What to do with those JFETs?

[rankot]

Recently bought 100 MMBF5457 JFETs from Aliexpress. Measured them with transistor tester and they are all in range of 0.10-0.15mA for I_dss and for -0.1V for V_gs(off). Any suggestions what to do with them? ROG JFET calculator says they can reach some 48 times of amplification, but for small signals only, since they will clip if input is bigger than 0.1V, if I understood that topic well. http://www.runoffgroove.com/fetzervalve.html

[merlinb]

Recently bought 100 MMBF5457 JFETs from Aliexpress. Measured them with transistor tester and they are all in range of 0.10-0.15mA for I_dss and for -0.1V for V_gs(off).

You are not measuring what you think you are measuring. That's not V_gs(off) or Idss, it's just a gate test voltage used by your analyser, and the resulting drain current.
The true figures for those JFETs will be larger. You can measure them with a multimeter and a 9V battery.
Looks like they should be good for most pedal JFET applications and are not too different from the popular J201 (maybe a little more selection needed).

[antonis]

ROG JFET calculator says they can reach some 48 times of amplification, but for small signals only, since they will clip if input is bigger than 0.1V, if I understood that topic well. http://www.runoffgroove.com/fetzervalve.html

You've understood it well enough but your measurements are not right - as Merlin already told you..

More gain realistic values should be 15-20 (marginally up)..

[PRR]

> That's not Vgs(off) or Idss, it's just a gate test voltage

Still, it is likely Idss is under 0.25mA. (so not a 5457 spec.) Vto is harder to guess.

But I thought these little testers *would* swing to near zero both ways and display more-right Idss and Vto numbers? I don't see that here.

[rankot]

Recently bought 100 MMBF5457 JFETs from Aliexpress. Measured them with transistor tester and they are all in range of 0.10-0.15mA for I_dss and for -0.1V for V_gs(off).

You are not measuring what you think you are measuring. That's not V_gs(off) or Idss, it's just a gate test voltage used by your analyser, and the resulting drain current.
The true figures for those JFETs will be larger. You can measure them with a multimeter and a 9V battery.
Looks like they should be good for most pedal JFET applications and are not too different from the popular J201 (maybe a little more selection needed).

Merlin, I have compared measurements with ROG method (I have a dedicated PCB for that) and those with GM328A and they are NOT very similar for TO-92 JFETs. I simply don't have a way to measure those little fellows that way. But, for the sake of science, I will solder legs to one of them and measure with ROG method and see what happens. When I have measured 2SK117GR with GM328A, it shown 0.41mA for I_dss and 0.33V for V_gs(off), while ROG measurement PCB shown 4mA and 0.62V. Quite different!

[rankot]

OK, I have just done it. Results:
GM328A: 0.16mA 0.1V
ROG: 1.1mA 0.36V

Still very low V_gs(off) (out of specs), and I_dss is near the low end.

[Rob Strand]

When I have measured 2SK117GR with GM328A, it shown 0.41mA for Idss and 0.33V for Vgs(off), while ROG measurement PCB shown 4mA and 0.62V. Quite different!

GM328A: 0.16mA 0.1V
ROG: 1.1mA 0.36V

From what I can see the GM328A does not produce VP ad IDSS.   

The V and I  measurements are just one point *somewhere* on the JFET's characteristic.   From the manual it speaks of a 680 ohm source resisistor.  To me that means it puts a 680 test resistor from the source to ground and connects the gate to ground and just measures the voltage across the 680ohm (and maybe the current down the JFET).  So if you calculate the V/I ratio of the numbers it produces,

    2SJ117GR:  Vmeasured / Imeasured  =  0.33V / 0.41mA = 805 ohm
    Dodgy MMBF5457= Vmeasured / Imeasured  = 0.1V/ 0.16mA = 625 ohm

and it's sure looks like the set-up implied by the manual with a 680 ohm resistor.

In the past you have posted IDSS and VP measurements from your ROG jig and they have looked OK.
At least you know what the ROG jig is doing.  The GM328A is just a black box.

Your MMBF5457's do look out of spec.  If you calculate r_min  = VP/(2*IDSS) using your ROG measurements you end-up with 163 ohms which is in the ball-park for that type of JFET.  So really they are just out of spec.   In the last few years we have seen a lot of low VP JFETs.

[rankot]

OK, now I have all the data - they are all very similar and all in this 1.0-1.1mA and 0.3-0.4V range. What to do with them?

[merlinb]

OK, now I have all the data - they are all very similar and all in this 1.0-1.1mA and 0.3-0.4V range. What to do with them?

Small Vp is usually a good thing for pedal builders. Use them like any J201, or use them for analog switching.

[antonis]

According to ROG optimal biasing, you can use them as high-input-impedance(1M)-medium-gain(11-15) amps..
(e.g. followed by another identical stage amp for a distortion circuit of about 41 - 47 dB gain.. )

[rankot]

Thanks guys!

I suppose I can't use them for a phaser? 

[Rob Strand]

I suppose I can't use them for a phaser?

Not ideal but I did manage to help someone get a phase 90 with low VP's working quite well a few months back.
The trick was to divide down the signal to the gate so the LFO swing wasn't too large.  Didn't notice any distortion on a guitar but quite possible it would show up on a keyboard. 

[merlinb]

I suppose I can't use them for a phaser? 

Yes you can, you can use them for anything. (If you do build a phaser, try to pick out the JFETs that have matched Vgs(off), and the highest Idss. The Idss does not need to be matched though.)