[ARC5] JFET in ARC-5 TX oscillator socket

[WA5CAB at cs.com]

1626's are cheaper than JFET's.  And for some strange reason work 
surprisingly well in Command Set transmitters.  You'd almost think that the sets had 
been designed to use them.

In a message dated 7/7/2009 10:32:03 PM Central Daylight Time, 
dennis.monticelli at gmail.com writes: 
> I forgot that it was a Hartley configuration.  You're right.  Output 
> would
> probably be low at the source node.
> 
> One way around that would be to choose a high Vp (pinchoff) JFET with a 
> high
> Idss rating and still protect it with a cascode device or else bias the
> drain from a separate voltage source.   I think the appropriate choice 
> would
> be a J111 (2N4856 or PN4856 is the same) or J108 (PN5432 is the same)  The
> latter is the most beefy.  In all cases the BV is at least 25V.  But these
> JFETs have more transconductance than the 1626 so the gate-source swing 
> will
> be less than the grid-cathode swing for a given current draw.  The optimum
> would be to change the tap on the coil but this is something most of us
> would prefer not to do.
> 
> Dennis AE6C
> 
> On Sat, Jul 4, 2009 at 10:31 AM, Ian Wilson <ianmwilson73 at gmail.com> 
> wrote:
> 
> >Not sure that beefing up the voltage dropped across a composite device
> >would
> >help, since it's a Hartley oscillator. The voltage swing is determined by
> >the
> >current through the device and the magnitude of the grid bias voltage
> >developed
> >(I think). Some other SS compound might have the right characteristics,
> >though?
> >
> >73, ian K3IMW
> 

Robert Downs - Houston
wa5cab dot com (Web Store)
MVPA 9480