Generating Upper and Lower Harmonics

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
scott2000 said:
I'm having issues with my image attachments but I wanted to ask what kind/types of harmonics/distortion are present across a push pull stage biasing resistor as it relates to monitoring this point for balancing????

edit//// let me try this again....
I surmise you're referring to a cathode-bias push-pull with a single common cath resistor. The AC voltage across it, which should be null in a perfect class-A circuit, in fact shows a rectified harmonic (at twice the fundamental frequency); the amplitude of the peaks should be as identical as possible, since they represent the summ of the tubes' current.
This method is valid only for un-bypassed cath resistor.
 
I found a very simple tube based sub harmonic generator somewhere ,I think in Wireless world or another English publication  , It was two tube sections , first tube distorted the second and low pass filtering was applied to the outputs ,which was then mixed back in with the clean signal,
the details are sketchy in my mind due to time , but 'Synthetic bass generator' might have been the title . Ill have a dig around and see if anything pops up. 

http://www.douglas-self.com/ampins/wwarchive/wwarchive.htm#exley

'Bass without big baffles '

I think that was the article I spoke of , just re reading it now myself ,
 
Tubetec said:
I found a very simple tube based sub harmonic generator somewhere ,I think in Wireless world or another English publication  , It was two tube sections , first tube distorted the second and low pass filtering was applied to the outputs ,which was then mixed back in with the clean signal,
the details are sketchy in my mind due to time , but 'Synthetic bass generator' might have been the title . Ill have a dig around and see if anything pops up.
I recall an old school efx that generated lower frequency (sub harmonic?)  components at 1/2 the fundamental pitch (effectively a div by 2 circuit, with smoothing).  I never took these efx very seriously.

JR
 
You have to love the proviso the author put in  ;D
 

Attachments

  • capture 6.PNG
    capture 6.PNG
    115.3 KB · Views: 32
Yeah Doug's site  has some of the best of Wireless world audio related articles , I think there is a link PRR posted to a huge repository or Wireless world stuff somewhere here too  . I found it interesting reading the US wireless magazine from the 1930's the other day ,theres such a huge contrast between the language compared to Wireless World ,even the 1930's US speak isnt much different to modern language , where the British style seems almost Shakesperian in its delivery .
 
JohnRoberts said:
I recall an old school efx that generated lower frequency (sub harmonic?)  components at 1/2 the fundamental pitch (effectively a div by 2 circuit, with smoothing).  I never took these efx very seriously.

JR

Thats the DBX120A, or maybe a derivative.
 
buildafriend said:
hey all,

we know that certain circuit topologies and pieces of equipment have more or less harmonics. that said, what circuit blocks are cool for generating more upper and lower harmonics intentionally and not just as an artifact of the circuit at hand?

I'm looking for less obvious things than just eq boost the fundamental or use a tube circuit and something more like a circuit that is made to do this specifically without adding much odd order distortion

cheers,
An underbiased push-pull stage will make crossover distortion that is both high frequency and odd-order, but if you add a DC bias the result has less odd-order and more even-order harmonics.  Depending on the effect you want, you may want to mix this at a low level to the original signal, something like the Aphex aural exciter thing.

FETs have already been mentioned as substitute for the classic class A vacuum tube circuit. Here's a video ...
https://www.youtube.com/watch?v=7u9OKPL1ezA

Do you have a specific purpose/application in mind, or do you just want to generally learn what all is possible? One use I've heard of is generating higher harmonics in "restoration" of older recordings that have limited bandwidth - kind of the opposite of de-essing.

This looks like an interesting patent:
https://patents.google.com/patent/US4150253A/en
 
> for example, single ended class A triode circuits tend to generate mostly second harmonic distortion, single ended class A pentode and semiconductor stages tend to generate mostly third harmonic distortion ....

Very simplified and contextual.

All single devices make even-order. All amplifiers with non-infinite supply will make big odd-order at the limit. 3rd grows faster than 2nd, so the most common result is mostly-2nd at medium levels going over to mostly 3rd at high levels.

Naked transistors make tons of 2nd. A tube will run 5% before 3rd dominates. A naked transistor goes to 25% before 3rd ears its ugly head. This is SO much that we "never" run transistors naked. (Also the gain can be intolerable.) Transistors "always" used with NFB. An emitter resistor will easily cut THD from 25% to few-%.

There are very few ways to generate simple (musically useful) distortions at very low level. This is "natural", how the world works. It suits music because it allows the musician to add emphasis without actually blowing the listener's ears out.

The simple theory is about single simple stages on single tones. Cascaded stages are more complicated. Complex signals (most music!) is more complex. Multiple low-THD stages in cascade produce a low hash of inharmonic intermodulation; why large recording consoles pursue point-oh-oh-oh THD numbers.
 
Thank you so much guys

I have a lot to look over here and I'll need to probably review it all a few times make sure it sinks in. Please feel free to add any more information that comes to mind.

Mainly, I thought it might be nice to have a standalone circuit that creates harmonic content. I'm not after copying the culture vulture, there's a plugin version of it ;) I'm not after copying the Aural Exciter either. The VCA idea sounds really smart but I see what Abbey is saying about how it was designed while considering the nature of an oscillator or some kind of more predictable input signal coming into it.

If anything comes out of this I'll gladly make boards and BOMs of it. Maybe there's a cost effective idea we could play with that could fit into a little hammond box? It would be nice to have a block like this to fit into other circuits too, like mic preamps or compressors.
 
I like (on paper) John Broskie's  'aikido harmonic restorer'  ...  basically a take on his circuit (aikido)  ..  to add a nice thd signature ..

https://www.tubecad.com/2013/06/blog0265.htm

I would think it is a very mild form of mostly h2 plus little h3  distortion  within the context of a very clean signal and some 6dj8/6922 'noise-reducing'  gain staging.

For the non-tube application, it could be interesting to do a reverse 'tube-fet equivalence' process and see if it does anything useful in the small signal fet world.

Just as soon as I have my bench tube rig 'reset', I'm planning to have a go at the 'aikido'  ..  I really need to see it for myself!

 
buildafriend said:
Mainly, I thought it might be nice to have a standalone circuit that creates harmonic content.
I believe almost anything that can be done with distortion has been done. There are thousands of distortion plug-ins, some of them using ophisticated algorithms (much more than what can be put in a Hammond box) and they all sound like crap to me.
We accept distortion on electric guitar and some keyboards because, after so many years of being exposed, it has become an acquired taste.
Maybe after many years of being exposed to distorted voices, I could find them acceptable, but I think I'll be dead before.
 
scott2000 said:
"A frequency selective Harmonic Restorer"
Abusive term. According to the author, what CD's are missing is the content between 22 and 27kHz, and I tend to agree with him. They are buried in the noise floor (because being attenuated >90dB by the anti-alias filter), and no technique can recreate them without adding new formants that are alien to the original signal. Even using digital frequency shifting methods generates unpleasant harmonics, since it applies frequency-shifting to existing odd harmonics.
Current disaffection of Aphex and similar enhancers is proof that it's not a long-term valid proposition.
 
I kind of like some BBE process on my acoustic guitar that has a piezo  :)

Useful to tame and extend that sparkle at the top.

I believe it does do some generation of some hi harmonics and so forth.  There is a really cheap pcb on eby with a bbe implementation - could be interesting to see what that does in a system with some other distortion modules

..  or not  ;D

And what about those 'color modules' for the 500 series ..  I think some of those could be interesting to play with.

Getting good distortion is definitely a whole-of-life thing!  I'm still trying.
 
alexc, I agree. some things sound better with a little distortion. there is a reason the decapitator plugin became very popular..

https://www.soundtoys.com/product/decapitator/

I have to take a look at the PCB you just mentioned.. Can you link us to it?
 
..  cheap eby bbe modules of various types ....

https://www.ebay.com/sch/293/i.html?_from=R40&_nkw=bbe&_sop=15

Integrating into a system could be interesting  ....  parallel streams ....  mixing down to mono or stereo, where a stream can be quite heavily eq'd and/or  with a distortion module ...  I like jfet pcbs for that purpose.

Apart from guitars and keyboards, which I frequently have some crunch going, I use it a lot on bass and sometimes on drums :)

Typical for me is 'crunch' on the bottom end and a touch in the mids,  with clean plus some sparkle at the top end.
 
alexc said:
I kind of like some BBE process on my acoustic guitar that has a piezo  :)
The BBE process does not rely on distortion; it is a dynamic expander operating only on treble (combined with a midrange notch).


I believe it does do some generation of some hi harmonics and so forth.
Since it incorporates a VCA, it actually produces a very tiny amount of distortion (typ 0.05%)
 
The bbe  (rack) also had a  'low' end enhance ... I thought that the 'low contour'  control  was of the 'sub-harmonic generating'    kind  ..  adding harmonics at 1/2  of the 'low end dominant freqs' or something like that.

The top end 'process' sure does a lot of boost/cut - I'm pretty sure the harmonic distortion from that would not be unchanged.

But - I haven't measured, only 'used' it for a long time back in the mid 90s.  It seemed quite novel then, and I used it alot for years  ...  it was the best I had access to  at the time .... eventually bin-ed it after it went awol.

It must have done 10 years service as my 'lounge hifi' enhancer, however  ;D

I do like these kinds of processors  ...  my progression  was    bbe enhancer  then spl vitalizer  then  several  finalizers

...


I also had a yamaha fx750 which was a 90s guitar processor  ...  it did have a multiband distortion block.  It was quite interesting as a  distortion effect, a bit more refined than most at that time. 

I like the principle ... nowadays I use that kind of thing as a plugin.

..

One of the most versatile distortion processors of all time, imho, is the one in the Roland VG-8 guitar processor.

It's polyphonic, requiring the GT-6 'divided pickup' [and a worthy 'donor' guitar']. It is 'mult-band' to the max, for guitars.

Once you get going with the 'models' and 'resonators' and stuff, it gets very 'out there', but polished sounding and  studio friendly. 

I use mine with an Art Compressor/Limiter/Gate - the Roland VG-8 has a ferocious dynamic range when going from preset to preset ... the Art box adds the sorely needed analog functionality.

Here is Australia, it was made famous by a local artist mr Ed. Kuepper, who used it for many years live performance.

He was, and still is a force to be reckoned with in the guitar world .. from 1975's    'Im Stranded'  (emi uk release) to the present day.

A real 'delta blues' man at his most introspective  and all round  serious, working man guitar/comp/singer    guy.

...

And speaking of generating upper harmonics ..  the FInalizer has a 'digital radiance'  function block in it ...  it adds/subtracts 'harmonics'  in the digital domain  (of course).  It's quite subtle ..  certainly doesn't damage the sound  :) 

I am doing some more measurements of the Finalizer with REW  ...  there's literally an ocean of audio in that 1RU to get busy with.
 

Latest posts

Back
Top