The problem with jfets used as common source amplifiers is that their pinch off voltage (the voltage required to shut them off) varies greatly between devices of the same type. This means that the amount of voltage applied to the gate to put the drain in the center of the operating range also varies. We often compensate for this by swapping bias resistors until we get the voltage we want. The problem is that the gain of the common source amplifier also depends on the biasing resistors. So by adjusting the bias, we also change the gain (and the EQ as well, usually), and the circuit may sound different than intended.
The amount of difference it makes depends on what you are using the jfet for. If you are simply amplifying without clipping, and you can bias the jfet with enough headroom to accomplish this, you won't notice that much difference. But if you are trying to use the jfet for clipping, the difference will be noticeable, because the onset of clipping will depend on the gain and the bias.
This is what happens with Wampler's J201 circuits. They are designed to clip at a very specific voltage, which gives them the frequency response of different amplifiers. Rebiasing the jfets in these circuits often changes the response very drastically, either by reducing the gain (wimpy tone), or increasing it (oscillation). The trick to these is to buy big bags of jfets, and swap them into the circuit until you find one that biases correctly with the given resistors.
The opposite approach is used with the
fetzer valve by ROG. It's a similar idea, using jfets to emulate tubes, but if you look at the bottom of the linked page, there is a recipe for measuring the pinch off voltage and Idss of the jfet, and a calculator that helps you find the the right biasing resistors for the measured jfet. The circuit will sound a little different with different jfets, but it's not a high gain circuit so the difference will be subtle, and it's not a clone of a commercial unit, so it won't be 'wrong' in any case.
Note that none of this implies that the jfets in a given circuit have to match each other, just that they have to bias correctly. If you build Wampler circuits with identical gain stages or parallel jfets in a single stage, then the jfets have to match only because they each have to bias correctly in their given stages, not because the circuit depends on them matching each other.
Other circuits do require matching jfets.
Mu-amps require two jfets in a single stage. They should use matching jfets, but the match doesn't have to be all that close. Circuits like
phasers and
fet octave doublers require some of the jfets to be matched because their correct operation depends on both jfets affecting the signal identically.
There's plenty of good info on
how jfets work and
what to do with them on the web.
Locked
).
Any further info from knowledgable forumites would be gratefully received. Meanwhile, Vince has certainly whetted my appetite for the RAH.