octave up idea....

[Quackzed]

Posted on: Today at 06:24:25 AMPosted by: Processaurus 
Insert Quote
How about a FWR octave, where the pointy side of the signal gets soft clipped, but the threshold of the clipping is varied somehow by the note's envelope, so that the resulting waveform is shaped the same at different points in the notes envelope.  So that sustain would be preserved instead of getting that gated, misbiased sound.  Big signal = higher clipping level, small signal, lower clipping threshold.
Posted on: Today at 03:24:32 AMPosted by: A.S.P. 
Insert Quote
one of the most terrific views on my scope screen
was watching a 400Hz sinewave
through the E-H HOG,
morphing into the various octaves
while playing with the slide-pots...
 
Posted on: Today at 02:20:01 AMPosted by: gez 
Insert Quote
There's no easy way to do this...not well, at any rate.  Most methods involve squaring off the signal, doubling it then shaping it to a sine wave.  All methods involve a fair bit of circuitry and the ones that get top results will only be monophonic if they're analogue.
Posted on: June 10, 2006, 07:52:03 PMPosted by: Quackzed 
Insert Quote
oh yeah...     theres really no way to get those 'points' of a rectified signal back in touch with their original 'smooth' sine wave selves... what about a diode feedback clipper? to smooth out the harsh high frequency junk?? or even hard clipping everything below 0 v? 
Posted on: June 10, 2006, 03:02:28 AMPosted by: gez 
Insert Quote
Quote from: Quackzed on June 10, 2006, 01:10:18 AM
what if you took 2 identical signals and inverted one... then full wave rectified both... so you have one positive "arch" signal and one negative "divit" signal? 

When mixed, they'll simply cancell each other out (nothing).
Posted on: June 10, 2006, 01:10:18 AMPosted by: Quackzed 
Insert Quote
              octavers all rectify the signal somehow right? , well i noticed that the resulting wave didn't look like a sine wave, so i wondered...
what if you took 2 identical signals and inverted one... then full wave rectified both... so you have one positive "arch" signal and one negative "divit" signal?  then hard clip the 'points' side to ground of each signal before blending them back together... 
you would get a closer approximation of a sine wave??! i guess the clipping would add artifacts ,but at least you still have the original signal mixed in there...  sort of like having 2 crossover points instead of one for every period... would this lessen the unrelated harmonics? or make them worse... i'm gonna bet theres something out there that does this!
??

[Quackzed]

i moved (??!!) this thread from the OT/Lounge.
       I kinda got it all backwards...
But basically i was asking if there was any way to get a full wave rectified signal back to a more sine wave looking wave... by clipping the 'pointy' half of the signal or other creative ways to smooth out the somewhat harsh edgy sound of a full wave rectified signal...
   following the original envelope sounds like an interesting way to achieve some 'rounded' clipping..
i was thinking more along the lines of clipping all the 'arches' to ground of an inverted version of the signal. then mixing the opposing 'pints' back in...  that would cancell the points and be like hard clipping, but what about the inverted signal at half strength... wouldnt that make them 'half' as pointy? or 'rounder'?

[Joe Kramer]

Hi!

I've also done some thinking about how to restore the symmetry of a FWR signal.  Here's what I came up with, but it involves a few steps:

1) FWR the signal and "fold" it in half, which gives you the basic octave-ized signal.

2) Level-shift the signal to restore the zero-crossings.  This can be done simply by AC coupling with a capacitor.

3) Half-wave rectify this signal again, keeping the rounded top half and discarding the spiked lower half.

4) Invert the above signal and shift it 180 degrees out-of-phase.

5) Mix the non-inverted and inverted signals produced by steps 3 and 4 back together again.

This will give you something like a symmetrical waveform that's an octave up from the original.  The problem, as far as I can see, lies in the 180-degree phase-shift step.  You have to phase-shift the inverted signal before you mix it or else it will simply cancel both signals out.  There would be two ways of doing this.  The easiest way would be to use a fixed phase shift network, but the problem here is that only certain areas of the frequency spectrum would be shifted while others would not be: some notes would sound, and some would be nulled-out altogether.  You might be able to get away with a control that lets you dial-in the frequency range that you expect to play in, but that might be too restrictive.  The second approach would be to use a very slight fixed delay.  This would give a uniform phase shift across the whole frequency range, but would be sort of like going around the world to get across the street--once you start using BBDs in your simple analog octave device, it's not so simple any more.

Anyway, those were my thoughts on the subject.  Probably others might have ideas for solutions more elegant than mine. 

Regards,
Joe

[christian]

But basically i was asking if there was any way to get a full wave rectified signal back to a more sine wave looking wave... by clipping the 'pointy' half of the signal or other creative ways to smooth out the somewhat harsh edgy sound of a full wave rectified signal...

Only problem here is that a guitar doesnt make sinewaves
You can run a triangle into FWR and get triangle with 2x frequency out, saw-wave would produce triangle with the same frequency,
square-wave generates a pulse-wave 2x frequency etc..

But guitar has a complex waveform that you cant bring back with any kind of hackery.

Just my 2c

ch.

[GibsonGM]

Don't know if this helps, but I built something like this a while back while playing with OA's, and it works really well for leads above the 10th fret...easy to build, has that 'chimey' octave sound, but not too clipped...I put a sound sample up, too...the output's a fairly nice sine wave.  Wish it worked better for lower notes.  I've been meaning to tweek it.   Added bias to the output OA to restore zero-crossing and a blocking cap. Can re-post the new schem is anyone's interested.  Between all of us, I bet we could come up with something decent! 


http://www.diystompboxes.com/smfforum/index.php?topic=45483.0

[Joe Kramer]

Hi,

There is also the route somewhat suggested by Christian, which is to re-shape the guitar signal into a sinewave first.  You can do that with the circuit found here:

http://img.photobucket.com/albums/v317/StephenGiles/EHsynth1.gif

Thanks Stephen Giles!  I haven't tried this, but if it works as advertised, you would get a perfect octave all over the guitar neck.

Regards,
Joe 

[johngreene]

If you use a multiplier instead of a full-wave-rectifier, it will do the same thing.
Here's a graph output from an excel spreadsheet that has a sine-wave input, a 'squared' output, and a FWR (absolute value) output.



--john

[Quackzed]

Now THATS what i'm talkin about!
look at that purple line... smooooooth.

[A.S.P.]

now beware the multiplied/squared amplitudes,
and the non-sinusoid waves` behaviour...

[Quackzed]

DOH!!!!
i'm guessing that that means it's not as simple as it seems?

[johngreene]

It's been done before. It's not that difficult. Since you are multiplying the signal by itself, the amplitudes are easy to keep in control. You just have to be careful of the DC offset between the inputs to get a nice octave.

BUT.....

People I know of that have done octaving using a 4-quadrant multiplier say it's not very exciting because it's so clean.

--john

[Joe Kramer]

Hey John,

Do you know of a good and fairly simple multiplier circuit, say 3080-based, that you could point to?

Thanks!

BTW, just found these two PDFs (the first two entries on the page), by Ray Marston, which have some very basic 2 and 4 quad multipliers.

http://www.google.com/custom?q=ray+marston+OTA&btnG=Google+Search&cof=L%3Ahttp%3A%2F%2Fwww.nutsvolts.com%2Fimages%2Flogo.jpg%3BAH%3Acenter%3BGL%3A0%3BBIMG%3Ahttp%3A%2F%2Fwww.nutsvolts.com%3BT%3A%23000000%3BGIMP%3A%23FF0000%3BGFNT%3A%23666666%3B&domains=www.nutsvolts.com&sitesearch=www.nutsvolts.com

They'd have to be worked up a little bit to be usable for guitar though.

Joe

[johngreene]

Not OTA based, but this is one of the 'simplest' 4-quadrant multipliers to use that I've seen.

You'll just have to think of something to do with the other 3 multipliers.....

http://www.analog.com/UploadedFiles/Data_Sheets/431190299mlt04.pdf

--john

[Gus]

John

was that 2001 with R.G. GFR posting a bit about jfets?  People do need to read all of geofex.

[Joe Kramer]

Wow--cool!  Guess it doesn't get any easier than that.  I have to get hold of one of those chips and give it a try.   Thanks John!

[johngreene]

John

was that 2001 with R.G. GFR posting a bit about jfets?  People do need to read all of geofex.

Actually it was a lot of bit. But, in this case I think the question was asked in a way that didn't directly correlate to 'clean octave'. The scope was much more narrow.

At least that is how I saw it so my response was geared toward pointing out the connection of "smoothing out the pointy part" and actually creating a 'clean octave'.

--john

[christian]

Multiplier acts pretty much the same as FWR, at least thats what the outcome is.

These to expressions are the same:

- Multiplier

Code Select Expand


x = x * x

- FWR

Code Select Expand


x = abs(x)


where abs(x) is just

Code Select Expand


if ( x < 0 ) x = -x


If you multiply a negative value with itself, you get positive value, so the x is same in both expressions.

BUT.....

People I know of that have done octaving using a 4-quadrant multiplier say it's not very exciting because it's so clean.

I recon that. Clean octave is sorta like this "holy grail" to designers and isn't quite worth the effort in the end. The thing in not-so-clean octave-uppers is probably just that that they have this cute overtone kinda thing going on.

For absolutely clean octave-up, you should change the whole concept and use something like analog pitch-shifting with delay-lines?

ch.

[GFR]

Various ways to skin a cat.

How to round the pointy part;

You can use a diode clipper for that:

http://www.geocities.com/munkydiy/mutantesoctave.pdf

You can use a squarer based on fets:

http://www.geofex.com/Article_Folders/jfetdoub/mudoubler.htm

Or you can use a squarer based on multipliers, like mentioned earlier in this thread.

One thing you have to keep in mind is that all this supposes you have sine waves, not guitar notes. Also, smoothing the pointy side may get you a tone that is "too clean" (dull).

I suggest you read:

http://www.geocities.com/gfr.geo/octave.html

and listen to the sound samples.

[Quackzed]

wow, cool links... that last geocities site has a really good explanation and description of some different methods of 'taming' an octave up... i'll admit, the octave up (multiplier) wave isn't exactly 'wild' but it does seem to emphasize the octave... and you can always add some gain post octave??!! seems like 2 steps forward and one step back... i mean, why bother cleaning up the wave if your just going to dirty it up again??!! ....the mutantes article was good too, but i was surprised at how big the circuit was.. not huge but not a simple build either...

[Hugo L]

Have any of you ever build one of these?
http://www.lynx.bc.ca/~jc/pedalsHarmonic.html