jfets in opamp circuits
Jul 24, 2009 at 2:01 PM Thread Starter Post #1 of 10

sanderx

100+ Head-Fier
Joined
Apr 27, 2009
Posts
115
Likes
12
I've been thinking about a new amplifier project lately. I have also been thinking about using jfets in an otherwise op-amp amplifier for buffering and switching. Anybody got pointers to something similar / useful hints? I have so far really only read about transistors and not used them much.

The goal is high gain minimal distortion amplifier with switchable bass boost that would be by design tolerant of component errors / mismatches.
 
Jul 25, 2009 at 12:48 PM Post #2 of 10
Discrete opamp? I had a good article, but I think it's on my other computer; will look for it. In the meantime, Google "JFET discrete opamp" for preliminary reading.

In terms of current schematics, the Headwize library has a number for you to browse. Amb's ubiquitous beta22 is also excellent and has JFET inputs.
 
Jul 25, 2009 at 9:30 PM Post #3 of 10
Well, not so much a discrete op-amp as a hybrid jfet and op-amp based amplifier. A mixture of trying to get best of both worlds and getting extraordinary performance without it filling the table or breaking the bank. (jfet input)->(op-amp gain)->(driver)->(120ohm hp out)

The idea started from using jfets to get rid of the need for dual-gang switches, went through adding an input stage to ppa v2 and is now slightly somewhere else
 
Jul 26, 2009 at 2:45 AM Post #5 of 10
there are plently of jfet input monolithic op amps: opa627/37, ad743/5, ad8210, opa134 for just a few recognized in audio apps, the opa604 even claims all of the V gain comes from jfets and had a audfiophile following in the past but gets less respect today

a good match is to use a monolithic jfet input op amp as the input with a discrete or integrated buffer or high current output op amp in a multiloop composite amplifer as long advocated by Walt Jung
 
Jul 26, 2009 at 4:35 AM Post #6 of 10
The following was one of the concepts I toyed with in the early stages of Mini³ development. The idea is to combine a discrete JFET (2SK170) differential input stage (for low input bias current) with a high output current bipolar opamp. I think it would have worked very well, but the idea was scrapped because of insufficient PCB space.

JFETinput_concept.png
 
Jul 26, 2009 at 11:38 AM Post #8 of 10
^If you're going to use a Current Feedback Input stage why not hammer in a monolithic buffer such as the LMH6321 for better performance?

EDIT: You need a symmetrical complementary current mirrors for dc shifting though.
 
Jul 26, 2009 at 12:33 PM Post #9 of 10
Quote:

Originally Posted by TzeYang /img/forum/go_quote.gif
^If you're going to use a Current Feedback Input stage why not hammer in a monolithic buffer such as the LMH6321 for better performance?

EDIT: You need a symmetrical complementary current mirrors for dc shifting though.



It's a no feedback input stage, just a buffer to cope with the low input impedance of the inverting amp.
 

Users who are viewing this thread

Back
Top