WHY won't Q3 function with a 5457 in my BSIAB2??

[analog kid]

This has been driving me nutz long enough...so I'm asking for help from the experts who have experience with this circuit.
WHY won't a 5457 work in only one position (Q3 the bottom half of the second Muamp in the BSIAB)  J201's OR 5457/8 will interchange fine in EVERY OTHER Q position. But as soon as I drop one in Q3  , it's instant gating and sputtering. And  no the Q5 trimpot will not remedy this sputtering no matter how it's adjusted.
My only substitutions are a 250pf polystyrene rather than 220pf ceramic at input
And a 4.7k instead of 5.6 on Q5 (signal?)
My voltages on Q3 And Q4 with a J201 are
   D  4.2v           Q4 D 8.8
   S  .05v                S 4.51
   G   0                   G 4.32
And like clockwork With most any 5457 the whole stage changes to this
   D  1.5 +/-      Q4  D 8.8
   S  .22                 S 1.47
   G   0                   G 2.1
Please Note the change to Q4's S and G  when using a 5457  in Q3 position, that is NOT a mistype, the voltage drop between S/G actually does a flip-flop from the correct vltgs for Q4 when a 201 is in there.
-also to help with any assistance here's the link to the vltgs GGG has
http://generalguitargadgets.com/diagrams/bsiab2_voltage_diagram.gif
I'll also note that Q4 happens to be the only Q that i don't have socketed so I have a J201 in there which I can't take out. Now I've never heard anything about HAVING to change BOTH Q3/Q4 to 5457's if one is changed, but this is something I've thought about(?)  Mismatching J201 and 5457 has NO SUCH ill effect on the first MuAmp stage. so..... ??

[brett]

Hi.
Double-check that the pinout is DGS.  A few fets and mosfets are SGD.  You probably haven't made this mistake, but I mention it because this friend of a friend of mine.....

In my experience 5457s have Vgs(on) close enough to that of a J201 to substitute ok.  Better than 2N5485 or 86 for example.
Good luck

[Ed G.]

I don't know why, but sometimes fets won't bias right without the same type fet in the upper position. I was fooling around with different fet types and found the same thing. Moreover, it changed when it went from breadboard to PCB. Maybe some of the real experts can weigh in on this.

[spudulike]

http://www.geofex.com/Article_Folders/foolwfets/foolwfets.htm

[analog kid]

Thanks guys,  And thanks for telling me that Ed!

sometimes fets won't bias right without the same type fet in the upper position

I had been wondering if that could be a possibility , since I have never been able to switch Q4 to a 5457 at the SAME TIME as I was switching Q3 from J201 to 5457. (don't know why I left only that position unsocketed?)
I guess I hadn't really entertained the fact that them not being the same JFET causing the problem that seriously Since it does NOT seem to cause any problem with the first Muamp stage of Q1/Q2  (?)  Maybe there 's something that allows this on the 1st pair but not the second  . I will get a socket in there on Q4 and let you know.  OH, and what did you mean that it 'changed' going to PCB? Mine IS.

BTW , off topic, have any of you messed with getting rid of the 1M Dividers on the muamps and instead using just One pair of like 10k's on just the 1st Mu and a couple large R's from there to the Drains of the Mu's?? . ie; Noiseless Biasing  have been doing it and can tell there's something to it(lower noise,etc..) but can't seem to get the right R values to keep the 1/2v points at 4.5v! I keep getting over 5.5volts

[analog kid]

Oh yeah. That was definitely it!  Strange as it is, the 5457 weren't working in Q3 WITHOUT also having one(or even a 5458) in Q4! With a J201 there it just won't bias up.   I am VERY suprised that noone else has come across this and mentioned it in all the talk about the choice of J201's or 5457's in this circuit !!
I will likely try and see if the J201 / 5457 combo when reversed , will bias correctly. ie; the 5457 in the Q4 and J201 in Q3  ,  I will be 'befuddled' if it works that way but not the other