do i find another better F5

[john-china]

look at f5 circuit.
when i disconnect GND from resistor of JFET souce pin, i found better performance in my simulation circuit.

in orignal f5 circuit , i got distortion =0.02% ,gain = 5 , dc offset=250mv
in no-ground f5 circuit , i got distortion=0.009%, gain=0.8(too small) dc offset = 80mv


BTW, my attachment pic is not mine , which is from website.
i use multisim from NI to simulate F5.

can somebody tell why ,which one is better.

[analog_sa]

I don't have the original circuit but looking at your attachment it seems you have increased the negative feedback, hence the reduction in gain and distortion.

[Magura]

If you take a look at your figures, you'll see that the distortion per dB of gain, is higher in the 0.8dB gain version.



Magura

Edit: Not that I see a reason to lower the distortion of the original circuit??