Re: Valve frequency multipliers (followup)

Liz Tuddenham <liz@poppyrecords.invalid.invalid> wrote:

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
 

On 2025-04-07 13:22, Liz Tuddenham wrote:

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
 

On 2025-04-06 12:35, john larkin wrote:

On Sat, 5 Apr 2025 18:33:39 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
 

You may remember a couple of month ago I was trying to derive 150 Mc/s
from a 16.667 Mc/s crystal with two triode triplers.  I now think I have
found the cause of my problems:
 
The first tripler circuit didn't seem to tune up correctly and all sorts
of spurious harmonics were coming out of it.  There was never enough 50
Mc/s signal to drive the second tripler far enough into non-linearity,
so the 150 Mc/s output was utterly feeble.
 
Eventually I decided I was never going to get it to work in the space
available, which was only just big enough for one valve, so it would
have to be split, with the first tripler in the oscillator box and the
second tripler in another box.  I decided to use a pentode (EF91) for
the first tripler as it could be biassed to give a lot of distortion and
a large anode voltage swing.  Because there was now room available and a
trimming capacitor to spare (which had previously been used to tune the
150 Mc/s coil), I abandoned the ferrite slug-tuned 50 Mc/s coil and
wound an air-cored one instead.
 
The circuit gives a *huge* output, far more than ever before (and it is
not due to self-oscillation or any other vice like that).  I think the
cause of the previous low output must have been the ferrite tuning slug,
which probably wasn't rated for 50 Mc/s and was damping the circuit or
saturating to give lots of unwanted harmonics.
 
There's plenty of work still to do, but at least one link in the chain
is now working and I have an explanation of the probable reason why it
didn't work before.

 
Air core inductors, simple coils, would have best Q at your
frequencies.
 
 

Decent coax is pretty good too--to make an inductor at 150 MHz takes
only a few inches of coax with a short at the other end.
 
I like to tune coaxial stubs using thumbtacks--you stick it through the
shield and the center conductor.  It survives very well, so you can do

 
That's an excellent idea, I hadn't thought of doing it that way.  It
might be a bit tricky 'tapping' such an inductor for various loads.
 

 
You tap it Colpitts-style when necessary.  That isn't as good as tapping
a tightly-coupled coil such as a gapped toroid, because the impedance
change comes only from the tank Q, rather than tank Q plus transformer
action.

 
If you use an 'open' transmission line, such as a trough line, then you
do get a transformer effect.   Most UHF television tuners used trough
lines with tappings to match the aerial impedance to the transistors and
still retain a reasonable 'Q' for tuning.
 
 Leak even did it at VHF in their FM tuner but I don't know whether it
gave any advantages (I believe some of their designs were actually made
under licence from the BBC and weren't designed by H.J. Leak at all.)
 
 

You do get different impedances at different points, but that’s just on
account of the circulating power, as in a Colpitts.  It falls apart
completely at low Q.

With tight inductive coupling, as in an ideal Hartley, the coupling ratio
is nearly independent of Q.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs  Principal Consultant  ElectroOptical Innovations LLC /
Hobbs ElectroOptics  Optics, Electro-optics, Photonics, Analog Electronics