Injection Locking an Oscillator (was Novel Zeitgeber devive for antique clock)

Should anyone be puzzled over my "oscillator", it's the behavioral equivalent of a diff-pair, the explanation of which I posted yesterday (2/1/2014)...

...Jim Thompson

You fail to realize that a parametric oscillator is an oscillator.

BoomBox! Hilarious hairball. Why do something with a few parts, when you can use hundreds?

Trivial and unoriginal. It's more interesting to use the parametric pump as the gain element.

Of course, PAL, it works so-o-o-o well >:-} ...Jim Thompson

The point is that it does work. It makes an f/2 signal using only a diode as the active element, which was (some time ago) in question here. With a proper varicap diode and some optimization, it would be more efficient. But it works. It was sort of an homage to A. F. Boff.

I used power varactors in the early '60's (tripling) to get to

Drilled a hole in the middle of the roof of my '61 Renault Dauphine and mounted a waterproof BNC connector for a quarter-wave whip ;-)

But getting back to subharmonic inject of "oscillators"... I think you're deluding yourself calling that parametric pumping.

See for example, with an ordinary diode and input DC-skewed (as in your post, but in the proper direction :)...

But we'd all be interested in seeing your ASC file when it includes "a proper varicap diode and some optimization" >:-} ...Jim Thompson

Cool; a dozen quad LM324s, 48 crappy opamps. LM324's show visible sinewave crossover distortion at 60 Hz.

One question? Who would put a circuit like that, grossly distorting, with hundreds of parts, consuming about 5 watts, into a boom box?

OK, there is clearly an f/2 component in my output. There is none in the V1 source.

Where does the f/2 energy come from? If you can't figure that out, try asking one of your "young bucks."

Are you sure your oscillator is sub-locking to the sinewave source? It's a little hard to tell in that narrow time slice, but it sure looks like they are drifting in phase. So the "lock" may be an illusion of them being, by design, close in frequency.

Listen, PAL, it's too exotic for you to understand... REALLY >:-}

How about getting back to the original subject? ...Jim Thompson

YOU posted that circuit. Now you don't want to talk about it.

Can't say I blame you. It's another parts-rich hairball.

Parametric dividers are still for sale, I think. Somebody used to advertise them as "frequency halvers".

It's a pretty well known effect--see e.g.

Cheers

Phil Hobbs

Awwwwh! Now you done gone and showed PAL how to "optimize" his "hair-ball" >:-} ...Jim Thompson

You're looking after your own pissing contest, Jim. The rest of us have more or less outgrown junior high.

Cheers

Phil Hobbs

Zero content, as usual.

Is it indeed locked? Or does it just look that way in a narrow time slice?

Before flipflops were invented!

Neat. It's fun to discover stuff, even though it has probably been discovered before.

I was thinking about how to do a Spice sim that is an analog to the pendulum pumped by the 60 Hz coil. The tricky part is coming up with a nonlinear resistor (or whatever) that approximates the force that the coil applies to the pendulum as a function of the pendulum's angular position. That curve will need to be fairly spikey to get decent pumping action.

Some other day. I've got to fix our hairdresser's 12-volt track lighting system (the fried parts of which are in the kitchen of the Chinese restaurant next door) and I think something else is going on this afternoon maybe.

Dunno. No TV here. ;)

Cheers

Phil Hobbs

You are a big boy, PAL, so run it in LTspice yourself. Copy the following text into a file named "InjectionLocking_DiodeOnly.cir" (you may need to rename D1N4148 to whatever LTspice likes as a name for that common diode).

Then, in LTspice, File, Open (the above filename), then RUN, and see for yourself.

"InjectionLocking_DiodeOnly.cir" is the following text...

• C:\Projects\Expments\SED\InjectionLocking_DiodeOnly.sch
• Schematics Version 15.7.0
• Sun Feb 02 11:16:10 2014

• Schematics Netlist *

X_L1 0 N_1 L_PARAM PARAMS: L=10mH RS=2 SRF=8Meg C_C1 N_1 0 1uF IC=1 R_R2 N_1 N_2 10K E_E1 N_2 0 VALUE { TANH(V(N_1, 0)/1m) } D_D1 N_3 N_1 D1N4148 V_V2 N_4 0

.END ...Jim Thompson

Post a proper LT Spice schematic and I'll run it.

In the meantime, is it actually locked?