Custom Microwave Resonant Cavity Problems / Electromagnetic Radiation / Forums

Forums

4hv.org :: Forums :: Electromagnetic Radiation

« Previous topic | Next topic »

Custom Microwave Resonant Cavity Problems

Move Thread

LAN_403

Spurious J

Wed Sept 16 2009, 01:24AM

Registered Member #2367 Joined: Wed Sept 16 2009, 01:07AM
Location:
Posts: 4

I've designed and built a resonator for a 2.45 GHz wave.
The cavity is one wavelength long and one half wavelength in diameter. It's constructed from the mesh removed from the door of a microwave, shaped into a tube of the above dimensions, grounded, and has adjustable plungers on the ends to allow for fine tuning.
The magnetron penetrates the cavity radially at a distance of 1/4 wavelength from one end.

Obviously, the purpose of the cavity is to generate a single standing wave at 2.45 GHz (I'll be installing stub tuners to remove harmonics once the basics work).
The problem is that the microwaves are escaping (essentially unimpeded) from the cavity.
Why/how is this happening?
How then do I fix it?
And is there a book anywhere on cavity resonator fundamentals?

Thanks.

WaveRider

Wed Sept 16 2009, 12:43PM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

I think we need to know a bit more about how you have put everything together.

Also, do you have a load (lossy material) in the cavity? Operating a magnetron unloaded will damage it...but since power seems to be escaping by radiation, that is not a problem... Are all joints tightly soldered? Are there any..and I mean _any_...gaps or slots in your assembly?

A picture and description, perhaps? This would help with fixing your problem...

Proud Mary

Sun Sept 20 2009, 06:02PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

The two standard rectangular waveguide sizes covering the S-band are:

WG8 Internal dimensions 109.22mm * 54.61 for 1.70 - 2.60GHz

WG9A Internal dimensions 86.36mm * 43.18mm for 2.10 - 3GHz.

f(cutoff) and f(max) lie a few hundred MHz on either side of these figures

The precision and oddity of the dimensions is an artefact of international agreements about
flange sizes and so on.

So in the world of reasonable men, you'll see that internal dimensions of 90mm * 45mm will maintain
the 2:1 aspect ratio without the need for busting a gut with a micrometer, and do you very nicely for 2.45GHz.

Double-sided epoxy laminate or PTFE PCB board is a good place to start, using a combination of soldering and copper tape strapping
to maintain the integrity of the corners.

If you terminate the waveguide at one end, (i.e. solder a rectangular plate over it) you can put your magnetron probe into the guide through a small variable slot (i.e. so you can move the precise position of the probe backwards and forwards two or three cms from the theoretical quarter wave point. Once the precise position has been empirically determined, you can just cover over the exposed lengths of the slot with adhesive copper tape, so you can peel it off later if further tweaks prove necessary.

Now you are in business, because once the microwaves are safely contained within the guide, you can make them go wherever you like. The experiment could be many metres away from the magnetron and its supporting controls and supplies if need be.

One last thing thing, standing waves and magnetrons don't get along too well together and the situation should be avoided at all costs. Power reflected back into the magnetron will probably destroy it and/or force it into unstable mode, so you won't have a clue what is going on unless you've got a pile of costly kit.

There's no theoretical reason why you shouldn't use a circular waveguide made of copper pipe or heavy aluminium tube (as second best) of suitable internal dimensions but from a practical point of view it's harder to clamp down, much harder to add junctions to, and lacks the advantage of the rectangular waveguide from a constructional point of, since it's much easier to bolt circuit elements onto a rectangle than to a cylinder, and generally easier to cut large rectangular holes than round ones in metal.

Once you've got your magnetron accurately installed in a waveguide, you can conduct them to your plasma chamber or whatever, make variable attenuators to control power throughput (since you cannot vary the RMS power of the magnetron) and all sorts of interesting things.

Wonderful things! as Howard Carter put it when he first looked through a hole he'd illegally hacked through a plaster wall in the tomb of Tutankhamounn.

So many possible experiments, so little time!

johnf

Fri Sept 25 2009, 08:22AM

Registered Member #230 Joined: Tue Feb 21 2006, 08:01PM
Location: Gracefield lower Hutt
Posts: 284

I've played with this myself

checkout

read all messages in the thread to see what I did

Spurious J

Thu Oct 22 2009, 04:23AM

Registered Member #2367 Joined: Wed Sept 16 2009, 01:07AM
Location:
Posts: 4

Thanks all. I've begun work on another version, which will be rectangular and constructed to much finer tolerances. Is there a good book describing resonator theory or modes of standing waves, or anything of the sort?
I've thus far found essentially nothing approaching a comprehensive guide to the design/function of cavity resonators.

A question about plunger design, are there methods of preventing/minimizing arcing between the plunger and the waveguide other than just making the edges touch?

I'm not particularly concerned with destroying a magnetron or two in early experimentation, I have a surplus anyway, but I want to avoid generating enigmatic modes in the cavity. Would inputing the RF through a coaxial cable be sufficient to avoid this? Or what would be best?

I'll get some pictures of the new device up as it's constructed.

John, why is your construction mounted within a microwave oven? Safety? Ease of construction? Durability?

Steve Conner

Thu Oct 22 2009, 08:48AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Spurious J wrote ...

A question about plunger design, are there methods of preventing/minimizing arcing between the plunger and the waveguide other than just making the edges touch?

As far as I know you use finger stock to ensure a good electrical contact.

Thu Oct 22 2009, 01:26PM

Registered Member #2372 Joined:
Location:
Posts: 62

There is probably some stuff in this book

Proud Mary

Sat Oct 24 2009, 04:45PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

You may find this page helpful:

WaveRider

Sat Oct 24 2009, 08:58PM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

I would recommend using a plunger with a 1/4 wave choke. Finger stock becomes unreliable at high power.

I would also recommend getting a good basic book that discusses microwave cavities (Ramo, Whinnery and Van Duzer, _Fields_and_Waves_in_Communication_Electronics_ is a good undergraduate text that can get you started.) The resonant frequency of the cavity will be heavily dependent on the modes that are excited....and this depends on the positioning of the probe in the cavity.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob