AS3320 Filter Questions

[idy]

So this started in my "vocal megaphone effect" thread but led to the 3320 chip which allows you make 4 pole filters. Thanks to all for the response on that thread. So I am making a 4 pole highness. But I have lots of questions. On the breadboard it makes a high pass but the Frequency and Resonance controls have no effect that I can discern. I am probably messing up some basic things but here I go.

Here is the Druid's redrawing of the data sheet circuit (actually the compatible CEM3320)

I am running this on single supply 9v. I just put a voltage divider (two 100k resistors) to give a midpoint voltage to the ground pin. I see the spec sheet says:
Voltage at negative supply pin: min -2.4v max -2.9v
So the midpoint voltage should be only 2.5ish?
I though about a buffered midpoint voltage but not yet tried it.

It has been stated the Freq and Res controls are really current and not voltage inputs. The spec sheet says:
Input bias current Freq control: min .2 max 1.5 micro amp.
Input impedance of Resonance control input min 2.7 max 4.5k

It also says:
sensitivity of pole freq control scale:57.5-62.5 mv/decade

I am told the Freq is a current input which should be thought of as "zero ohm" impedance. The schematic shows a 100k resistor from the CV to the pin...
To get inputs on the order of a micro amp I think need resistance on the order of (for 9v) 9M ohm. That can't be... So I tried a 1m pot feeding the 100k resistor both as a variable resistor and as a voltage divider but did not hear anything happening. I know I'm in the dark.... Should the control V start form 0 or the midvoltage/ground?

[idy]

sorry no image...

[PRR]

> sorry no image...

Image:
https://i.postimg.cc/hQcZHTW8/CEM3320-Highpass-Filter-1024x335.gif

[Rob Strand]

We need the enbiggend version,

https://postimg.cc/hQcZHTW8



Old thread,
https://www.diystompboxes.com/smfforum/index.php?topic=125885.0

From reply 89,
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"

My vague memory from the old thread was the *voltage* on pin 12 sets the frequency.    The voltage range is quite small and because of that any voltage dividers on pin 12 would need to connect to the ground of the filter chip, pin 3; otherwise the DC offsets would be unmanageable and you would never be able to set the frequency reliably.    See fig 2 in the datasheet.

[idy]

Thank you, that  helps. The CV range needs to be very slim...
And that trimmer on V-, "CV reject" has the effect of fine tuning the bottom of that range? I guess I'll cross that river when I get the CV range.
I have 1 1M pot hooked up from 9v to 0 with the wiper feeding the 100k/1.8k on the CV pin (12).
I'm blindly fiddling trying to get that tiny voltage range and not succeeding.
Not sure why... tried 500k and 1.8k and still the CV between .49 and .5 volts.

[Rob Strand]

Thank you, that  helps. The CV range needs to be very slim...
And that trimmer on V-, "CV reject" has the effect of fine tuning the bottom of that range? I guess I'll cross that river when I get the CV range.

When you adjust the frequency it shifts the DC voltage at the filter output.   That's undesirable for an envelope control.   So to set the CV reject you would look at the filter output pin, adjust the frequency control voltage over the full range and try to adjust the CV reject current to reduce the amount of DC shift.   Probably not a big issue for you but you might need to have a ball-park resistors for the chip to work.  See page 5 of datasheet: Pole Frequency Control Rejection.

I have 1 1M pot hooked up from 9v to 0 with the wiper feeding the 100k/1.8k on the CV pin (12).
I'm blindly fiddling trying to get that tiny voltage range and not succeeding.
Not sure why... tried 500k and 1.8k and still the CV between .49 and .5 volts.

Hmmm, doesn't sound good.    Do you have a resistor on the CV reject pin?   Maybe it needs one.


EDIT:
OK, I can confirm it needs an REE resistor.   There's a formula in the datasheet,

Early on it gives,

    REE = (VEE - 2.7V) /  8mA

but later it gives,

    REE  =    (VEE - 3.2V) / 12mA    where the 12mA can be 5mA to 12mA


I'm not sure if VEE in the equation is positive in the calculations.

If you have +/- 9V supplies and we treat VEE as positive I get: 
REE = 787 ohms for the first and a range of 480 ohms, 1160 ohms for the second.

[idy]

I think supply is the problem, it was loosely suggested I could use 9v, but the sheet says:
Pos supply:9 to 18
Neg supply: -4 to -18.
So a total of over 13v would seem in order.

[Rob Strand]

I think supply is the problem, it was loosely suggested I could use 9v, but the sheet says:
Pos supply:9 to 18
Neg supply: -4 to -18.
So a total of 13v would seem in orde

Ah, yeah 9V alone might not work.  If you split a 13V rail you will have to ensure it's skewed to get +9V and -4V, +/- 6.5V won't meet the +9V requirement.   With 4V the REE resistor might need to be quite small it would be best to measure voltage across and make sure V / R is within the 5mA to 12mA range.   Also, if the supply varies from 13V it will quickly screw-up the required REE resistor.  In the formulas you have VEE - 3.2V  so 4V - 3.2V = 1.2V  but if that rail went to 5V you would get 5V - 3.2V = 1.8V, giving 1.5 times higher REE current.  If you choose say 6mA at 4V then it might hold up a bit better if the supply goes up.

Hopefully it's just the power.