Replacement For Toshiba 2SK170/2SJ74

At the moment, apart from Linear Systems offerings in the DIYaudio store,
this is the most reliable offering I know, a known manufacturer of
our acquaintance who bought tons of these new before they were
discontinued and who does a fussy job]

I was keeping that as my secret stash.
Well I did share it earlier a couple of times but no one ever believes what I say so it was even better than a secret, people assumed they must be fake.
Now you've ruined it.
Hahaha
 
Disabled Account
Joined 2004
I'm using exactly these in my Babelfish J

Thanks, I was not aware of this project.

After spending some time reading there, it seems it should be possible to buy a lot of J270 (or J271) and J308 (or J309) and with careful matching come up with working pairs, e.g. for a F5.

Of course, depending on how many usable pairs we can find in lot, it might not be cheaper than just buying the real thing...
 
Official Court Jester
Joined 2003
Paid Member
they're good enough in LTP , CCS biased , in power amp input stage

for follower ...... naah , source resistance (read - DC offset thingie) too high in value , so no go ....... Rout goes south

if you take output cap in game , connected ditto to source , then OK
 
Last edited:
I've tested the 2SJ498 in an MC cartridge headamp circuit, and the results suggest that somewhere between 7-10 2SJ498 in parallel will give similar results as a single 2SJ74V.

Regarding noise, most 2SJ498 appear comparable to the Toshiba 2SJ105 (which presumably has the same die as the 2SJ103, but with less testing, and in a smaller package).
Again like the 2SJ105, there will be considerably noisier outliers. The outlier rate seems to be around one for every 30 good 2SJ498s.

But even the outliers should be perfectly usable in applications where no voltage gain is required.
 
Jim,

The less known InterFET does have a very interesting P channel jfet, the IFP44, advertised as replacement for low noise NEC 2SJ44, Yfs 9mS, about half of J74 and no direct complementary. The Toshiba TO92 2SK117 and SOT346 2SK209 could be close enough to complement it.

There is also a low noise N channel IFN147 advertised as replacement for Toshiba 2SK147 with Yfs 30mS.

Never seen anybody selling (or even mention using) them though.
 

Attachments

  • DS_2SK113_P44.pdf
    86.5 KB · Views: 276
  • DS_IFN147.pdf
    89.1 KB · Views: 233
Disabled Account
Joined 2004
Scott Wurcer once mentioned that Fairchild, now On Semi, p channel jfets have somewhat higher noise, no idea on the Interfet's or others, ZM may have more details. However, the noise issue is of much less significance when used as a follower.

I found 0 yield on Fairchild P-FET's (J177) to noise. Remember to compare gfs at the same Id, those TO-18 FET's are probaly not cheap.
 
Last edited: