Hello Hugh, Jim, Alan et al

Hugh M0WYE wrote:

I wonder how you demodulate CW signals - do you allow the circuit to
oscillate as you would do with a conventional regenerative circuit ? The
basic Spontaflex seems to be an AM receiver.

Yes, works just as a "conventional" regen, receiving CW whilst gently
oscillating. It's a lot more sensitive in this mode than it is when
demodulating an AM signal (set just short of oscillation) - much more so
than I would expect, in fact. Have yet to work out why this should be -
there's certainly a number of quirks in that circuit!
The only real audio filtering I have at present consists of a 22nF capacitor
across my pair of SG Brown "Type F" 4kohm phones, the combination resonating
at about 750Hz. This has reasonably decent Q when used as the collector load
of the audio output transistor, which helps in more ways than one.

Jim M0BMU wrote:

The small ferrite cored inductors that are fairly readily available can be
used to make passive bandpass filters of several hundred Hz BW quite
easily. If you don't mind winding some pot cores, and have some test
equipment (audio generator & voltmeter), passive filters with bandwidth
down to 50Hz are not too difficult. With limited dynamic range, you might
be better to go for a passive preselector, which can also be made with
several 100Hz bandwidth, if you can find suitable pot cores.

Alan G3NYK wrote:

As to the effects of DCF39 ....(and it may be being useful as a BFO in

that

circuit !!) it is possible at the LF frequencies to build a bridged-T

notch

filter that will remove DCF39 with little loss in the 136 band. You will
need a high Q inductor  as its is only 1030Hz above the band , so a Q of
about 130 will still leave you with 3dB loss at the top of the band. It
should be possible to make a circuit with a Q or 500 to 600, which should
make the in-band loss almost negligible. There is even an example in the

AC

analysis ZIP download (I think) using a 50uH inductor. The only difference
might be it is easier to centre-tap the coil that use a capacitor "tap"
(which was easier in the program)

Hmm, interesting (particularly the notch idea). It's obviously better to put
as much filtering as possible right up front, so I'll look at combining a
narrow RF bandpass filter with one or maybe even two bridged-T notches to
reduce the effects of the nearby commercials. It'll increase the overall
size of the receiver several-fold, but then you can't have everything! I've
got gain to spare in the receiver itself, so hopefully a little loss with
moderate coil Q should not be too much of an issue (in fact it might even
help). Will report back in due course.

73, Kevin G1HDQ