diyAudio (
-   Pass Labs (
-   -   Thoughts for replacing the scarce 2SK389 in the pearl (

h_a 12th March 2010 10:17 AM

Thoughts for replacing the scarce 2SK389 in the pearl
So recently the 2SK389 dual jfet blew in one channel of my pearl. Reading the forum, it seems this happens now and then. Looking at the schematic it becomes clear that with some 13.5-14V VDS a highish drain current (IDSS) BL jfet may run a little hot (above 100mW) as degeneration is rather low to maintain high gain.

So what to do, especially as the matching and thermal coupling of the jfets in the scarce 2SK389 is not really needed in this position due to the fact that this stage is essentially a copy of the K170 quad?

Thinking a little bit about the circuit, it becomes clear that the K389 has to deal with rather large signals, in the some 100mV range - and that needs rather large pinch off voltage Vp to avoid clipping. Pinch off voltage is the gate source voltage at which the jfet completely cuts off and is way away from its linear region.

It's often said the K389 can be perfectly replaced by 2x K170, but that's not exactly true. Looking at the datasheet, the Vp of the 2SK389 is about -0.8 V for IDSS=9.2mA. The 2SK170 offers about -0.5 V - so almost only half of that - for comparable IDSS. The 2SK369 comes with a little more than -0.3V for that IDSS.

Doing a little simulation of this common-source stage shows nicely that Vp has a big impact on distortion, it is twice as high for a part with -0.2 V in comparison to a -0.4 V part.

So simply replacing a K389 in this position with 2x K170 will increase distortion - so what todo?

My suggestion to tackle that is
  • use the highest IDSS parts that you can find (Vp increases with IDSS!)
  • or/and use the even higher transconductance 2x 2SK369 with increased degeneration (20-40 Ohms) as that will decrease distortion by local feedback

These two points should bring distortion back to about original levels, though headroom is reduced by some 4dB or so. If time permits I try to measure distortion figures for comparison.

Have fun, Hannes

PS: not a workable option is to use the large pinch off parts like 2SK246 as they have 10 times lower gain.

h_a 14th March 2010 08:58 AM

Funny, I just noticed that in the same datasheet the curves from different plots do not fit together.

I took the pinch off voltage from the ID vs VGS curves and now I see that there is below a separate plot for the pinch off voltage vs. IDSS ;) - and these two data do not fit together, likely due to production tolerance of the used sample jfet. So there is no general trend for larger Vps for 2SK389s.

From the Vp vs IDSS plots the 2SK170 is indeed a 1:1 replacement for the 2SK389 as is the 2SK369.

While my recommendation is still the same, I no longer see why the dual jfet was used at all in this position...

Have fun, Hannes

All times are GMT. The time now is 06:46 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright ©1999-2017 diyAudio