vocal megaphone effect

[Rob Strand]

What I've discovered by playing with the chip is that the filter stages themselves are inverting (very useful for pole-mixing filters) and also that the resonance VCA is inverting. So the resonance feedback path *is* inverting, because the resonance VCA flips it.
I found this because I was experimenting with doing resonance volume-drop compensation (In many synth filters as resonance increases, passband gain decreases).

Excellent info, thanks.  There's no way to find that stuff out without having real chips.   (The problem with a lot of special chips is the finer details aren't in the datasheet.   All you can do is expect something will get you.)

I'll have to check the way I've set-up the sim represents reality in terms of signs.    (Maybe after New Year.)

The typical way this is done is to feed some input signal to the resonance VCA as well as the input, so that as the resonance increases, the input level also increases to compensate the passband drop. I found I needed to use a differential mixer to combine the output signal and the inverted input signal. This implies the resonance VCA is inverting, putting the signal back in phase with the main input.

Nice solution to a tricky problem.    You get that level balance issue on Bass-amp contour controls but it's a much easier problem to solve than your case.   BTW,  you have clearly spent a lot of hours deciphering some of those Oberheim filters.

BTW Rob, I love the Sallen-Key sections done with the 3320. That's excellent work, nice one. I've not seen that done before.

Thanks.

[idy]

I'm getting closer to putting those 3320s to use. Most of the example circuits call for 300pf, one calls for 330pf, and a note suggests "good quality film." I am not finding these; DigiKey and Mouser seem to only have 330/330pf film caps with minimum orders of 1000 (still close to a buck each), small bear has 330pf only, I do see silver mica 330pf in small lots on eBay for about $1 each.
I have 300pf ceramic discs, these are readily available. I should probably try these on the breadboard.

Suggestions? There must be a reason why the design had to use these values...?

[Rob Strand]

You should be pretty safe with ceramics provided they are COG or NPO types.
Film caps below 1n aren't that common.

Suggestions? There must be a reason why the design had to use these values...?

I haven't gone through the fine details but it often comes down to the range of frequencies.

In fact, I'm pretty sure 300pF is too small for the range of frequencies you want.

You might want say 200Hz to 2kHz range or so.

[anotherjim]

I've got Wima polypropylene in 330pF from Rapid in the UK, but I'm sure the rest of the world has them if we do. They're nice caps with good consistency I find. The only problem is their cases are quite wide so you can't stack them side by side on a 0.1" pitch.
Maybe searching for "film" or "Polyester" is hiding the polyprop' option?

[amptramp]

I have seen polystyrene capacitors in the 330 pF range.  They are very temperature stable but they require care in soldering as they are only good to about 85°C above which, they melt.  They should still be available.

[garcho]

The exact value of the caps isn't super important. Do you have a breadboard? Try anything you have or can find close, 220, 470. You'll be dialing the cutoff in with CV and a pot, right?

[idy]

That's what I thought, the frequency is dialed in so why the special value... but still, if the frequency is dialed, why the special value?.

Thanks to all.

I am seeing the 330pf at mouser now, but 300 is still the buck a piece minimum 1,000. Funny.

[PRR]

It is a "megaphone effect". Don't polish poop. Crap-caps will be fine.

[garcho]

That's what I thought, the frequency is dialed in so why the special value...

These ICs are based on old designs, might not have been special in 1983 for those designing pro audio. Or Mr Curtis was a perfectionist and the math told him 300p. These were intended for serious gear from major companies in the synth world.

[Rob Strand]

The caps can be used to shift the range of frequencies to suit requirements.

CEM3320 datasheet
http://www.bustedgear.com/images/datasheets/CEM3320.pdf

page 3, "Operation of Each Filter stage"

RF = 100k
REQ = 1M parallel with 100k  = 91k
AIO = 0.9
VT = thermal voltage = 26mV

Filter frequency,

fP  =   AIO * exp( - Vc / VT ) / (2 * pi * REQ * C)


page 4,  "Pole Frequency Control Scale",

"For best results the over a thousand to 1 (filter frequency) control range,
the voltage on pin 12 (Vc) should be maintained between -25mV and +155mV"

1/(2*pi*REQ*C) = 5830 Hz

Vc                      AIO * exp( - Vc / VT )     fP
-25mV                2.354                              2.354*5830 = 13720Hz
+155mV             0.00232 = 1/431           5830/431     = 13.5Hz

Notice 13720 / 13.5 = 1016  which agrees with the 1000 to 1 filter frequency control range.

So a 300pF cap includes our range.

We can work backwards to get an idea what control voltage is require for some specific filter frequencies,

fP                 Vc
250Hz         79mV
500Hz         61mV
1kHz            43mV
2kHz            25mV
 
Using a more convenient value of 330pF will just shift the frequency down by 10% with no ill effects.

[idy]

Thanks all. Answers for those without understanding indeed! Fine distinctions for the  kludger. Am fiddling on the breadboard already.....

This has kind of strayed from the megaphone effect, maybe time to start a new thread. We'll see what use I find for a 4 pole filter.

For me, both voice and acoustic guitar like highpass. I am also looking at a notch filter for the guitar. 3320 can make a band pass... I am reading about notches.