Determining if my 2N5457’s are FAKE

[awgearhart]

But in terms of future use, would they be applicable? Not just for the SWAH... although I am very excited for that...

[pinkjimiphoton]

i'm of the opinion that yeah, they can always be used for something. if they're completely useless in any other way, they often still work great as clipper diodes.

to be completely honest, if i have to deal with jfets i usually opt for ic's. i hate dealing with all the crap to match them and test them and stuff... i much prefer to just be able to stick a part in and have it work, and to this day don't find personally that jfets sound more "tube like"... and i despise the way they hard clip, even uglier to my ear than normal transistors. they sound good unTIL they clip... kinda... but i've yet, after easily 600 circuits under my belt, to find any circumstance where i really cared for jfet "tone". jmo, ymmv, and really should

pretty much tho, with jfets, if ya don't need to match them, they should work fine. i would add a trimmer to bias each one tho... one more reason i hate dealing with them.

[Eb7+9]

Is it possible to tell just by my Vgs(off) and Idss readings?

your Idss values are reasonable for a 2n5457 but not your Vgs(off) ones ...
so, that's a labeling or calibration mistake

even with bone fide 2n5457's I measure and catalog these two numbers
which can lead to things like matched sets for use in differential pairs and phasors, and also low/high headroom constraint indexes for use in various single-supply SE circuits

just the nature of jFET's ...

where you use which is up to you the circuit designer

if you want to use a jFET in a current source capacity Idss is of course of primary importance ... if you want to use a jFET in a single-supply voltage follower/buffer capacity then Vgs(off) is a primary variable there ... and same for OD circuits, etc ...

the label on the front is almost secondary in nature at that point, since many/most of the various jFET's we use in low'ish voltage applications have overlapping parameter rectangles anyway

---

at least now we have a slightly more precise basis for establishing heuristic comparisons (ie., w/o need for circuit math)

of course, you can do one better by using a signal generator and scope on a very simple follower circuit
making sure (ie., choosing Rs appropriately) to never exceed the device's Idss
(Rs = Vdd/IdssMAX works as a safe and acceptable minima)

removing the Drain load and insuring Rs is chosen large enough just simplifies the test results ...
you should see obvious one-sided clipping here

the same clipping mechanism applies in the full voltage gain stage later when a Drain load is re-introduced
they occur independently as they occur at opposite ends of the resultant large-signal output transfer curve

this way you should be able to correlate in-circuit behavior with your prior dc testing efforts
and then conclude something about "predictable" input headroom

---

if you need pictures to go with that take a look here:

http://www.lynx.net/~jc/transferCurvature-TubeSimulation.html

---

if you're curious you SHOULD do/re-do all this and previous dc testing at Vdd = 18v
confirming Vgs(off) is hard fixed
and Idss nearly the same

this is for your own edification

if you go there make sure to wear safety glasses just in case these suspected fake devices can't handle the 18 volts
(it's unlikely, but ... who knows)

a real 2n5457 should handle 25volts without issue here - just saying
good luck

[Rob Strand]

But in terms of future use, would they be applicable?

Some circuits will work but most will need a part charge or adjustment to compensate for the low Vgsoff.

If you match the JFETs you can get a phaser to work but you will need to reduce the swing of the LFO by tweaking a part or two.

[awgearhart]

It sounds to me like it's possible to use these fakes depending on the circuit. I know to socket the transistors, so if they don't work with the particular circuit, I'll need to get MMBF5457's in their place.

Anyone have any problems with the MMBF5457's? Other than them being SMD and not in the TO-92 package?

[pinkjimiphoton]

hell, if you gotta BUNCH of these fakes, let me know, i can use them for clipper circuits if the diode junction part is functional. i gotta design i had used 5457/5458's in just as clippers, getting harder to find by the day. <pretty much ANY jfet will work as a clipper, i've found.... work great with shotkes like bat41 or 1n60, or ge's like in  zen drive.... little brighter sounding than mosfets.... more broken glass, less molten purr>