CGS04 DX Mixer distortion?

[metalboxhead]

I'm looking a building a DIY mixer. The Ken Stone design here http://www.cgs.synth.net/modules/cgs04_mix.html is a simple one that will suit my needs. I've seen an almost identical version:
http://aisynthesis.com/wp-content/uploads/2016/06/mainboard.pdf

However the claim from the manufacturer (see point 11) http://aisynthesis.com/ai002-diy-eurorack-mixer-module-build/ is that the mixer will distort when the feedback pot P4 is wound all the way up, which I've replicated on a breadboard.

What I cannot understand is that for a mixer which should have a gain of -1, why it would distort? Ken Stone doesn't mention anything about this.

[Grumble]

The reason why Ken Stone doesn't mention it is because the schematics are different from each other. Where Ken Stones amplifier has a fixed gain, the other diagram has a potmeter in its feedback loop.
Why this mixer will distort when the runner of the potmeter is all to the left (in this diagram) is that the output of IC1A gets shorted to ground.
I think they made an error drawing this diagram.
See the attachment for the better version.

---

mixer.JPG
Description:	Removed the wire from the left connection of P4 to GND
Filesize:	113.47 KB
Viewed:	213 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[metalboxhead]

Thanks Grumble. Disconnecting that pot from ground helped keep the gain under control and it no longer drops back to 0 when turned all the way up.

When I run a couple waveforms from a WMD Spectrum in everything looks and sounds perfect on my scope. However, when I have a waveform and a kick or snare from a Mutable Peaks, if the feedback pot is set over 50k I get a lot distortion which is noticeable on the drum sound. The scope is saying it's hitting less than 20V peak to peak. Any ideas?

[metalboxhead]

I found the issue I was hitting the output voltage max limit.

http://www.ti.com/lit/ds/symlink/tl074a.pdf

Increasing the output resistor to 2.2k / 3 worked

[AlanP]

Had a suspicion it'd be to do with the opamp's output limits.

I'd say tweak that last resistor so that the last 10% of the feedback pot's rotation is distortion. Options are nice.

[Grumble]

I had no idea what the amplitude of your input signal is, so I just concentrated on the diagrams where the only error was the wire from pot to ground.
But I understand that the opamps used are not reel to reel types, if they were than the amplitude of the output signal would have been wider.
Good you figured it out yourself, must feel good!

[metalboxhead]

Yeah am happy and a lesson learnt, study the spec sheets!

Thanks for your help.