EMRFD Message Archive 3509

Message Date From Subject
3509 2009-08-31 05:57:36 timshoppa Theory/modeling of regenerative detectors
I had some fun building N1TEV's regenerative receivers, e.g. N1TEV's "high performance regenerative receiver design" over the weekend. The circuits use a "throttle capacitor" as a regeneration control, giving a smooth action superior to the Space Spanner I had as a kid. They have impressive audio clarity when receiving AM shortwave broadcasts. Spent an hour or two winding different coils for 49M and 31M last night and listening to all sorts of stuff with a two-transistor receiver!

What's interesting is that in real life, the regenerative detector I build on the bench seems to have a much higher gain than the SPICE models I play with.

In my SPICE models of J310-based regenerative detectors, I see very little shift in overall DC current draw by the detector as I twiddle the regeneration control up and down. Yeah, there's some shift, but it's very small, a few percent. I do see a fairly smooth advance in oscillator level as regeneration goes above the "start" point and things saturate.

In my real-world circuits, there is a substantial (20%, 30%) change in regenerative detector DC current as the regeneration kicks in and cranks up.

As far as I can figure out, the detection in a regenerative detector results from the change in average current through the detector/oscillator. At least that's my off-the-cuff description.

One possible hint that I don't fully understand: the real world regenerative oscillator waveform starts looking pretty gnarly, lotsa little bumps and spikes visible watching the 6MHz detector on my 100MHz scope, as I advance the regeneration control past the "start" point. The SPICE model seems to lack these waveform details, showing a more smooth progression towards saturation. Is it possible that parasitics and nonlinearities not so well modeled by SPICE are the larger part of the detection process?

Any other thoughts? I know better than to trust SPICE all the time in the face of real world data, but I like to understand how it differs from the real world.

N3QE Tim
3510 2009-08-31 06:48:33 timshoppa Re: Theory/modeling of regenerative detectors
3511 2009-08-31 09:57:33 Paul Anderson Re: Theory/modeling of regenerative detectors
On 31-Aug-09, at 9:47 AM, timshoppa wrote:

> I just wanted to add that the observed change in detector/oscillator
> current is quite smooth - and feels almost linear in terms of
> rotation of the regeneration control - despite the gnarliness of the
> RF waveform. Very interesting.
One thing I have noticed is that some designs of regenerative
detectors seem, to me, identical to the basic design of a tesla coil.
If we look at a tesla coil from an electrical perspective, we have two
LC tank circuits. The primary has a large capacitance and low
inductance, the secondary has a low capacitance(the terminal at top
forms the capacitor) but a large inductance. They are inductively
coupled, and tuned to resonance. At this point it produces the
desired sparks. If the two are not resonant, it either doesn't
function at all or does so only very poorly.

Mind, I'm far from being an expert on either tesla coils or
regenerative detectors. I may be completely out in the wilderness on
that point.

> (Sorry, if you never saw a hit-and-miss engine start up you gotta go
> to youtube, but you'll miss the physical presence of the vibrating
> and smoke-blowing gizmo that is so essential!)
I have a 4HP vertical Goold, Muir and Shapley with the square hopper.
Love those engines, there is something about the sound of one that
does good for the soul.

Paul Anderson
3512 2009-08-31 10:27:24 iq_rx Re: Theory/modeling of regenerative detectors
Hi Tim,
Thanks for your hit-or-miss engine analogy. Another one I use is to compare a phasing rig with a two cylinder engine, with the two cylinders 90 degrees apart. That prevents it from running backwards--that is,
3527 2009-09-03 10:27:09 davidpnewkirk Re: Theory/modeling of regenerative detectors (long)