Balanced SE All-JFET Line Aamplifier

Stefano · 18th March 2006, 11:32 AM

Hi folks!
Now that we all are wandering about j-fet amps (Zen9) and pre, I would like to see a j-fet version of the balanced line stage (BoSoZ) and I've much appreciated the Krazy project, but has anyone tried to build someting similar to E. Borbely's EB-2000/402?
This is the link to the kit version:

http://www.borbelyaudio.com/eb2000402.asp

and this is the link to the original article (on audio amateurs):

http://www.borbelyaudio.com/adobe/ae599bor.pdf &
http://www.borbelyaudio.com/adobe/ae699bor.pdf

What do you think about?

flg · 18th March 2006, 04:25 PM

I have also been interested in something along that idea. I wanted close to 25V or more output swing though to drive a Circlotron follower output and that becomes difficult with JFETS. Naturally I don't want any caps in the signal line so I gravitated toward a folded cacode to shift the level for the output stage. It simulates pretty good open loop or with feedback, but it looks to much like an existing patent of the one and only's (the feedback version). I guess that is no suprise... Thank you very much N.P.. Currently I have an unfished ZV9 proto to complete and software for PC soundcard audio analysis to play with or I would be building it...
Grey once said he had some similar ideas, who else. He may have some comments if we can pull him out of the JFET distribution bis...

Stefano · 20th March 2006, 07:10 AM

I'm thinking of using lovoltech power fets as followers, by the way could you any schema of your own project? It seems to me very, very intristing, 'couse I like very much quasi-complementary topologies...

JoshK · 30th March 2006, 08:30 PM

Quote:

Originally posted by Stefano
... I would like to see a j-fet version of the balanced line stage (BoSoZ) and I've much appreciated the Krazy project....

Ditto! Ultimately, I think we are going to have to DIY! But seeing as I have never really modified someone else's design before (newb), this will be an undertaking. I also don't have much spare time these days.

What I'd really like to see is a balanced in/out line stage (after the BoSoZ) with J-Fet transistors and little to no gain. The pieces are all here but need to be put together by us.

If those in the know wouldn't mind giving a couple of hints about what kind of J-Fets would be best, it would be most appreciative. I have about 20 of the Lovoltech's and am willing to use them or acquire something more suitable.

flg · 30th March 2006, 11:03 PM

Well, I've been doing this type of amp design for a few years now. I was not much of the solid state forum member but Ive been doing that, speakers and systems on and off for my whole life. I work in the semiconductor industry during the day. There is no better tool, other than your brain, than something like Pspice or some circuit sim package. I just recently got a hold of Pspice and I have an excelent mentor, it's great. In my opinion, if your prototype is amplifying in the first 15 minutes, your pretty good at what your doing. With Pspice, your circuit is working and close to optimised, before you build it.

It's not perfect, but, by the time you build it, you know the circuit very well from playing on the simulator. All the guessing is transfered into trust in the simulator and you get much more time to debug and optimize sitting at your computor. Rather than cursing at the circuit board. You get more time to listen also. A mandatory concern in this forum.
That having been said. All I have at this moment is a Schematics file. A Cadence freeware front end for Pspice. I can't seem to get it into something normal like jpg or something. I will attach it and I will try to get a real pic on here shortly. Well, It does not like .sch files... I'll redraw it in capture... Sorry
I should have said up front: THIS IS A WORK IN PROGRESS!!
Is it to the thread? Is it Balanced in? SE????? ??? All JFET?? well, the signal path at least... Line AHHH Well, Right now I think the gain is about 300. Right there it says that this guy may not be ready for this forum...ooops, Like I said, a work in progress... I have to throw away some parts and maintain the linearity, rather than use 40db of feedback... I think at about 3V out @1khz the simulated THD is .007% and -3db is 750khz+.
this is very similar to my Circlotron front end... What do you poeple think???

flg · 31st March 2006, 12:01 AM

Lets try again... The work in progress balanced in, balanced out line amp.... Sorry, the quality is a little strange. Still trying to get some better conversion results from the Cadence output...

JoshK · 1st April 2006, 01:43 AM

Thanks for the advice....I d/l'ed Pspice eval version and plan to play around.

flg · 3rd April 2006, 11:59 PM

Well, I should have been doing something else this weekend but after all this preamp jazz, I decieded to take a look at the BOSOZ, the Aleph P 1.0 and the aleph P1.7. I mostly used the simulator and at least to begin with I looked at all three as published. I will need one soon anyway, especially one that will go SE to Bal. but I am also still on the market for ideas to drive the current design... I looked at the L also but I really need that bal drive for the amps. I have been playing with the Circlotron topology but have not settled on a drive mechanisim. I currently think a circuit like the XXX899 Pat. would do well. Similar to the one I posted without the output current source...
Anyway, my initial results(without any data here!) were something along these lines: All three will do the 25V P-P Balanced level I need. Woaa. I had never noticed that before??? Both the BOSOZ and the P1.0 are disticntly Zen in nature but can produce these high levels at less than .05%(without quoting any numbers here) Thd (from Pspice). And with really low, low level numbers, increasing steadily with higher level. Add an abundance of current sources (The Aleph P1.7) and the Thd is about an order of magnatude lower!!!
My initial conclusions: With the low distortion and great headroom these circuits have, (on the sim at least) all of them are wort the listen. My folded cascode JFET Circlotron version gets about .005%THD @ 1kHz, @ 10VP-P and .06% THD @22VP-P into 8 ohms. My initial tests using the BOSOZ as published, with the Circlotron output did a little better

jam · 4th April 2006, 03:53 AM

You could try this..............with mosfets and a jfet differential.

Jam

flg · 5th April 2006, 12:12 AM

Yes, very similar to what I've been experimenting with, and it looks like it wll work, but I really like the idea of the ZEN simplicity. Even what I posted has 9 devices. The 4 device solution looks more like what I was looking for though...
The BOSOZ driving the Circlotron follower seems to be shaping up better than the folded cascode, on the sim...

18th March 2006, 11:32 AM	#1
Stefano diyAudio Member Join Date: Jun 2002 Location: Cagliari, Sardinia	Balanced SE All-JFET Line Aamplifier Hi folks! Now that we all are wandering about j-fet amps (Zen9) and pre, I would like to see a j-fet version of the balanced line stage (BoSoZ) and I've much appreciated the Krazy project, but has anyone tried to build someting similar to E. Borbely's EB-2000/402? This is the link to the kit version: http://www.borbelyaudio.com/eb2000402.asp and this is the link to the original article (on audio amateurs): http://www.borbelyaudio.com/adobe/ae599bor.pdf & http://www.borbelyaudio.com/adobe/ae699bor.pdf What do you think about? __________________ Let's make something special

20th March 2006, 07:10 AM	#3
Stefano diyAudio Member Join Date: Jun 2002 Location: Cagliari, Sardinia	I'm thinking of using lovoltech power fets as followers, by the way could you any schema of your own project? It seems to me very, very intristing, 'couse I like very much quasi-complementary topologies... __________________ Let's make something special

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Balanced jfet buffer for Vout DAC?	jonners	Solid State	7	22nd October 2010 07:57 AM
FS : Borbely EB2000-402 Se Balanced Jfet Lineamp	ANALOG GUY	Swap Meet	1	12th July 2006 04:40 AM
Balanced Zen v5 + balanced Zen line stage	Adrie	Pass Labs	0	30th May 2004 05:27 PM
Zen Balanced Line Stage Balanced vs Unbalanced	macka	Pass Labs	28	11th December 2002 06:18 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

18th March 2006, 04:25 PM	#2
flg diyAudio Member Join Date: Jul 2005 Location: North East	I have also been interested in something along that idea. I wanted close to 25V or more output swing though to drive a Circlotron follower output and that becomes difficult with JFETS. Naturally I don't want any caps in the signal line so I gravitated toward a folded cacode to shift the level for the output stage. It simulates pretty good open loop or with feedback, but it looks to much like an existing patent of the one and only's (the feedback version). I guess that is no suprise... Thank you very much N.P.. Currently I have an unfished ZV9 proto to complete and software for PC soundcard audio analysis to play with or I would be building it... Grey once said he had some similar ideas, who else. He may have some comments if we can pull him out of the JFET distribution bis...

30th March 2006, 11:03 PM	#5
flg diyAudio Member Join Date: Jul 2005 Location: North East	Well, I've been doing this type of amp design for a few years now. I was not much of the solid state forum member but Ive been doing that, speakers and systems on and off for my whole life. I work in the semiconductor industry during the day. There is no better tool, other than your brain, than something like Pspice or some circuit sim package. I just recently got a hold of Pspice and I have an excelent mentor, it's great. In my opinion, if your prototype is amplifying in the first 15 minutes, your pretty good at what your doing. With Pspice, your circuit is working and close to optimised, before you build it. It's not perfect, but, by the time you build it, you know the circuit very well from playing on the simulator. All the guessing is transfered into trust in the simulator and you get much more time to debug and optimize sitting at your computor. Rather than cursing at the circuit board. You get more time to listen also. A mandatory concern in this forum. That having been said. All I have at this moment is a Schematics file. A Cadence freeware front end for Pspice. I can't seem to get it into something normal like jpg or something. I will attach it and I will try to get a real pic on here shortly. Well, It does not like .sch files... I'll redraw it in capture... Sorry I should have said up front: THIS IS A WORK IN PROGRESS!! Is it to the thread? Is it Balanced in? SE????? ??? All JFET?? well, the signal path at least... Line AHHH Well, Right now I think the gain is about 300. Right there it says that this guy may not be ready for this forum...ooops, Like I said, a work in progress... I have to throw away some parts and maintain the linearity, rather than use 40db of feedback... I think at about 3V out @1khz the simulated THD is .007% and -3db is 750khz+. this is very similar to my Circlotron front end... What do you poeple think???

1st April 2006, 01:43 AM	#7
JoshK diyAudio Member Join Date: May 2004 Location: Toronto, ON	Thanks for the advice....I d/l'ed Pspice eval version and plan to play around.

3rd April 2006, 11:59 PM	#8
flg diyAudio Member Join Date: Jul 2005 Location: North East	Well, I should have been doing something else this weekend but after all this preamp jazz, I decieded to take a look at the BOSOZ, the Aleph P 1.0 and the aleph P1.7. I mostly used the simulator and at least to begin with I looked at all three as published. I will need one soon anyway, especially one that will go SE to Bal. but I am also still on the market for ideas to drive the current design... I looked at the L also but I really need that bal drive for the amps. I have been playing with the Circlotron topology but have not settled on a drive mechanisim. I currently think a circuit like the XXX899 Pat. would do well. Similar to the one I posted without the output current source... Anyway, my initial results(without any data here!) were something along these lines: All three will do the 25V P-P Balanced level I need. Woaa. I had never noticed that before??? Both the BOSOZ and the P1.0 are disticntly Zen in nature but can produce these high levels at less than .05%(without quoting any numbers here) Thd (from Pspice). And with really low, low level numbers, increasing steadily with higher level. Add an abundance of current sources (The Aleph P1.7) and the Thd is about an order of magnatude lower!!! My initial conclusions: With the low distortion and great headroom these circuits have, (on the sim at least) all of them are wort the listen. My folded cascode JFET Circlotron version gets about .005%THD @ 1kHz, @ 10VP-P and .06% THD @22VP-P into 8 ohms. My initial tests using the BOSOZ as published, with the Circlotron output did a little better

5th April 2006, 12:12 AM	#10
flg diyAudio Member Join Date: Jul 2005 Location: North East	Yes, very similar to what I've been experimenting with, and it looks like it wll work, but I really like the idea of the ZEN simplicity. Even what I posted has 9 devices. The 4 device solution looks more like what I was looking for though... The BOSOZ driving the Circlotron follower seems to be shaping up better than the folded cascode, on the sim...

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)