What's this JFET doing for me?

[passtheducky]

I cargo culted an input stage for a pedal I'm designing and was a little surprised when I measured it. It's a JFET (2sk208) set up as a source follower into an OPA1652 in an inverting topology.

The THD was something like -43db, mostly 2nd and 3rd harmonic. I fussed with the bias and was able to get it down to -65db or so. However, when I bypassed the JFET and measured just the op amp stages, that distortion was completely gone and THD went down around -100db and was all 3rd harmonic.

I understand I could get the input impedance a bit higher by keeping the JFET, but there is a really slim chance that this particular pedal would ever be the first thing your guitar sees. So I don't really see the advantage there. I guess there are people who would consider extra 2nd harmonic distortion as a feature, but that's not the goal for this particular pedal. I also didn't see any advantage in noise performance, but that might be at the limit of my measurement system.

Is there something I'm missing here that would persuade me to keep the JFET in this design, aside from "it might sound cool" ?

 

[JTEX]

Might I ask why you're using an inverting topology?

 

[passtheducky]

Might I ask why you're using an inverting topology?

Yea. That stage has some filtering and a gain < 1, as it has to drop the level a bit to go into a codec.

 

[JTEX]

I'd consider ditching the FET and using the op amp non-inverting. You can add a pad after it to drop the level as needed for the codec. Non-inverting is always less noisy than inverting, and you also get the high input impedance.