Electronics > RF, Microwave, Ham Radio

Resources for Designing Cavity Resonators

<< < (3/3)

It depends on how you do it.

If the tape is very long / wide so you end up with mostly a single piece in there, that might not be so bad.

The issues are:
1: Surface roughness -- as you say with a 3um skin depth anything that isn't polished flat to within a small fraction of that creates signal disturbance and added resistance no matter if there is a thick solid copper plane just below it.

2: Seams.  Since the skin depth is small a solder joint along an edge does produce a "long" region where the signal transitions from the surface of the Cu to the surface of the soldered seam, over the top of the solder, down to the Cu layer where the other end of the joint is.  Not the end of the world, but lossy.

"pro" RF cavities are not only say brass or copper or what not inside, but they're then electroplated with a polished layer of something like ultra high conductivity silver lining on the inside then that's possibly coated with something to protect the silver from oxidation.  Just the difference from pure polished copper to pure polished silver is enough to matter for the Q in not a few cases.

But it depends on your application and quality and stability requirements.  There are enough amateur radio articles talking about making cavities out of pipe or metal cans / bins.  And at minimum some people make VHF cavities others up into
high UHF or cm wavelengths.  Lower frequency is more forgiving dimensionally and in materials.

But just get any old paint can, or maybe tin can attach a probe, hook up your VNA and you'll probably see a strong resonance at some relatively predictable dimension just like an old grid dip meter operated somewhre roughly near a lumped tuned circuit.  So you absolutely can experiment with this with simple materials. 
You just might not achieve more than a fraction of the stability / Q which is possible with more careful materials and designs.

You can also experiment with coaxial line resonators or patch / plate resonators and so forth and get relatively high Q.

Dielectric loaded cavities are another possibility where you can fill it with low loss wax or polystyrene or silica or some such thing that has low loss @ F vs its Dk and maybe thereby tune a smaller (easier to make) resonator down to resonate at some significantly lower frequency.

Reentrant designs can be useful / interesting, too, or making a cylindrical one ywith a tuning piston and maybe a quarter-wave choke flange for the tuning piston if you're so inclined.

--- Quote from: WideBandwidth on June 14, 2021, 03:41:46 am ---All,

This is great! Thank you all for the suggested readings, they will keep me busy for a while.
And @evb149, thanks for the information on coupling/tuning methods.

I suppose one follow-up question: is it feasible to prototype a cavity with a plastic shell lined with copper tape, or would I encounter high loss and other badness? I'm currently mainly interested in the UHF range where skin depths in copper are around 3um. I imagine tape thickness somewhere around 50um would make loss a nonissue, but perhaps I'm not thinking of something else.


--- End quote ---

Lining inside a plastic enclosure sounds awful. Wrapping the outside though, that has some promise.  Use EMI tape, the kind that makes contact through the adhesive side, or if nothing else, solder over the seams.

The plastic being on the inside, has a minor effect on resonant frequency.  Choose plastics with low losses, i.e. avoid PVC and PET, most hydrocarbons are okay.

Can also solder together sheets of copper clad PCB, again doing up the seams fully.  Can't exactly get a soldering iron down inside a shape, you'll have to do it from the outside which means the inside copper is kinda just dead space.  Might rather use single-sided, actually?  Maybe glue up the seams for strength, then solder foil around the corners?  Oh, or solder foil around the edges of double-sided, then solder it up the rest of the way.


You might want to have a look at the cavity filter pages on Scotty's Spectrum Analyzer site:


Scotty Sprowls came up with a design for a high Q ~1GHz coaxial cavity filter that many people have been able to reproduce. For the spectrum analyzer application, multiple filter stages were needed.

There are many articles around about "pipe cap" cavity filters for microwave ham radio use.  Most of these pipe cap filters operate as lowish Q coaxial resonators with the "tuning screw" being the resonant structure.  Higher Q could be achieved from pipe caps by operating them at resonance of the cavity, with a small screw only perturbing the resonant frequency, but I don't think that is done very often, possibly because higher modes of the cavity are closer in than are higher modes of coaxial resonators.

Here is what I would do, I'd get an idea of the rough size, throw it onnthe nanoVNA and tune it that way.   Every time I saw a big metal garbage can or something, Id make the tuning screw an make two ports, then throw it on the VNA and see if it could b e tuned to act as a high Q filter at some useful frequency, then mark on the outside what it is, fairly soon you would have made some useful resonators out of junk and could just pick one out of your junk box when you needed one. You could also use it for cooking food

W1GHZ has some web pages on making them cheaply out of plumbing fixtures.


[0] Message Index

[*] Previous page

There was an error while thanking
Go to full version