Grounded grid VHF front-end

My current receiving aerial system is very inefficient at 2 metres (144
Mc/s) and I have thought about making a sleeve dipole for that band.  My
VHF receiver is an Eddystone 770R, which covers the band but only in a
small portion of the whole scale.  While I am improvomg the aerial
system, I could also make a crystal-controlled down-converter, that
would allow me to use an HF communications receiver or the lower ranges
of the 770R, so that the  band 2 Mc/s wide would cover a much greater
scale length.

It's been a few years since I designed anything with valves, so I
thought I might have a go at making a down-converter using valves - but
not necessarily the expensive 'cult' ones which everyone seems to regard
as having magical powers.  The EF91 is plentiful and cheap as New Old
Stock, so that seems like a good valve to start playing about with.

The EF91 was used as an RF amplifier in the input stages of television
sets working at about 45 Mc/s, so it can't have too bad a noise figure
(although Mullard don't quote one in their data sheet).  If I
triode-strapped it and ran it in grounded grid mode, that would reduce
the noise and increase the maximum frequency it could usefully amplify.
From the data sheet, with 200v on anode and grid 2 and an anode current
of 6mA, the gm is about 6mA/V, which gives an input impedance at the
cathode of  160 ohms.  A 75-ohm feeder could be matched to this with a
Pi tank or by tapping the L or the C of an input tumed circuit.

The voltage gain may not be as high in this configuration as in grounded
cathode mode, but it allows the valve to be triode strapped for low
noise without instability problems or the dependence on neutralising
that a cascode stage would have (especially the need for correct
neutralising to obtain the best noise figure).  If I also use an EF91 as
a mixer, I might need one more stage of RF gain to get the signal up to
a level where the mixer noise is negligible - but this isn't such a bad
thing because it would allow extra tuned circuits to give better image
rejection and allow a lower output frquency if I wanted one.

Anyone with experience of doing something like this with valves?


--
~ Liz Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk