Some other Source Follower Configurations

[kinku]

Is there any Good Samaritans who knows LT spice ,can design a BF862 version of the follower with TCS?

 

[AndrewT]

What's "TCS"?

 

[kinku]

Taylor current source. Patrick made the designs with other JFETs.Post #473

 

[RCruz]

Here you have a BF862 sim made by ELFishi:

 

[RCruz]

And another link: http://www.diyaudio.com/forums/solid-state/296335-ltspice-models-worst-bf862-jfets.html#post4821746

 

[kinku]

Rcruz that was an awesom link. But I don't know how to use LT spice . So finding the resistors value etc using LT spice is challenging for me.But will give it a try this week.

 

[kinku]

How does this look like. Can someone look at it and let me know.

 

[EUVL]

DC offset is 0.11V.
AC Current through R3, R4 not nearly equal amplitude.


Patrick

 

[kinku]

I guess you left clues to fix it in original spice model itself. Thanks, Patrick!
Did I nail it this time?

 

[EUVL]

If you accept 32mV offset, yes.
But it is immaterial, since you need to trim on circuit anyhow, depending on Idss.

Not so painful after all.


Patrick

 

[ElFishi]

Here's another source follower:

A JFET Triadtron after John Broskie. It has a Zout of 28 mOhm (mind the milli). I prefer my BF862 to be cascoded as they tend to gate leak with higher Vds.

Let me know your thoughts.

 

[EUVL]

Do you think it will be DC stable ?


Patrick

 

[ElFishi]

Sim says DC drifts with temperature, if that's what you mean.

A test circuit has a bug and currently I don't have time to hunt it down.

 

[EUVL]

Many of these circuits, including the TCS, do not always justify their complication when driving high impedance loads.
You can achieve the same performance by just using 2 pairs of matched NJFET in parallel.
And they, when thermally coupled, have very stable DC.

They come to play when you want to drive 30 ohm loads, as in the DAO.


Patrick

 

[ElFishi]

I've come to play with low Z buffers when ZenMod said you need one for his IronPre. But I'm not clear how low it needs to be.

 

[EUVL]

A TCS follower is push-pull, like a complementary follower.
So it halves the Zout of a NJFET follower.
Using 2 pairs of NJFET in parallel achieves the same.
The only difference is even harmonic cancellation in push pull.

My first DIY preamp using 5 pairs of 2SK170/2SJ74 in parallel.
Zout is < 3 ohm.


Patrick

 

[kinku]

Many of these circuits, including the TCS, do not always justify their complication when driving high impedance loads.
You can achieve the same performance by just using 2 pairs of matched NJFET in parallel.
And they, when thermally coupled, have very stable DC.

They come to play when you want to drive 30 ohm loads, as in the DAO.


Patrick

The reason why I got attracted to TCS follower is predominant Second harmonic distortion.
What is the type of distortion in B1 and 2 Pairs of matched Jfets in parallel? Anyone measured or simulated them?

 

[EUVL]

Use Spice.


Patrick

 

[kinku]

Patrick ,the current sensing SE NJFET follower also uses Push pull topology?does it uses NFB?

 

[ElFishi]

Do you think it will be DC stable ?


Patrick

OK, for all practical purposes it is DC stable. But it needs a capacitor between base and collector of Q1 to not oscillate. I used 1nF.