Good gate control voltage level that, shall be sufficient. Without the diode the gate may crawl up open again. So it has it at least for that.
Saying that because I did some quick and simple breadboard experiments with -5V control from bench PSU, few NJFET types that I had laying around (2SKxxx, BF2xx, J1xx). Plus one 1N4148 gate diode (cathode connected to control's side).
I saw the lower Vgs (OFF) JFETs shutting down completely but when on they were only good for passing 500mV RMS clean signal before kinda soft clipping it soon after.
For more signal like 1V to 1.5V RMS high Vgs (OFF) types were necessary. They made me increase the OFF voltage to -8V even. Else, the negative going sinewave's part nosed out as the test signal was turned up high. THD with a J112 sample of 41Ω RDS(ON) was 0.045% when passing 500mV RMS 1KHz to 10kΩ load.
I saw the lower Vgs (OFF) JFETs shutting down completely but when on they were only good for passing 500mV RMS clean signal before kinda soft clipping it soon after.
For more signal like 1V to 1.5V RMS high Vgs (OFF) types were necessary. They made me increase the OFF voltage to -8V even. Else, the negative going sinewave's part nosed out as the test signal was turned up high. THD with a J112 sample of 41Ω RDS(ON) was 0.045% when passing 500mV RMS 1KHz to 10kΩ load.
There is a trick if slow switching is acceptable, something like driving the gate from a large resistance and decoupling it to the source to improve signal handling.
A nice trick. But not for auto muting I guess? Many neat solutions from those old mainstream Hi-Fi engineers. My tape deck for instance has a BRT transistor and two 2SK246 JFETs for tape/source monitor switching. The T/S control logic line goes to the "digital transistor's" base and to one JFET's gate. The other gate is on the BRT's collector. What polarity one gate gets the other one gets it flipped that way. Both drains are common to the deck's line out op-amp +in. Source pins to the Dolby chips outputs.
2SK246s seem to have a rather HIGH on-resistance, if I am interpreting the datasheet correctly. It shows drain current @ 1.2 to 14 ma with Vds@10 volts---that would seem to indicate a minimum of 714Ω.
Not quite. A JFET's drain current versus drain-source voltage graph is similar to a straight line through the origin at small drain-source voltages, then gradually levels off to a more or less constant current. According to rough first-order models, the more or less constant current part starts when the drain is more than the opposite of the threshold voltage above the gate (for the N channel case, all polarities reversed for P channel).
When the drain-source voltage is 10 V, you are probably already in the sort of constant current part, rather than the close to linear resistance part.
When the drain-source voltage is 10 V, you are probably already in the sort of constant current part, rather than the close to linear resistance part.
I don't know what they were thinking or what they were measuring but they list 2SK246 or 2SK381-D in the service manual. Maybe smoothest reaction to fast T/S monitor button activation was their priority. The K381 was a Mitsubishi SGD pinout 250Ω class JFET. The output shunt mute transistors I have changed to higher VEBO Panasonic with a small gain in THD (0.12% from 0.15% at 2VRMS or 5dB red in this deck) but those four for stereo JFETs to test and replace with as good functioning but surely better final result its a little project on its own with a lot of dismantling. Also the control DC level is 5V only. Not expecting it easy to do better. Sometime I may try it out of curiosity though. Cassette tape will have much much more THD of its own already vs the electronic loop at this level, but anyway.
It is not an on resistance because it is in the saturation region (constant current region).
On resistance of JFET is calculated by
Ron=-Vp/(2*Idss) （Vp is the pinch-off voltage）
2SK246 data sheet
On page 2, the upper right graph of Id-Vgs, the curve on the far right has a curve of Idss = 1.3mA, which is almost the lower limit. Vp reads approximately -1V.
Than this is calculated as
Ron=1/(2*1.3m)=385Ω
Also, if Idss = 5.6mA, it can be read as Vp = -2.75V from the leftmost curve.
Ron = 2.75/(2*5.6m) = 246Ω
That's what it means.
Of course other breeds can be calculated in the same way.
