trimpots to match JFETS

[breather-resistor]

I recently read somewhere (I can't find it or I would link to it) that there is a way to use trim pots to match JFETS.

I am thinking about building some JFET phasers and I wondered if this would be applicable.  It seems a little too good to be true, and I am assuming until I hear otherwise that there are some downfalls to this method.  Obviously, it would cost a couple bucks to get the trim pots but not nearly as much as overstocking on JFETS just so that some of them can match.

Plus, you could get them to match EXACTLY with a meter.

Anyone dabbled with this? 

[petemoore]

  Are you referring to bias adjustment or Jfet matching [as in phaser Jfets]?

[R.G.]

Search function.

If it were as simple as just throwing in a trimpot, do you think that there would be JFET matcher circuits and long harangues about matching JFETs?

The short answer is that any JFET to be used as a variable resistor, like in a phaser, has a range of drain resistance which is about right, that being a few hundred ohms to what amounts to an open circuit. Alls you gots to do is drive the gate-source perfectly for this one JFET to match the others and you don't need to match. The problem is that each JFET then needs a custom DC bias on its gate-source to get it to the right static or no-signal condition, and also a custom size (and shape, but we won't get into that) of LFO signal to track what the other JFETs are doing.

You need at least two trimpots per JFET. And you need a way to set them correctly, which is no trivial task if you have nothing to compare them against. A simple listening test while you twiddle the trimpots is going to take a long time, and has a lot of blind alleys. You would need to build a custom tester circuit to get them to match EXACTLY (sic) with a meter.

Let me put it to you this way: If you're going to put a meter on your JFETs and try to get them to match, how about you build a smaller amount of circuitry (the JFET matcher) and use your meter to pick the ones that already match without needing a board full of trimpots? JFETs are cheap, under $0.10 each and often under $0.05 each. Trimpots cost money - and board space that's at a premium in a pedal and time diddling, and failures and drift.

I guess I have an opinion here, and it may be showing through. 

[breather-resistor]

RG,

I certainly considered that if it was as easy as a trimpot there would be no need for the JFET matching circuit.

However, it did occur to me that maybe people preferred the JFET matcher over the "trimpot method" for some of the same reasons that you mentioned like board space and cost.

I know it is a too good to be true idea but I figured it couldn't hurt.

I have used your Jfet matcher with success before and it looks like I will continue on that path.  The responsibility of adjusting 4 trimpots simultaneously seems much more difficult than plopping JFETS into your matcher on a bread board.