What's the distortion adding feature of the new $$L console?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Svart

Well-known member
Joined
Jun 4, 2004
Messages
5,134
Location
Atlanta GA USA
I post this here because I proposed a circuit that adds nothing but 2nd order distortion to an audio signal some time ago, (years?) and no one took interest at that time. Now SSL has incorporated it into their new consoles. What do you think they used to create this new option?

I was thinking of how to extract 2nd order distortion from a circuit, we could fool around with a balanced type of circuit that would subtract the normal signal out of the audio leaving the harmonics intact. amp this and we have something to work with. the only thing I don't understand is how we can create the distortion cleanly.. if you can imaging distortion being "clean"..

What do you think?
 
Frequency doublers of various sorts have been around for decades. But if by doing it "cleanly" you mean that the generated waveform is of the exact same shape as the input, but at twice the frequency... well, that's not so easily accomplished in the analog domain.

If phase and amplitudes can be matched (which may be a challenge in itself), a bridge-type circuit can be used to extract the distortion products from the signal. But of course it won't discriminate between the different harmonics.
 
Don't know how relevant it really is to the SSL-method, but it made me think of the info Rossi gave a while ago:

http://www.groupdiy.com/index.php?topic=17795&postdays=0&postorder=asc&start=45

Look for DE000019631892C2 & DE000019631892A1 (from a certain Uli B.)
at www.depatisnet.de

Die Erfindung betrifft eine Schaltung zur Erzeugung der für den Röhrenklang von Audioverstärkern und dgl. typischen Harmonischen auf der Grundlage eines Nutzsignals, mit zwei Signalpfaden 1, 2, die eingangsseitig vom Nutzsignal beaufschlagt sind und jeweils eine Röhre 8, 9 enthalten, und die ausgangsseitig an den invertierenden bzw. nicht invertierenden Eingang 4, 5 eines Differenzverstärkers 6 angeschlossen sind, wobei in einem 2 der Signalpfade ein Inverter 12 angeordnet ist, der die Phase des Nutzsignals um 180° dreht, und wobei in einem 2 der Signalpfade ein Pegelsteller 15 angeordnet ist.
 
[quote author="mediatechnology"]I remember doing this some time ago with analog multipliers. I think I used either an MC1495 or 1496. You can also full-wave rectify and add that back in to the "dry" signal.[/quote]

I recall this: http://www.lynx.bc.ca/~jc/pedalsHarmonic.html
One of those circuits I really should build one day... '1496 indeed.

Regards,

Peter

BYW, it's been a while I visited that JC Maillet site. Saw a few nice new additions:

http://www.lynx.bc.ca/~jc/pedals.html

thefamilySml.jpg
 
Hi Wayne,

I've been messing with the 1496 IC as well times ago, when I was lusting for an eight-string bass. It looked a lot like the circuit you posted.

I also recall Elektuur (aka Elektor) presenting an alike circuit but with some 90 degrees phase shifters in front in order to use another gonio-formula (to avoid the DC-component (0 Hz); which the circuit you posted tackles by means of C1). IIRIC they did +45 & -45 degrees to get the 90 degrees difference.

I gave up when I realized a typical signal (bass or gtr) is more than a one component spectrum since I was after a natural sound, not an FX.

I'm afraid I've also been trying PLLs and a divide-by-two in the loop. Not suited for what I wanted, but at least it got me acquainted with the NE565 :wink:

Note that the MC1495 is disco. RIP
Ah, wasn't aware of a brother to the 1496.

Regards,

Peter
 
So is the $$L circuit a simple exciter circuit like this or something more involved?

It seemed to me that they were adding 2nd order distortion, not pure harmonics..

Who knows, I guess we will have to get someone that has access to the new console to take a look..
 
o.k here is the SSL offical explination on what is happening.

The unique ‘Variable Harmonic Drive’ circuitry introduces controlled distortion of either second or third harmonic, providing variable valve-type warmth or transistor-type distortion on every channel.

so it's introducing 2nd and or 3rd harmonic distortion into your signal not just adding a 2nd or 4rd harmonic.
 
3rd? most people try to stay away from that at all costs.

I doubt $$L used a cheap mod IC, just not their style. What other ways could we come up with the same effect?
 
[quote author="Svart"]3rd? most people try to stay away from that at all costs.

I doubt $$L used a cheap mod IC, just not their style. What other ways could we come up with the same effect?[/quote]

tubes :twisted:
 
[quote author="mediatechnology"]Second order "distortion," if I'm not mistaken, introduces a DC component. You see it as waveform asymmetry. Like a saxaphone on a 'scope with its' strong octave resonances.[/quote]
not sure exactly what you are refering to there
but
I don't think 2nd order distortion and DC offset need to be tied like that

if the original waveform or the reason for the distortion is asymmetrical in nature then a DC shift can occur
:roll:
but then it could be argued that if the original signal is symetrical and the reason for the distorion was balanced on both sides of the waveform then the distorion spread may be even, across 1st, 2nd, 3rd harmonics .. and so on
:oops:
I've just confused myself
I'll shut up now
 
[quote author="Kev"][quote author="mediatechnology"]Second order "distortion," if I'm not mistaken, introduces a DC component. You see it as waveform asymmetry. Like a saxaphone on a 'scope with its' strong octave resonances.[/quote]
not sure exactly what you are refering to there
but
I don't think 2nd order distortion and DC offset need to be tied like that[/quote]

FWIW:

sin(wt)*sin(wt) = 0.5 - 0.5*cos(2wt)

so a DC-term.


The Elektuur-freq-doubling circuit I thought to recall above uses sin & cos, hence no DC-term:

sin(wt)*cos(wt) = 0.5*sin(2wt)


Regards,

Peter
 
[quote author="mediatechnology"]I think the FFT of an asymmetric waveform yields a DC term.
Edit: Generalized: Even-order produces asymmetry; odd-order symmetry.[/quote]

and further to what Peter has said

No AC-coupled waveform, symmetric or asymmetrical, would yield a DC term.
Also a symetrical waveform can have a DC bias.

I had a long talk with my father and we argued in circles. I could try to type the endless math and include the references to locked in phase with the variations of resultant waveform but I don't think it would help or keep us on track to the original point.
Wayne's reference to the full-wave rectified waveform is not really ... I can 't find the word for it ... it doesn't help here.

:roll:

I've given this a lot of thought but can't find a fast way to present it here.

NOT convinced that DC appears naturally. Just because I blow a sax in a room doesn't give rise to an increase in preasure in the room. The variations in pressure is what we perceive as sound but the long term presure in the room will remain at 14psi ... well STP and all that.
You know what I mean.

Also NOT convinced that symmetrical is odd and asymmetrical is even. You may find anecdotal evidence but I'm not convinced that it is proved across the board.

Start again
the only thing I've seen in this thread that leads to higher levels of 2nd than perhaps the other orders is
soft clipping
whether it is in the positive or negative ... or both.

Gut feeling tells me that soft clipping is the trick of the trade here
whilst minimising the distortion across the zero ... crossover distortion
and the any fault that causes a hard instantaneous change to the waveform.
 
Oh damn, what is going on here?!

This is my thread and you stay put Wayne! :green:

We shouldn't be disagreeing at all at this point. everybody gets a say and it all helps lead us to the answer no matter how close or far off you are.

lets get back to work eh?

This forum can't stand to lose any more awesomely smart people because of disagreements..
 
[quote author="mediatechnology"]

If you stretch the opening of a balloon over the bell of a sax (no small feat) and play it, it will blow up the balloon until the air leaks out through the keys, the reed, or the player.

Bye[/quote]


eww. air leaking out of sax players is not good. :shock:


OK comic relief over. Go back to what you were discussing. this is very intresting (but over my head like 90% of the stuff I read here!)
 
interesting. makes me wonder if $$L used both even order distortion AND harmonics or just one.

So I guess a simple question is in order, how does a tube give 2nd order harmonics and/or distortion and how can you emulate it with solid state.

Since a tube works in a similar fashion as a Jfet, I assume there is a way you could emulate the tube and then extract the wanted effect completely.

doable?
 
Since a tube works in a similar fashion as a Jfet

Well, kinda. The drain curves are like the plate curves of a pentode, but the Gm is much lower. They're "similar" only in the most general sense.

The manner in which a single-ended triode stage generates second harmonic has to do with its 3/2-power law. The JFET has more of a square-law relationship between drain voltage and drain current.
 
Back
Top