Weird Results Measuring JFETS?

[jafo]

...for the Fetzer Valve. These are pretty bizarre. I think. Well, I guess. Maybe. They certainly seem to fsck up my plans for substituting MPF102s for 2n5457a in my next build, a Supreux Deux.

First, an MPF102 from Rat Shack. Per ROG's article, I should get in the neighborhood of
average Vp: -2.34V (spanning from -1.76 to -3.11)
average Idss: 5.65mA (spanning 3.91mA to 7.15mA)
average Rs: 344 ohms
average Rd: 688 ohms
average Gain: 1 times or 0 dB

Instead, I got
Vp (V): -3.478
Idss (mA): 9.98
which gave me
Rs (ohms): 289
Rd (ohms): 184
and a "gain" of -9.9 dB. Yikes!

The second MPF102 gave me
Vp (V): -3.303
Idss (mA): 8.97
which calls for
Rs (ohms): 306
Rd (ohms): 240
and an alleged gain of -8.1 dB. Yikes again!

Is this likely? The JFETs are from the Shack, but they're stamped with Fairchild's seal of approval...

Next, I grabbed a random J201 I bought from the forum store. This gave me:
Vp (V): 0.80
Idss (mA): 0.50
Rs (ohms): 1328
Rd (ohms): 13320
and a gain of 14 dB. Much better, and completely consistent with other J201s, but prolly way too high for the SD. Still, this confirmed that my multimeter and jig were working correctly.

I don't mind a gain of 1 or 2 dB, but negative friggin' nine?! Yeesh. Then again, these results are the, err, result of biasing in a very specific manner which need not be optimal -- they're tuned for tone, not gain. But still...

[waltk]

I think the ranges for Vp and Idss cited on ROG were just for their limited-sample test batch.  According to the datasheet, you could theoretically get values far out of this range.  I've had MPF102s with values like yours also - so it's not unheard of.  J201s and 2N5457s are much closer in specs, so you should be able to sub j201s for the 2N5457s.  You should always socket JFETs because of their variability - that way you can test different combinations to get the performance you're looking for.

[jafo]

Words to DIY by. ty

[ubersam]

Try the J305s. The V_GS(off) and I_DSS specs are pretty close to the 2N5457 specs.

[PRR]

> Per ROG's article

It is highly unlikely you would want Rd of a few hundred ohms.

Or that you really want to run a high Idss device near Idss for a voltage-gain stage.

Or that you want a broooooad-spec device like MPF102 (0.5V-7.5V) as a guitar-amp input stage.

To pose a rude analogy: you don't carve a guitar out of random chunks of firewood.

There's other ways to design a JFT stage, including the approximate triode-law correction. But since the ROG Fetzer is well documented and well-regarded, and suitable JFETs seem to be available, why not start with suitable lumber instead of random logs?

[jafo]

There's other ways to design a JFT stage, including the approximate triode-law correction.

Err, isn't that what the Fetzer method aims to accomplish?

and suitable JFETs seem to be available, why not start with suitable lumber instead of random logs?

Just trying to use what I have on hand while I save up my pennies for what's suitable (Rat Shack really drains your wallet!)... and discovering how shockingly unsuitable what I have on hand actually is.