P channel JFET as audio switch

[diydidi]

P channel JFET as audio switch
Hi.
I recently built a 4 channel audio switch for a client. I was in a rush.
I decided to use a circuit described on the net “Audio switch notes”. It uses a P ch JFET. See the first of the two attached circuits.
It works great except that the signal gets clipped at louder transients.
Im wondering if this is caused by the absence of a gate diode in the first attached circuit. The second attached circuit has the diode along with an N ch JFET.
Could this be the cause of premature clipping of the signal?

[Jazz P Bass]

Is the clipping symmetrical?

[nickb]

Yes, you need the diode to prevent the gate becoming forward biased. Also, you ideally bias the FET so that it can never be turned off by big signal swings otherwise you get a sort of crossover distortion. That's the point of the Vb in the original circuit. For small signals it will OK to leave it at zero.

[diydidi]

Is the clipping symmetrical?

I will check and report back. Client only brought it back today

[eschertron]

Yes, you need the diode to prevent the gate becoming forward biased. Also, you ideally bias the FET so that it can never be turned off by big signal swings otherwise you get a sort of crossover distortion. That's the point of the Vb in the original circuit. For small signals it will OK to leave it at zero.

Does some of this have to do with the Vgs_off of the fet? My rudimentary understanding is that the signal swing can't be bigger than the 2x Vgs or the fet turns off under best conditions.

[diydidi]

I found the problems.
Wave was clipped on the bottoms.
Decided to bias G &S at 1/2 supply.
Then added a diode on Gate. Problem solved.
Thank you all.