MPP

Mr marsh, you're right about the distortion mechanism, but with mc, signals being so low that you only work at a ver narrow part of the transfer-curve, distortion will be so low that it's unmeasurable from the circuit, swamped by the performance of the cartridge. in simulation and in practice, this is the best circuit I/we have ever had.
 
We need a high quality tone control and a mono sum rumble filter i think.
Of cause we also need a bypass switch and it needs an expert to adjust the EQ.
I have a Cello Palette. It makes bad recordings sound good but i do not use it any more.
It is just so huge and the many cables when you have it close to the listening seat.
Also it is quite noisy.
Maybe a Quad tild control is a good compromise.
 
Joachim, would it be possible for you to post a curve trace of Exicon and/or Toshiba Lfets? It needs to be Vgs vs Id to very low currents and have logarithmic scales.

We need this because we can use it to make accurate Lfet models in the crossover region.

If the VDMOS simulation model accurately models subthreshold conduction for Lfets, then we will have very accurate models.
 
The best would be the 2SK1056/J162, because Cordell already has SPICE models for those. If I can see the low-current Vgs behavior, I can just add in the subthreshold parameter and ideally they should be accurate in the crossover region, which AFAIK is the last major complaint of Lfet models.

Any other popular Lfet you can add in will help greatly in getting the ball rolling to get all these devices modeled properly.