What would be the best distortion for aural exciter like harmonics?

[Mark Hammer]

Yes, R11 is a pot for mixing the "excited" signal in with the "unexcited" one.  The rest of it checks out.

Were one to "redesign" it, I suspect the preference would be to have the input stage be non-inverting, for a higher input impedance.  By doing so, that would eliminate the need for inverting stage IC2D.  2D is really only there to compensate for the fact that 1A is inverting.

However, it works just fine, as is.

[erikb1971]

Hey Mark, thank for your reply. I am going to google on the inverted/non inverted thing you are saying. If I still do not get it then I will get back to you.
This is the schematic with the suggested mods.

Could you maybe look at the way I put in the pots, cause that was some wild guessing to be honest.
Would it be possible to set R11 so that it would only have the excited signal? Since in my design Iam already mixing all processed bands with the original signal, Iam trying to have as little overlap as possible... cheers

erik

[Gus]

Have you looked at this for ideas?

http://www.paia.com/ProdArticles/quadrafz-design.htm

[erikb1971]

Have you looked at this for ideas?

http://www.paia.com/ProdArticles/quadrafz-design.htm

I think I used to have it as a plug in in an earlier version of cubase actually. It is very nice, specially for guitar. For my purposes however, I do not see a lot of use. The link is very much appreciated!

[Mark Hammer]

The pot, or whatever it is, wedged between C1 and C2 has me pretty confused.

The variable tuning of the first 2-pole section, built around IC1B, involves one resistance from the output (pin 7 in your drawing) back to the junction of C1 and C2, plus a second variable resistance from the non-inverting input (pin 5) to ground.

As originally drawn/designed, the 4n7/4k7 networks provide a rolloff starting around 7.2khz.  That's unlikely to yield anything audible with a great many sources (though decent results from some).  If you replace the 4k7 fixed resistors in that first filter stage with 3k9 in series with half a dual ganged 10k pot (so that resistance in each path ranges from 3k9 to 13k9)  you'll be able to tune the filter corner frequency of that section from 2.4khz up to just under 8.7khz, which will likely cover the most usable and relevant parts of the spectrum.  Keep in mind the second section, built around IC2A will still have the 7.2khz corner frequency, so the variable first section will have modest (though still usable and musically relevant) impact on the overall result.

[erikb1971]

The pot, or whatever it is, wedged between C1 and C2 has me pretty confused.

The variable tuning of the first 2-pole section, built around IC1B, involves one resistance from the output (pin 7 in your drawing) back to the junction of C1 and C2, plus a second variable resistance from the non-inverting input (pin 5) to ground.

aahhhh so I should replace both those resistors with a pot, or a dual gang (just found out the dual gang can replace both resistors with one pot!) I see...

As originally drawn/designed, the 4n7/4k7 networks provide a rolloff starting around 7.2khz.  That's unlikely to yield anything audible with a great many sources (though decent results from some).  If you replace the 4k7 fixed resistors in that first filter stage with 3k9 in series with half a dual ganged 10k pot (so that resistance in each path ranges from 3k9 to 13k9)  you'll be able to tune the filter corner frequency of that section from 2.4khz up to just under 8.7khz, which will likely cover the most usable and relevant parts of the spectrum.  Keep in mind the second section, built around IC2A will still have the 7.2khz corner frequency, so the variable first section will have modest (though still usable and musically relevant) impact on the overall result.

....

It took me a while to understand what you meant.. but I think you mean:

[Mark Hammer]

You got it.

An alternative would be to use a toggle or slide switch to select from a few audibly distinct rolloffs, via fixed resistors. 

This has a few advantages.  First, pot sections on dual-ganged are not particularly well matched.  Second, there are more choices in switches than there are in pot configurations.  A pair of on-off-on DPDTs (one per filter section) would easily give a variety of filter responses.

The dual-ganged thing is probably simplest for the time being, but if you have difficulty scoring the dual ganged pot, there is absolutely nothing wrong with using something like a 4-pole slide switch to select between "guitar" and "percussion" settings.

[soggybag]

...
As originally drawn/designed, the 4n7/4k7 networks provide a rolloff starting around 7.2khz.  That's unlikely to yield anything audible with a great many sources (though decent results from some).  If you replace the 4k7 fixed resistors in that first filter stage with 3k9 in series with half a dual ganged 10k pot (so that resistance in each path ranges from 3k9 to 13k9)  you'll be able to tune the filter corner frequency of that section from 2.4khz up to just under 8.7khz, which will likely cover the most usable and relevant parts of the spectrum. ...

Mark, how did you calculate the rolloff frequency from these values? Is there a page somewhere with RC math?

[Mark Hammer]

Someone will likely correct me but I just used the good old  F = 1 / [2 * pi * R * C].  It helps that it is a fairly simple Sallen-Key filter with equal-value resistors and caps.

Look here for details:  http://en.wikipedia.org/wiki/Sallen%E2%80%93Key_topology

[erikb1971]

what programs do you guys use when you want to go from a schematic (mine is made in Eagle) to a perf / vero  board prototype?

cheers

Erik

[erikb1971]

You got it.

An alternative would be to use a toggle or slide switch to select from a few audibly distinct rolloffs, via fixed resistors. 

This has a few advantages.  First, pot sections on dual-ganged are not particularly well matched.  Second, there are more choices in switches than there are in pot configurations.  A pair of on-off-on DPDTs (one per filter section) would easily give a variety of filter responses.

The dual-ganged thing is probably simplest for the time being, but if you have difficulty scoring the dual ganged pot, there is absolutely nothing wrong with using something like a 4-pole slide switch to select between "guitar" and "percussion" settings.

Hi Mark

Glad I got something right :-)
For anyone interested, the thread for the build of the complete 19" unit is here: http://www.prodigy-pro.com/diy/index.php?topic=45195.msg565858#msg565858

On the selector for a few frequencies the following. I have done some experimenting with the set up of the multiband distortion unit with a behringer crossover and it was shitloads of fun playing with it, even with my somewhat limited setup. So I am a bit in doubt about switches. On the other hand, the harmonic sweetener part is probably the least usable for live use (the most subtle effects) so that favours the switches. Also, I love the idea of having four "lit when on" push buttons for freq selection, with a surprise setting when you press them all at the same time (like the nuke setting on 1176). Choices choices choices.

One other thing I would like to ask you is the following: does the harmonic sweetener need the lower frequencies for the production of the higher harmonics? I mean, Iam experimenting now with the behringer. It has two variable xover frequencies. So it splits the sound in three bands. Would it make sense to put the harmonic sweetener on the higher frequency band (e.g. from 3khz up). Or should it be inserted a copy of the complete signal?

BTW, iam reading for hours the past days/weeks on www.hammer.ampage.com and I have seen so many things I might want to incorporate... that place is a goldmine!

Cheers

Erik

[Mark Hammer]

One other thing I would like to ask you is the following: does the harmonic sweetener need the lower frequencies for the production of the higher harmonics?

Actually, no.  Jules Ryckebusch's original circuit was intended to be a standalone unit so naturally it contains/passes the entire spectrum.  If, however, your bass and mids are finding their way to the exit sign via another route, then you can feel free to leave out the clean/unaffected signal from the HS, and simply feed your input signal to the highpass filter /clipping section and mix that in at the final mixing node with the other things.

Glad you're enjoying what's on my site.  I'm pleased to make a lot of that old stuff available to beginners.  It's gold.  All I did was recognize it and scan it.  Thank the folks that wrote it in the first place.

[erikb1971]

so that would be like this:

so that would be like this:


cheers

Erik

[erikb1971]

would this be a good schematic for adding a noise gate?

http://freestompboxes.org/viewtopic.php?f=1&t=3787&sid=987cc1e690111db13932f53dd68a1bbc

[Mark Hammer]

Probably more complicated than you really need.

A simple gate might be something like this.  You take the single op-amp envelope detector from something like the Dr. Q or Nurse Quacky, and instead of a diode for whatever is used to feed the base of the transistor in those circuits, you use the LED half of an optoisolator.

The LDR part of the optoisolator is used to essentially replace the 10k mixing resistor you see in the Harmonic Sweetener schematic, just after the Mix level pot.  When you play, the LED is lit up and reduces the resistance of the LDR, allowing signal to be mixed in.  Stop playing and the LDR resistance goes high again, placing whatever residual hiss is coming from the highpass/clipper sections waaaayyyyyy down in the mix.  Hard to get much easier than that.

Realistically, one would probably want to use some sort of resistance in parallel with the LDR so that when it goes low, the combined LDR/parallel-resistance is low enough to pass a healthy amount of exciter signal.

Does that make sense?

[erikb1971]

Probably more complicated than you really need.

A simple gate might be something like this.  You take the single op-amp envelope detector from something like the Dr. Q or Nurse Quacky, and instead of a diode for whatever is used to feed the base of the transistor in those circuits, you use the LED half of an optoisolator.

The LDR part of the optoisolator is used to essentially replace the 10k mixing resistor you see in the Harmonic Sweetener schematic, just after the Mix level pot.  When you play, the LED is lit up and reduces the resistance of the LDR, allowing signal to be mixed in.  Stop playing and the LDR resistance goes high again, placing whatever residual hiss is coming from the highpass/clipper sections waaaayyyyyy down in the mix.  Hard to get much easier than that.

Realistically, one would probably want to use some sort of resistance in parallel with the LDR so that when it goes low, the combined LDR/parallel-resistance is low enough to pass a healthy amount of exciter signal.

Does that make sense?

Almost! 

I had to look up almost every word on wikipedia and stuff. But I think I understand most of it. This is a schematic of a nurse quacky:



The little blue circle is the base of the transistor, so everything left of it feeds it. This one seems to be fed by a led (huh.. I see two?). Before that I see what is an op-amp,  so that should be the single op amp envelope follower you are talking about. So I could use the bit from the sensitivity pot to the le before the 51k resistor. That led becomes the led half of a optoisolator. More light means less resistance, means the effect is mixed in. I order to make sure to have enough resistance I have to wire a resistor parallel to the optoisolator to make sure that it has enough resistance to actually push the effected signal down enough.

So the part roughly in the green circle is added to the circuit, fed by the clean signal input. It is inserted with the LDR replacing the mix pot in the original schematic. So what part within the green circle do I use?

[Mark Hammer]

I can't see pictures here at work so I'll need to wait until I'm home to respond.

[erikb1971]

seriously, you think working and making a living is more important than the noise gate on my harmonic sweetener? 

[Mark Hammer]

Not at all, but that still won't get me past the damn filters here. 

[erikb1971]

hahaha.. If you are really bored I could email you the picture :-)