Page 1 of 1

JFET Biasing?

PostPosted: 08 Oct 2013, 04:15
by commathe
Hey everyone,

I have been running some simulations of JFETs in order to study a bit about how to bias JFETs properly and also to teach myself along the way. I'm having problem with some of my calculations though. I am currently able to accurately calculate the operating point of a simple JFET circuit so long as it only has a source resistor. However, I've noticed that as soon as the drain resistor gets large enough to limit the current to be lower than the expected current draw for some source resitor and Idss value, then the operating point of the transistor becomes elusive and strange to me and I'm unable to calculate it. I feel that it has something to do with changes in drain-source voltage drop (maybe gate-source too?) but I can't figure out exactly how to calculate it accurately.

Similarly, I am confused as to how to bias a JFET well enough as to basically negate the differences between individual components - similar to how you can by using a voltage divider bias network for a BJT.


Re: JFET Biasing?

PostPosted: 08 Oct 2013, 07:13
by Nocentelli
There's a lot of reading to do over at DIYSB, but unfortunately the search function over there is pretty poor. The gist seems to be that wide variances in the manufacture of JFETs mean that it is quite difficult to design a biasing scheme that will suit whichever individual JFET you pull out of the stash: Even JFETs from the same manufacturer vary wildly. The "solutions" appear to be either to pick a suitable source R, and use a trimmer on the drain for each individual JFET (like almost every ROG project) or pick suitable approximate source and drain resistor values, buy a massive bag of JFETs and individually test each one by sticking it in a test circuit and monitoring the drain voltage produced to find suitable individual JFETs : This is apparently how Wampler does it for his pedals.

Re: JFET Biasing?

PostPosted: 08 Oct 2013, 07:31
by commathe
Thanks, that helps a lot! I guess it also explains the popularity of MOSFETs over JFETs, though I haven't really touched upon MOSFET circuits yet. I am aiming to design things I can do atleast small runs (~50-100) of and the idea of that many trimpots makes me a sad panda.