Need help with R.G. Keen's simple parametric EQ

[megadar]

I am breadboarding R.G. Keen's simple parametric EQ with the intention of building a pedal to help shape the tone of my guitar setup.  It seems like the perfect solution to what I need but I am not getting the results that I was expecting.  I only added one band for now to keep things simple. I ran a white noise generator through the circuit and monitored the output through a frequency analyzer and here is what I am getting...

1. Signal is passing through the circuit so it seems that the input buffer and output are working.
2. with the Q and volume pots turned all of the way up I can hear the filter sweeping back and forth as I turn the frequency pot.

However...
3. With the resonance set at a low Q there is only about a 2db boost/cut throughout the frequency range.
4. With the resonance set at a high Q there is about 7db boost to a 5db cut throughout the frequency range.

I was expecting to be able to get more of a reduction of the resonant frequency than 5db at a high Q and I am wondering if I either have a problem with my circuit, or if this design is just not capable of cutting the resonant frequency by that much.

Here is the schematic that I am using  http://www.geofex.com/article_folders/eqs/parmet.gif

I used TL071's for all of the op amps with the only modification being that I didn't add the first frequency band.

Input buffer IC voltages
1 - 0.15
2 - 4.50
3 - 4.46
4 - 0.00
5 - .015
6 - 4.50
7 - 9.00
8 - 5.91

Simulated inductor IC voltages
1 - 0.15
2 - 4.48
3 - 4.12
4 - 0.00
5 - .014
6 - 4.48
7 - 9.00
8 - 0.00

Output IC voltages
1 - 0.15
2 - 4.50
3 - 4.45
4 - 0.00
5 - .013
6 - 4.51
7 - 9.00
8 - 0.00

Any help with this would be much appreciated!  I used to use a parametric EQ that had +/-12db and was hoping that I could get something like that out of this circuit.  If it is operating as designed I wonder if there are any mods that I could do that might yield the desired results.

[PRR]

There should be 15dB boost/cut with resonance pot set to zero resistance.

Yes, 2dB when res pot set to 10K.

White noise is a poor tool for circuit response testing. Can you use a sine-tone generator and an AC voltmeter?

What does it _sound_ like? A 15dB bump inside the guitar range is pretty strong. 6dB may be mild.

[megadar]

Okay, so if +/-15db is what the circuit should be doing, it looks like I must have something wrong.  It sounds clean and I can hear the the boost/cut as I mess with the pots, but barely.  The EQ I used to use was only +/-12db and had more impact on the signal than this one does.  So far anyway.

Good to know about white noise not being a good tool for response testing.  I have a function generator and a good meter, but I guess I don't understand how exactly to test with a sine wave.  Do I set the frequency of the sine wave to match the frequency of the EQ and measure the AC voltage difference at that point?

At 440hz
Input: .132v
Hi Output: .554v
Lo Output: .101v

At 2.1khz
Input: .124v
Hi Output: 1.438v
Lo Output: .026v

[PRR]

> set the frequency of the sine wave to match the frequency of the EQ

No. Start at 20Hz and go to 20KHz. Use a moderately low level, perhaps 0.1V (100mV) if your meter goes that low.

First: do just sig-gen into meter. A function-generator should be "flat" (constant output over nearly full frequency range) but some meters (especially DMMs) tend to roll-off above 400Hz or 4KHz or something. 10% variation is no big deal, major roll-off above 400Hz is bad.

Anyway, with the EQ between, you should be flat and unity-gain over most of the audio band, with a bump (or dip) in the octaves around the resonance.