Pedja Rogic's J-fet buffer for dummies.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
After my bad luck with the Les Sarge circuit, I decided to give Pedja Rogic's J-fet buffer a try....

Well it worked straight away, and is sounding very mature... It seems to hang with the music relentlessly, and there is none of the hash in transients, I could hear with my other preamp attempts, except for maybe the ECC88 buffer this is replaceing.

:D

So, I thought I will add it here for the future, and other noobs struggleing to get a good preamp going.

Pedja-Rogic-Buffer.gif


Pedja-Rogic-BufferB.gif


An externally hosted image should be here but it was not working when we last tested it.


http://www.pedjarogic.com/gc/index.html

Thanks PEDJA, and NUUK, for pointing me in this direction with your site and layout...mine is a little diffirent, trying to reduce wire jumpers, putting power connection on output side, and accommodating diffirent size caps I had... you can see that it is easy to adapt the outut filter caps to whatever size you have...
 
Why not use a dual matched JFET like the LSK/2SK-389? Also, a source resistor on the top JFET would let you null out the DC offset and you could eliminate the coupling capacitor.

For the cascode MOSFET, how 'bout something like the Supertex DN-2540, which is a depletion mode device which would make it self-biasing which would eliminate the need for the voltage divider?

For that matter, you could also use one on the current source and end up with no more parts than you're using now.

se
 
Sorry Steve you are getting ahead of me there... offset is minimal about 6mv... allthough there is a 2V DC spike at power on.

From what I understood, the lowpass was a neccesary element to get the circuit to work well with a chipamp which is internaly frequency limited...


What I would like to know is, if one should add decoupling caps to the power rails?

Will try to understand what you said later when I'm awake...
 
How do I know when I have a groundloop?

I suspect so, as the circuit is quiet when using earphones but have a slight hum with my My_ref C.

Could this be because of the lack of bypass caps...? once again, didn't see any in the schematic.

WHat would be the easiest tests for groundloop problem?

Input and output sockets are isolated from chassis.. have exactly the same earth decoupling network in the preamp as in the main amp... both connected to mains ground... no problem with my valve buffer, using same arrangement.
 
:)

Pedja Rogic JFET Buffer is a good circuit.


Here is my own approach for a DC-in and DC-out Buffer.
I use a one JFET + one BJT Follower.

At 5 mA, BC550C has VBE like +0.660V.
At like 1.65mA, JFET BF245A has VGS like -0.660V.

Add those, and I get 0.000 Volt offset at the output.

The BC550C does what it is best for: Provide current.
The JFET BF245A does what it is good at: Provide High Input impedance.

Circuit can be made very fast. 1 MHz is no problem!
Also shows very nice low distortion, without any feedback.
Most of the low level distortion is in fact 2nd harmonics.


Enjoy :cool:
lineup
Lineup Audio Lab
 

Attachments

  • lineup-zerobuffer.png
    lineup-zerobuffer.png
    9.8 KB · Views: 4,669
Thanks Steve... that worked 100%

Wow these sound good... shame about the crappy PSU.... totaly went minimalistic getting the other buffer to stop oscillateing...

Its like super stereo.... listeing to some CDs now, and its pretty freaky how diffirent they sound... lots of hidden detail comming out...

Brilliant!

Incredible bass, crystal treble... very atmospheric... its like a sound effect generator... its like some of the sounds are just apearing in the room... no way my speakers can make em :)
 
Nordic said:
Thanks Steve... that worked 100%

Wow these sound good... shame about the crappy PSU.... totaly went minimalistic getting the other buffer to stop oscillateing...

Its like super stereo.... listeing to some CDs now, and its pretty freaky how diffirent they sound... lots of hidden detail comming out...

Brilliant!

Incredible bass, crystal treble... very atmospheric... its like a sound effect generator... its like some of the sounds are just apearing in the room... no way my speakers can make em :)

Just came across this thread, anyway glad you sorted the problem:)
I've built a few chip amps including My-RefC.
If you like Pedja's buffer you should try building the rest of that amp including his discrete regulator for the LM3875.
Its still my favourite out of all the chip amps I've tried so far
 
Nordic said:
You reccon it could perform better than the My_ref C?


Only you can answer that, the My_ref C is a very nice amp but for me and my opinion Pedja's was better.
I find the My_ref C a little too smooth for my taste although theres nothing wrong with that and is probably the sort of thing some people like, to me Pedja's has more speed and detail.
I actually built Pedja's circuit but used LM3886 instead of LM3875.
I wouldn't use MUR860 diodes with that amp/discrete reg, they didn't sound as good for some reason with the regulator
 
Sorry about all the posts on your thread:)
One thing I wanted to try was removing the Zeners for the input stage of the My-RefC and fitting shunt regs but never got around to that yet.
Also today I may alter some of the track on the My_RefC amp and try out 4-pole T-network capacitors instead of the bog standard two pole ones
 
I'm still waiting on some better PSU caps for mine, but I did replace the 220uf caps close to the chip with low ESR Vishay caps... great great improvement by itself....

I do enjoy the smoothness of the My_ref, the extreme trebbles of the solidstate stuff is a bit novel to me, and generaly I can't listen to them too long, valve gear is the opposite...

Please post here all you like... I am still learning about how this circuit works (except for very well)

I know people are using bigger zeners, at higher voltage, with good results, I have not had the urge....but I have a number of components on order for it from TP... even ran it all this time with 1% resistors in 0.1% places...
 
Nordic said:
I'm still waiting on some better PSU caps for mine, but I did replace the 220uf caps close to the chip with low ESR Vishay caps... great great improvement by itself....

I do enjoy the smoothness of the My_ref, the extreme trebbles of the solidstate stuff is a bit novel to me, and generaly I can't listen to them too long, valve gear is the opposite...

Please post here all you like... I am still learning about how this circuit works (except for very well)

I know people are using bigger zeners, at higher voltage, with good results, I have not had the urge....but I have a number of components on order for it from TP... even ran it all this time with 1% resistors in 0.1% places...

I use all Rubycon ZL low esr caps for the lower values and currently Panasonic types for the 10000uf's.
I also don't like bright harsh although I don't like things veiled either;) a shunt reg should be much better than a zener for the input op-amps, I think (could be wrong) Mauro uses shunts to regulate the input stage for his latest amp.
I was going to use ALW super regs but feel thats a bit overkill for this little amp.
I've tried quite a lot of PSU caps in my UCD amp and found the 4-poles to be way superior to any of the others, only problem is they are not cheap
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.