wanted a dual differential amplifier with all fet

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
The Borbely amp has a dual (complementary) differential input stage. If you want an amp that is complementary differential from input to output, check out:

http://www.diyaudio.com/forums/showthread.php?s=&postid=334963&highlight=#post334963

(At then end of the post is a zip file that you can download.)

Also, James Bongiorno had a design in Audio Amateur about 20 years ago (Ampzilla 200???) that was complementary differential from input to output, although it was 100% bipolars.
 

fab

Member
Joined 2004
Paid Member
thanh said:
Borbely's project has not got value of components.I is not good at designing JFET.Thank Charles Hansen and fab!
I still need that so i still need all your help!Again,thanks a lot!


Maybe you did not check in the proper site menu. Check in more precisely in http://www.borbelyaudio.com/index15.htm. There are the "SERVO-50" and "DC-102". It includes all information including value of components.
 
disctete FETs are really not good in "cookbook" projects, even professional engineers avoid them in mass production designs because of the wide individual and lot-to-lot parameter variations - simply copying a circuit based on fet part # alone could easily leave you with something that doesn't work at all due to bias errors

for diy and "hand crafted" audiophile projects they can be good but you must know enough to measure sort/match and individually trim bias as appropriate for each stage in your circuit
 

fab

Member
Joined 2004
Paid Member
jcx said:
disctete FETs are really not good in "cookbook" projects, even professional engineers avoid them in mass production designs because of the wide individual and lot-to-lot parameter variations - simply copying a circuit based on fet part # alone could easily leave you with something that doesn't work at all due to bias errors

for diy and "hand crafted" audiophile projects they can be good but you must know enough to measure sort/match and individually trim bias as appropriate for each stage in your circuit

Yes, and that is why I have indicated the "SERVO-50" and "DC-102" which only includes jfets at the input stage.
 
Borbely's project are way over priced. Heck you can buy a used Krell or a Levinson off eBay for what this guy want for his parts.
I believe that www.audioXpress.com still have some of his PCB's from passed project. You could change inputs from bipolars to FET that's is not a big deal.

Now if we only could get Charles Hansen to offer a PCB for his no feeback amp design, that would be cool. :D
 
this is JFET's formula.
Is not it a absolute formula?
 

Attachments

  • aaaaaaaaaaaaaaa.jpg
    aaaaaaaaaaaaaaa.jpg
    1.6 KB · Views: 497
a fine formula, but still an approximation that is limited to large Vds and “long channel” fets

the problem is that Idss and Vp can vary 5:1 for the same part # between batches - and smaller but still significant variations within a lot

you have no guarantee that you can select what you want yourself from any given lot of the same part # fet

you can buy fets sorted by the manufacturer to smaller ranges but fets are inherently more sensitive to normal process variations than most of the bipolar Q we use
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.