helical resonators?

What frequency? Helical resonators only make sense between 50 to about 2GHz. Large wire diameters are needed. Since one end is unconnected, it has to have some mechanical rigity. See various examples at: They mostly look custom made (or home made) but there are a few:

I think Toko went to Temwell:

Depends on your loss, bandwidth, group delay, matching, cost, etc requirements. Helicals require manual tuning, which tend to make them less than optimum for robotic assembly and production:

Temwell in Taiwan.

Steve

NeoSid is still in business as well.

Steve

I made a tunable one for the FM band ages ago. As a guess,about 8" tall and 5" diameter; used 1/2 inch copper tubing for the helix which i wound by putting sand inside the tube. Copper cylinder shield, solid copper bottom and adjustable top "plunger" with that "J" spring stuff for contact at the periphery of the moving disk. Been so long i do not remember the name but that stuff was used for doors in shield rooms. Construction dimensions derived from an old (then) IEE/IRE article.

Thanks, now I feel old again. I used to build them by hand from fat Cu wire or small dia tubing, inspired by an article in the ARRL Handbook.

Mark L. Fergerson

Nice!

What was the application? FM "DX-ing?" with a strong nearby transmitter making life difficult?

"Finger stock". Typically made of beryllium copper, I believe.

The best stuff is silver- or gold-plated - this is what's used for repeater duplexer cavities.

You can still get them but they won't likely fit the price range you had in mind:

Pretty much. They went the way of the buggy whip.

Oh well, I guess I'll use a coaxial ceramic doobie. The impedances are really low, but I guess I'll make it work.

I'm making an oscillator that has to start and stop fast. A SAW would have phenominal accuracy and TC, but I don't think they would start/stop fast.

Am 17.12.2014 um 22:03 schrieb John Larkin:

Look into the circuits of the HP5370A/B time interval counter. They are in the net, maybe bama or snipped-for-privacy@febo.com

regards, Gerhard

Not sure how fast "fast" is but starting could be done using a gooser, or impetus, where the resonant circuit receives a salvo of pulses from a starter circuit which then recuses itself just like a car starter after you let go of the key. Fast stop should not be an issue.

John Larkin:

Multivibrator.

I have a couple of 5370Bs that still work. Got manuals, too.

They used an ECL delay-line oscillator. The coaxial ceramic resonators are really just delay lines, but really small and very low TCs. The reasonably-sized ones are 500 MHz and up, which I guess I can work with, divide it down maybe. A tunable helical at maybe 200 MHz would have been fun.

you havn't given us much to go on how accurate does the freq need to be? does the osc have to start with a particular phase relationship? if not, why not just leave it run and gate the output on/off? can't start/stop faster than that.

That's a fact.

It will be phase-locked to a 10 MHz TCXO, after it starts. It needs to be pretty close, under 10 PPM, for a while until the phase lock kicks in. There will be a varicap to fine-tune it.

HP did that in the 5370A and the 5359A, namely started an oscillator at some arbitrary trigger time, and then phase-locked it to a crystal osc reference. They used a delay-line oscillator trimmed by a varicap, which shouldn't work but does. They actually ran the delay-line oscillator continuously, to servo out the horrible delay-line temperature drift. When the trigger arrived, they stopped the oscillator (for 75 ns? I forget) and then restarted it. I just want to start it.

It has to start ASAP, preferably instantly, when triggered.

I plan to buy some SAW resonators and play with them, but I'm not optimistic. Anything that involves physical motion will tend to be slow to start and stop. The interdigitated transducers in a SAW must have many cycles of delay before they can launch or absorb a wave.

HP made one instrument that used a triggered quartz crystal oscillator. It was pretty awful.

I have a $2000 electro-acoustic delay line around here somewhere, 6 us or some such. It works: drive the input hard and look for a small blip at the other end. But it rings like a church full of bells, takes milliseconds to recover.

Digikey still has a bunch of Murata helical filters, but they're all listed as non-stock, i.e. DK won't be buying any more of them.

They also have a lot of real genuine time delay.

If you can't predict the arrival of the trigger pulse, one approach would be ping-ponging two oscillators, so that one is calibrating itself while the other is ready for triggering.

Well, you could use a spring reverb for that. ;)

Cheers

Phil Hobbs

The coaxial resonators are pretty good, tempcos like +-4 PPM. LCs are typically closer to 100, and playing around with NTC capacitors is a real nuisance.

Some of the SAW resonators specify astounding numbers like 0.04 PPM/K.

How about a delay-line oscillator using a spool of optical fiber? Fibers have pretty good tempcos, 15 ppm or some such.

Now I just need an optical equivalent of a varicap.

rata

u had

re

m a

er

Easy - free-space delay. Two moveable fibre ends, each at one at the focal points of a lens. Put the lens+fibre assemblies on a "travelling microscope set-up, and vary the spacing between the lenses. It's a small but finely a djustable delay. The speed of light in free space is about one nanosecond p er foot, so about one nanosecond of delay should be attainable.

If you've got a decent-sized spool of optical fibre you can adjust it's opt ical length just by heating and cooling it. Refractive index increases with temperature - at of the order of 10ppm/K. It's physical length will increa se too.

Plastic fiber and a motor-driven stretcher :-)

Maybe a piezo-driven mirror could work.