over many octaves containing many intervals) with the higher notes gradually diminishing in volume.  
This has always fascinated me and still does.  

In practice it would be possible set up some signal generators, sync them together, add the sine waves in a mixing desk and produce any tone you could imagine.  I don’t suggest you try this at home,  we’ll leave that kind of weird shit to people like Brian Eno (the fifth U2).

Hmmm….  I’ve rambled on a bit here haven’t I?

[Paul Marossy]

Hi Rob,

Thanks for the tidbits, that's interesting. I've read that a true sine wave is what a tuning fork produces a nearly perfect sine wave, and it does sound pretty boring, but at the same time kind of fascinating, I guess because it sounds so pure...

I did some poking around on the net, and did find several computer based spectrum analyzers. That would be the easiest thing to do.

I agree that the ear is a mini-spectral analyzer. I'm hearing things, but I can't see them on a scope, and after doing a little more research today, I understand why.

[Craig V]

Hey guys,

I'm sorry I have nothing to add to the discussion, but I was wondering what does the BSIAB sound like?  Is it kind of a fuzz face thing?  Or is it aiming for the Van Halen Brown sound?

Thanks.

ps.  If anyone knows a site that has sound clips, I'd love to hear them!

[Paul Marossy]

Hi Craig,

The sound clips that used to be available aren't anymore. But you are in luck! I happened to have the one that used to be up on that site on my computer. I uploaded it here:

http://home.att.net/~paulnkim/bsiab.mp3

It sorta has that EVH thing going. You decide. I really like it in any case.  

(I have a PCB layout that I designed for it, too. It's more compact than the one at GGG, if you're tight on room in your enclosure. It works, it's what I used in mine)

[Gringo]

Could you please send me your layout for the bsiab? It's on my build list for this summer

gringo@worldconquer.cjb.net

Thanks!!

[Paul Marossy]

Check your email, Gringo.  :wink:

[WGTP]

That is interesting stuff.  Our distortions are making harmonics of harmonics and if it is a 2 or 3 stage, harmonics of those harmonics.  You end up with a pretty dense or full spectum of harmonics.

If you look at the Blackstone Appliances page he has a Spectrum Analysis of a Marshal, a Tube Screamer and his Mosfet Distortion.  Paste it on a page and spread it out to see what is going on.

It's a combination of even and odd ordered harmonics that seems to be what most of us are after.  

[Elektrojänis]

As a general rule, complex waves which have pointy bits to their shape have higher degree of odd (dissonant sounding) harmonics.  Complex waves which have a smoother outline have more even than odd harmonics (pleasant sounding).

I disagree a bit... Complexity and sharpnes of the corners in waveforms does not relate too much to the even/odd -harmonics balance. But... Usually the even/odd balance can be aproximated a bit from the symmetry of the waveform.

Perfect square wave has only odd harmonics. You can smooth it with a filtter but unless the filtter is distorting it will not give out any even harmonics. Triangle waves (well... perfect ones) have only odd harmonics too, but a lot less of them. Sawtooth has all the harmonics (even and odd). Pulsewaves (square where one half wave is longer than other) have both, even and odd harmonics too.

Smoother waves usually sound more pleasant. They have less of the higher harmonics.

In practice it would be possible set up some signal generators, sync them together, add the sine waves in a mixing desk and produce any tone you could imagine.  I don’t suggest you try this at home,  we’ll leave that kind of weird shit to people like Brian Eno (the fifth U2).

Every one of those sine wave generators should have it's own volume envelope generator with a capability to make incredibly complex envelope shapes. And there should be a lot of those signal generators. We are talking about a massively complex setup here. (well... Kind of simple but yet very difficult to use because the amount of parameters.)

That is interesting stuff.  Our distortions are making harmonics of harmonics and if it is a 2 or 3 stage, harmonics of those harmonics.  You end up with a pretty dense or full spectum of harmonics.

Not neccesarily... If you clip a square wave, it is still a square wave (same goes for pulse waves etc.). Tonal shaping (filttering) between stages can change that though.

[Leftrights]

I did some poking around on the net, and did find several computer based spectrum analyzers.

I was just wondering reading through this if there were spectral analyzers that I could get for my PC.  Did you try anyof them?  Could you Give a name or web site to narrow my search?

[RobB]

Elektrojänis wrote:

Complexity and sharpnes of the corners in waveforms does not relate too much to the even/odd -harmonics balance. But...

and

Smoother waves usually sound more pleasant. They have less of the higher harmonics.

You are correct.  My memory failed me.

[Paul Marossy]

For a computer based audio spectrum analyzer, just do a search on it. I found several of those kinds of programs in just a quick search. They all cost something, no freebies that I am aware of.

There was a cool one on ebay, if you do a search for "audio spectrum analyzer", several of the same thing will pop up. It looks like a really cool program for the $45 (I think) that they are asking for it. I would get it, except for my computer is slower than the required 500mHz minimum clockspeed for the program to run.  

That's next on the list - a newer, faster computer...