Polarizing radar waves with egg slicers

[jimgraber]

If I bought a radar gun, a radar detector, and three egg slicers,
should I be able to demonstrate the three polarizer crossed polarization tricks?
TIA,
Jim Graber

 

[DaveC426913]

Is this the one where you put two grates perpendicular to each other, so there's no signal getting through, the put a third in between at a 45 degree angle?

That's so cool.

I did this with three polarizing filters when I worked in a camera shop.

My boss did not see how this would sell more cameras.

 

[sophiecentaur]

Yes. There is an excellent School demo, using 3cm (I think) microwave transmitting and receiving horns and a couple of gratings (a bit less than 1cm spacing and about 20cm square.
I am sure you could do it with your kit.
Just make sure the grid spacing is around 1/4 wavelength or less.

Camera shop managers are no fun!

 

[jimgraber]

Yes. There is an excellent School demo, using 3cm (I think) microwave transmitting and receiving horns and a couple of gratings (a bit less than 1cm spacing and about 20cm square.
I am sure you could do it with your kit.
Just make sure the grid spacing is around 1/4 wavelength or less.

Camera shop managers are no fun!

Is there somewhere I can read about, or possibly even buy this School demo?

 

[sophiecentaur]

I think the equipment (the whole caboodle) is available from Philip Harris (a UK firm). I guess there are other sources of the equivalent kit in other countries. I think Unilab make it.

 

[jimgraber]

I think the equipment (the whole caboodle) is available from Philip Harris (a UK firm). I guess there are other sources of the equivalent kit in other countries. I think Unilab make it.

Good, I found it. Thanks.
a bit pricey though.

Now I have another question:
What is the best/easiest way to produce and demonstrate circular polarization with a similar setup?
Thanks again.
Jim Graber

 

[jambaugh]

We did this experiment in Undergraduate Lab (when I was an RA) and students had a hard time with the seemingly counter-intuitive result, i.e. the direction of polarization must be perpendicular to the conducting strips for the waves to pass through.

It finally helped the students when I suggest they think of the conductor as acting like a mirror to the microwaves but only when they can conduct in the same direction as the E component of the wave. This seemed to help some integrate the results into their intuition. (FWIW)

Do remember that the polarization is in fact perpendicular to the wires.

If the egg slicers don't work because the spacing is too wide wrap some heavy paper around a section of PVC pipe (or other handy cylinder) wrap enameled wire around at the desired spacing, and paint with a couple of coats of varnish or epoxy. When it is dry enough to handle but still flexible cut along the length and you have a nice wire comb. (or three at once if you use long enough cylinder and cut into sections when you're done.) Glue onto a piece of cardboard or foam core board for stiffness and you are set to go.

 

[sophiecentaur]

"Good, I found it. Thanks.
a bit pricey though."
Yes, of course. We're talking 'shooting fish in a barrel'. Schools and colleges don't usually have the facilities to make this stuff and the suppliers can charge what they want. otoh, how much do you think the 'professional' version would cost?

 

[jambaugh]

Good, I found it. Thanks.
a bit pricey though.

Now I have another question:
What is the best/easiest way to produce and demonstrate circular polarization with a similar setup?
Thanks again.
Jim Graber

Hmmmm... You need a http://en.wikipedia.org/wiki/Wave_plate" which is a specific thickness of a material which is birefringent in the wavelengths you are using. Google-ing around I see teflon and liquid crystals mentioned as well as an interesting paper about using metal meshes with sub-wavelength scale features but that sound a bit exotic.

The basic idea of a circular polarizer is to start with a linear polarizer at a 45deg angle to the principle axes of the birefringent material. The linear polarized wave splits into two equal components one of which traveling at different speed accumulates a phase-shift relative to the other. If the thickness is sufficient for the phase shift to be a quarter phase then the output wave will be circularly polarized. (Different distances give you different degrees of elliptical polarizations.)

There are two directions 45deg from the principle axes of the birefringent material and orienting the linear polarizer along each of these will yield either left or right (cw or ccw) circular polarization respectively.

I would think you could use any plastic with an appreciable index of refraction in the microwave spectrum create a stack (with gaps) in the same way as one creates the wire comb (but thicker). The stacking will mean the dielectric value perpendicular to the sheets would be closer to air than that parallel to the sheets. The stacked sheets would then form a birefringent block.

You could also try making a resonant block to measure the wavelength of your microwave source inside the material.

Plastics are stretched during processing and so are naturally birefringent to some degree. This means using them as I have described as stacks you'll want to be consistent with the orientation, or you may want to bake them carefully to remove orientation dependent stresses, or you may want to use a setting plastic such as epoxy or lucite which will not have processing stresses. You may also be lucky enough to find a standard plastic source with the desired birefringence. Google around and see what you can find.

 

[jimgraber]

@jambaugh
Tomorrow I will look for that paper about subwavelength features. The whole idea of making the process depend on visible structures as opposed to microscopic chemicals appeals to me, even if it's not practical. Thanks for the suggestions.
Jim Graber

 

[jambaugh]

One final bit on microwave birefringence. It seems elliptical and rectangular wave-guides of suitable dimension are birefringent. It might be easiest to create a stack of rectangular waveguides of suitable length. You should find standard formulas for mode propagation speeds to do the length calculation for quarter wave phase difference at the frequency you're using.

 

[sophiecentaur]

You can produce circular polarisation of microwaves / radio waves using a whole range of structures. Crossed dipoles, separated by 1/4 wave and fed in phase should do it (in one direction, at least) and, at a few GHz, that isn't hard to make. You can do all these things because you have a coherent signal, so it is easier than for light.
There are various arrangements with circular waveguide which will give circular polarisation. And , of course, there is the axial mode helix (just thought of that one). I suggest you google and find a link that makes sense to you. There seem to be plenty.

If you are prepared to go to lower frequencies, sources and detectors are easier / cheaper to produce and so is the hardware (plus you can 'see' what's happening). Most of my life has involved me with RF and I tend to think in that direction rather than going for the 'light' model but, of course, both are valid.

 

[jimgraber]

You can produce circular polarisation of microwaves / radio waves using a whole range of structures. Crossed dipoles, separated by 1/4 wave and fed in phase should do it (in one direction, at least) and, at a few GHz, that isn't hard to make. You can do all these things because you have a coherent signal, so it is easier than for light.
There are various arrangements with circular waveguide which will give circular polarisation. And , of course, there is the axial mode helix (just thought of that one). I suggest you google and find a link that makes sense to you. There seem to be plenty.

If you are prepared to go to lower frequencies, sources and detectors are easier / cheaper to produce and so is the hardware (plus you can 'see' what's happening). Most of my life has involved me with RF and I tend to think in that direction rather than going for the 'light' model but, of course, both are valid.

sophiecentaur,
Very helpful, thanks.
I will follow up these suggestions as soon as I can find time.
I think I am looking for something like an axial mode helix filter which converts unpolarized radiation to polarized radiation.
I also like your radio idea, but CB radios work at a wavelength around forty feet.
Is there anything legal and practical around four feet? Microwave ovens transmit at around four inches, but what kind of receiver? I guess you could modify something else if necessary, but I am trying to keep it very simple, cheap and safe.
Thanks again for the ideas.
Jim Graber

 

[jimgraber]

Apparently there are Ham radio or amateur TV bands at the right wavelengths, including 125, 70, 33, 23, and 13 cm. I'll have to think about it some more.
Jim Graber

 

[sophiecentaur]

I was thinking more of UHF frequencies. There are some 'almost' legal frequencies which you could use and some lowish cost equipment for transmitting and receiving. I'd bet you could get hold of some pretty low cost, low power, UHF transmitting equipment and TV receiver modules are cheap enough.

The antennas, reflectors and grids would be fairly easy to make, with wavelengths of a few tens of cms.
I remember being shown some great demos of standing waves on model mast radiators using a Ham transmitter fed to wires fitted with 'pea' bulbs. That would have been in 1967!

Woops - just read the last post. Too slow in posting.

 

[jimgraber]

Simple circular polarizers Re: Polarizing radar waves with egg slicers

Back to the idea of a visible circular polarizer:
How simple can it be?
What would happen if I took a ribbon of aluminum foil,
Gave it a uniform axial twist and put it in a waveguide?

Or even just took a rectangular wave guide with a very narrow cross section and gave it an axial twist with a period of two pi?

I halfway think this might work, and I half way doubt it.
I wonder if I can set up a finite element simulation??
TIA
Jim Graber

 

[sophiecentaur]

I seem to remember a circular waveguide section (after a transition) with an arrangement of pins inserted part way worked well.
I found this link which has further links (the Ham way into this could well be the cheapest)
http://lea.hamradio.si/~s57uuu/emeconf/eme06.htm

Why not get onto a Ham Forum? Your requirement may be a bit specialist if you want a cheap / working solution.