Injection-Locked Oscillators

James Authur is our local expert on injection-locked oscillators. He seems to check out sed now and then. I have his email around here somewhere if you want to ping him directly.

Yes, please. Thanks! ...Jim Thompson

Injection is the method of injecting a periodic signal that causes a free running signal to align at that moment and thus the injected frequency and vco if you have it, will run together in time. example.

A flip flop circuit that is operating at high frequency can receive a periodic pulse on one of its inputs forcing the flip flop to restart itself and thus following output of the FF is aligned, for a while anyway..

Put it in Phils H. terms, the injection forces the VCO to wiggle one way or the other! :)

Jamie

Den torsdag den 29. januar 2015 kl. 00.36.22 UTC+1 skrev Maynard A. Philbrook Jr.:

-Lasse

See FloppyDataExtractor.pdf on the S.E.D/Schematics page of my website. I've done this with very long length shift registers to get fine resolution.

In the present application need, if the incoming signal was reasonably well-behaved I'd do that. But it's not :-( ...Jim Thompson

Dan Wolaver's "Phase-Locked Loop Circuit Design" discusses injection effects in oscillators. He does it primarily so he can discuss noise effects, but you could use the information to understand injection locking.

The basic notion is that if you inject a signal at some node in the circuit, you'll change the effective frequency/phase characteristic of the resonator (or feedback, if it's a multivibrator oscillator). Do the analysis correctly and you can then predict the amount of frequency pulling as a function of phase offset for any given amplitude of injected signal.

I suspect that any good PLL book would go into this sort of thing. Wolaver was an excellent teacher (I took PLLs from him), but I don't think he taught anything hugely unique.

'Wiggle' is a perfectly good technical term among people of good family. ;)

Google 'wiggler magnet' and you'll have the opportunity to learn about free-electron lasers.

Besides, being good at wiggling things is a key skill, for hams as well as physicists.

Cheers

Phil Hobbs

Pretty sure the term was used heavily during the very first electric motor lab experiments. :-)

(or was it the first TENS experiments done on sailors with girls tattooed on their inside arm)

Wiggle wiggle!

Perhaps not the tutorial that you are looking for, but I had this on my list of links on the topic

Injection locking is a nonlinear phenomenon. Topics which are inherently related to this are stability, bifurcations... Some years ago we investigated injection locking on a switched oscillator, with the objective of replicating a "synchronous oscillator", a nice circuit proposed by a guy named Uzunoglu.

Hope this helps!

Pere

There are a lot articles about injection locking (microwave oven) magnetrons. The first search result for "injection locking" and "magnetron" points to an IEEE article and at least according to the abstract, contains also equations for oscillators in general, not just magnetrons.

Thanks! Nice one! ...Jim Thompson

A study of locking phenomena in oscillators Adler,R.

Do you have a soft copy?

Thanks

Phil Hobbs

I keep seeing Adler's paper referenced, but haven't run across it.

It's on ieeexplore.ieee.org

I'm not a member :-( ...Jim Thompson

I'll dig it out and send you a copy.

Another form of injection locking....

Fun thing discovered this morning. We make a 4-channel ARB that uses THS3201s in the output stages. All four amps are oscillating at 1.408 GHz. That's exactly 11x the 128 MHz DAC clock; looks like a cooperative, injection-locked oscillation.

We'll replace the amps with THS3001s. That will cost us a fraction of a dB at 30 MHz sinewave out.

Moral: don't use more opamp than you have to.

or find a way to 'run' at an ideal 'locking frequency', and use the whole thing to your advantage.

Make a better camera flash storage cap charger with it. :-)

You guys only have luxury problems. ;)

Cheers

Phil Hobbs