So my tested example is actually 20% above average.
AND according to my database, the typical Yfs of 2SJ74BL is around 38~40mS.
A typical 2SK170BL is about 30~33mS.
Cheers,
Patrick
John Hall gave me a small bag of each grade of LSJ74's so maybe I could get numbers for you. I mostly design for a given drain current and the gm variation is quite small across grades at say 1 or 2 mA Id. I would find a conformance to square law comparison more interesting.
Jan - I hope to bring my proto-type of the all FET mic (using LSK parts) to BA.
Last edited:
Kaiser bought WCVB-56 in Boston and exported him every Saturday night. The night he put on a Nixon mask and just stared at the camera I nearly passed out laughing.
I did grow up in Milwaukee so I understood the silver ball buster and koo-basi jokes.
In NPs original article on the F5 (1/31/08), he wrote:
"The input JFETs used are 2SK170 or 2SK370 for the N channel parts (Q1), and 2SJ74 or 2SJ108 for the P channel parts (Q2). In these cases the Idss selection code is BL, although V and GR types will also generally work fine. The primary thing about these particular parts is the transconductance figure of 20 mS - many of the potential substitutes are much lower at 4 to 10 mS."
At least with regard to the F5, it sounds like transconductance in the 20 mS range is what he was expecting or planning on, in general. Of course, this doesn't necessarily hold for all other amp designs, but 6L6's point that the Linear Systems' components should be fine in many or most instances seems to hit home. In a very simplistic view, one needs to generate enough current from the JFETs across a couple resistors to get the appropriate voltage to the gates of the MOSFETs, and it seems that with appropriate bias and resistors, this shouldn't be a problem.
Please correct me and teach me if I am off-base. Thanks.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.