If anyone’s got some spare BF862s (not the sketchy ones from eBay - tried those, definitely not legit), I’d be happy to buy at least 10 pieces.
https://www.digikey.co.uk/en/products/detail/nxp-usa-inc/BF862-215/807482
Maybe time to look at the substitute/equivalent shown on many sites including Digikey, above.
Maybe time to look at the substitute/equivalent shown on many sites including Digikey, above.
Attachments
Thank you for that. These JFETs have even smaller input capacitance than BF862, but much lower gm, in my application it will affect CCS performance. BF862 combines low capacitance with high gm. However BF862 does not have super low lambda, but it's rather normal to JFETs comparing to MOSFEts. Lower lambda increases the AC output impedance of CCS.
Last edited:
I think the 2SK3557 is much closer to the BF862, with slightly lower transconductance. However, the maximum Vds is only 15V, compared to 20V for the BF862.
Thanks, the CPH3910 datasheet looks promising with higher transconductance (40mS vs 35mS for 2SK3557) and lower input capacitance (6pF vs 10pF) but the practical tests (common source gain) reported disappointing gain results for the CPH3910.
How can you get less gain from more gm, other than by presenting a MUCH
smaller load resistance?
smaller load resistance?
In vacuum tubes, two devices can have the same transconductance, but they won't necessarily be matched in gain because their plate resistance (rp) can be different.
Similarly, in FETs, the voltage gain is the transconductance (gm) multiplied by the drain resistance (assuming no external load is connected), just like in vacuum tubes.
Am I thinking about this correctly? 🙂
Similarly, in FETs, the voltage gain is the transconductance (gm) multiplied by the drain resistance (assuming no external load is connected), just like in vacuum tubes.
Am I thinking about this correctly? 🙂
Last edited: