The B1 thread is already bigger than 1500 posts. In this big pile of information I'm not able to find which JFET's can be used instead of the standard 2SK170/LSK170/2SK370 types.
Purpose of this thread is to make a list with suitable replacement JFET's for use in the B1 buffer. Any information on the performance of the replacement devices compared with the standard ones is welcome.
Some possible alternatives:
J309
J310
2sk117
2N5459
Purpose of this thread is to make a list with suitable replacement JFET's for use in the B1 buffer. Any information on the performance of the replacement devices compared with the standard ones is welcome.
Some possible alternatives:
J309
J310
2sk117
2N5459
BF861B suits perfectly (transconductance & Idss):
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=568-1967-1-ND
Datasheet:
http://pdf1.alldatasheet.com/datasheet-pdf/view/16240/PHILIPS/BF861.html
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=568-1967-1-ND
Datasheet:
http://pdf1.alldatasheet.com/datasheet-pdf/view/16240/PHILIPS/BF861.html
these two links have some other recommendations used in similar circuits:
this tech note talks in detail about the buffer circuit and proposes some different jfets.
the schematic here on page 22 features a similar circuit and uses yet another model jfet. (last time i bought some of these, they were pretty cheap).
it's a simple enough circuit that almost anything could be used, even the lowly 2N5457. I've seen the 2N5486 used over the years in various op amp test circuits (many years ago, you used to find useful things in data sheets besides data - imagine that
). you'll need to play with the biasing because fet characteristics are all over the map.
have fun ...
mlloyd1
this tech note talks in detail about the buffer circuit and proposes some different jfets.
the schematic here on page 22 features a similar circuit and uses yet another model jfet. (last time i bought some of these, they were pretty cheap).
it's a simple enough circuit that almost anything could be used, even the lowly 2N5457. I've seen the 2N5486 used over the years in various op amp test circuits (many years ago, you used to find useful things in data sheets besides data - imagine that
). you'll need to play with the biasing because fet characteristics are all over the map.have fun ...
mlloyd1
2SK369 is a VERY good suggestion.
BF861 has a bit too much noise. I would suggest using BF862 instead. You can read about measurements of the BF862 by Scott Wurcer from the Blowtorch thread.
Of course 2SK370, 2SK117 are all good alternatives.
You can also read my reasoning here :
http://www.diyaudio.com/forums/showthread.php?s=&threadid=140488
I personally would not waste time with the rest. They either have too much noise or too low Yfs. And 2SK170 is still sooooo easy to get....
Just my 2 cents.
Patrick
BF861 has a bit too much noise. I would suggest using BF862 instead. You can read about measurements of the BF862 by Scott Wurcer from the Blowtorch thread.
Of course 2SK370, 2SK117 are all good alternatives.
You can also read my reasoning here :
http://www.diyaudio.com/forums/showthread.php?s=&threadid=140488
I personally would not waste time with the rest. They either have too much noise or too low Yfs. And 2SK170 is still sooooo easy to get....
Just my 2 cents.
Patrick
EUVL said:
Patrick, if you look at at the BF861's noise figure (in attachment) you'll see that it's good even for phono stage and we are talking line-level buffer here
BF862 is a great part, but it cannot be used in B1 without some adaptations in schematic. In B1 JFETs are running at Idss and dissipation will be a problem - up to Id of 10mA will be OK
Attachments
mlloyd1 said:
That's just the way things has become.
This weekend the Mrs. and I went to a diyaudio meeting in Sweden (Gbg).
We were among the few under 50, I tell you we felt young in that company.
A bit on topic:
2SK170 is easy to find and cheap, a fine choice.
If you wanna go all over board, 2SK369 is the ticket, hard to find, and sure expensive if you manage to find such.
The rest are on par with the 2SK170 or below.
Magura
Here’s some data on JFET’s all recently acquired looking for substitutes in some applications. The data was taken with a QuanTech semiconductor noise analyzer. I took all data at 1mA drain current to compare the JFET “beta” parameter across the lot. My noise FOM is the flatband noise (100kHz) compared to the theoretical noise (from .66/gm). This should be ~1 but the fudge factor is computed from an ideal JFET (Shockley 1952), still they were pretty close.
Some comments - The J305 needed selection and still was not great but the J310 was surprisingly good. The 2SK222 is a great FET for microphone circuits since it’s noise is excellent and the C’s are substantially lower than the 2SK170. The BF862 was a surprise and folks here have already discovered it, also low C’s. I have not found any bad ones yet datasheet aside. J177 has something wrong like excessive gate series resistance.
Some comments - The J305 needed selection and still was not great but the J310 was surprisingly good. The 2SK222 is a great FET for microphone circuits since it’s noise is excellent and the C’s are substantially lower than the 2SK170. The BF862 was a surprise and folks here have already discovered it, also low C’s. I have not found any bad ones yet datasheet aside. J177 has something wrong like excessive gate series resistance.
Attachments
scott wurcer said:
I can't find the 2SK222 for sale anywhere - tried Avnet, Allied, Digi-key, Mouser, MCM, Newark, no luck. Any hints where to buy 2SK222 in volume today?
Any preference for brand of manufacture on the J310? Mouser sells Fairchild and OnSemi. I bought some of both, but wondered if anyone has compared. What brand was used for results in the table?
Also bought some BF862. Sorting these small suckers will be somewhat sobering.
Bob