Flexible RF Reflective material: what causes reflectivity?

[cprobertson1]

Good Morning,

I've been toying with the idea of making a portable, deployable parabolic antenna for UHF (70cm) band.

The design calls for a flexible material that, in combination with a lightweight frame will hold itself taught - but if loosened, the entire thing will be able to fold up like an umbrella.

I came up with several ideas for doing this: all with their pros and cons.

However, one fundamental question has plagued me: what causes reflection of the RF energy? What criteria must be met in order for it to reflect more energy that it absorbs/transmits*

*as in passing through it, rather than being re-radiated

The parameters I'm interested in are the:

• Conductivity of the material: must the entire reflector have electrical continuity - or can I use an amorphous material like metalized polyester (polyester with aluminium particles embedded in it - found in many tent fabrics to reflect heat from direct sunlight)
• Thickness of the material: I would assume that increasing the thickness of the material reduces the amount of energy passing through it - causing it to absorb/reflect more: but I also imagine you would get quickly diminishing returns with increasing thickness. Am I correct in these assumptions?
• Will any metalic surface reflect RF?: at a cursory glance, I would assume so - but this would tie in with the conductivity of the material (and a few other parameters, for instance, is the material absorbing and re-radiating the RF?)

My current ideas for a reflective surface are:

• Metalized polyester (tent canvas)
• Light (tent?) canvas with aluminium foil sewn into it (with overlapping joins between panels)
• Aluminium fabric (very lightweight mesh)
• Metalized polyester (tent canvas) with a layer of flexible metallic spraypaint over it.

Any help on the topic would be appreciated

[hamster_nz]

(mainly to follow, not really add)

I wonder if it is like optics - where the impedance mismatch is so high that no RF energy can penetrate into the material or be absorbed, hence it bounces.....

[cprobertson1]

I've got a feeling that would only apply if it were behaving like a transmission line: though I assume an RF guru could tell us a lot more about the subject.

So according to this electronics stackexchange post - a perfectly conductive surface would give a reflection.

I am not sure how this would work with my metallic paint/metalised polyester idea: which has conductivity, but only over very short ranges (the particle size of the aluminium). I would imagine it'd be the RF equivalent of a matt surface, with some reflectivity but a lot of absorption (and transmission too, depending on how thick it is). There will also probably be a lot of scattering - though that might not be a terrific problem. Heh, I wonder if there were will be the equivalent of libration fading on the surface of the antenna - I think it will just manifest as an inefficiency/loss though, rather than anything interesting.

If the antenna has to be conductive, I suppose there are also electrical interactions to consider.

You know what? I'm ordering a textbook on the properties/theory of electromagnetism. I expect it's going to be maths-heavy xD

I've also ordered a textbook on antenna theory for £12 second hand; it was the last one at that price. The others, also second hand, are £150. It's £250-270 new.

Jeez, why do textbooks cost so much!?

[TheUnnamedNewbie]

Well, one way of mentally ''simulating'' it can be done as follows: Remember that a highly conductive material supports, in theory, no E-field across it's surface. As a result, if an EM wave 'impacts' the surface of a metal, the metal must generate an ''opposite'' E-field to cancel it out. This E-field than generates a wave traveling away from the metal - Your reflection (this also explains why reflections at angle happen the way they do - the metal only cancels out the component that is along the surface (so, propagating ''into'' the metal). The component that is orthogonal (propagating along the surface) is ignored.


The reason for this has to do with the conductivity. When a E-field is applied across a conductor, we get a current. This current results in charge (electrons or holes, though almost all metals have pure electron currents) moving (in addition, the H component induces currents as well). These moving charges result in the 'opposite' field moving away, resulting in your reflection.

At some frequency, the charge is too slow (the electrons can physically not keep up). At this point we go from strong reflection to absorption. The frequency this happens at is the plasma frequency, and is in the UV wavelengths for most metals. Some semiconductors are engineered specifically to keep that plasma frequency low, to make translucent conductors.

Of course, a real metal is never perfect, and so not everything is reflected. Part of the incoming field will actually enter into the metal, but the metal is a very lossy medium, and we get a bit over 8 dB loss per skin-depth (it is actually 1 Neper/skin depth I believe).

The skindepth itself is tiny, and related to the conductivity of the material.
It is equal to:

\$ \delta_s = \sqrt{\frac{2}{\omega \mu \sigma}} \$

Where \$ \omega = 2 \pi f\$, \$\mu\$ is the magnetic permeability and equals \$4 \pi \cdot 10^{-7}\$ and \$\sigma\$ is the conductivity. At 10 GHz we are looking at skin depths less than 1 micrometer deep.


More mathematical study of reflection:

You can show that the reflection coefficient \$\Gamma = \frac{E_{ref}}{E_{inc}}\$ at an interface equals \$\Gamma= \frac{\eta_1 - \eta_0}{\eta_1 + \eta_0}\$ where \$\eta_0\$ and \$\eta_1\$ are the impedance of the materials (incoming wave traveling from material 0 to material 1). For air, \$\eta = 377\, \Omega\$. For a metal, \$\eta = \frac{1-j}{\sigma\delta_s}\$.

In Pozar's book, example 1.4 studies this for a copper at 1 GHz, and we find that  \$\Gamma= 0.9999382\$ and \$T = 6.181\cdot 10^-5\$ (\$T\$ is the 'transmitted' component. I'm ignoring the phase shift in both). So in dB that would mean \$S_{11}\$ equals to -0.000536 dB, and \$S_{21} = -84\$ dB. (and then after that we lose another 8 dB per 2 micrometers of copper).

For very thin materials (so when we have air - metal - air) it gets a bit more complicated because the propagated wave reflects again when we have the next interface between metal and air, and this second reflection will start interfering with the original wave. This is what gives very thin films nice rainbow colors.


I will read over this again, I hope I didn't make any mistakes in the math If you have any questions feel free to ask, I don't know how familiar you are with the physics involved so.

You know what? I'm ordering a textbook on the properties/theory of electromagnetism. I expect it's going to be maths-heavy xD

A great free starter resource is Sophocles J. Orfanidis ''Electromagnetic Waves and Antennas'', which you can download from his website: http://eceweb1.rutgers.edu/~orfanidi/ewa/. Most RF people I know tend to say Pozar's "Microwave Engineering" is the go-to reference. But, as you said, it is expensive. I must say it is also the one book that I have on my desk all the time.

[Howardlong]

How large are you expecting the reflector to be?

There's an practical engineering tradeoff between using, say, a yagi and/or phased array compared to a dish. The tradeoff is typically down to size and wind load, particularly if it needs to be re-oriented. Also the relative feed size will cause aperture blockage unless you use an offset feed.

I did a lot of work on dishes down to L band  (1.2GHz) some years ago, and the point at which you'd switch from a phased array or yagi was when the dish diameter hit about 4 * lambda, which equates to a half power beamwidth of around 23 degrees and gain of around 18dBi.

So for UHF you're looking at a minimum of 2.8m diameter to make it a reasonable proposition compared to other options.

You don't need a solid reflector, and for the purposes of wind load almost certainly you'd want to avoid it for a dish of that size. There's a rule of thumb to keep the mesh size to less than lambda/10, I'd go a little smaller if possible. The dual band 2.4GHz/1.2GHz feed dish I made some years ago used chicken wire which is about a 1cm mesh as I remember http://www.g6lvb.com/g6lvb_shack_spring_2002.htm. To make a parabola is quite easy as you can make one by simply stressing spars from a centre point, just like an umbrella. Try to keep the parabolic aberrations down to <lambda/10. At 70cm this won't be so hard.

I did make a portable umbrella style dish for 2.4GHz using a copper fabric which worked pretty well http://www.g6lvb.com/brollydish2.htm but this was only 1.2m, with a yagi co-located on the feed boom for 70cm.

Keep in mind that I found that commercial umbrellas (or similar) used as parabolic reflectors were quite shallow, so you need to be careful feeding them: they will have quite long feed booms as a result, and the feed itself will need increased gain as a result in order to efficiently illuminate the reflector. This is one place where reciprocity between transmit and receive doesn't quite work: for best receive performance it's common to under-illuminate the dish to avoid noise from overspill.

See also http://www.g6lvb.com/brollydish.htm which used fly screen mesh. I wouldn't necessarily recommend it as despite being a mesh, it is fine and presents high wind load.

Bonus: here I am with SKylab & Shuttle astronaut Owen Garriott http://www.g6lvb.com/dayton_demo_2002.htm. Coincidentally I gave his son Richard Garriott (space tourist, Soyuz & ISS) a lift from London to Warwick and back a few years ago for a talk shortly after his trip to the ISS. Never have I been so delighted to be stuck in a two hour traffic jam.

Edit: fixed a couple of links.

[TheUnnamedNewbie]

This is one place where reciprocity between transmit and receive doesn't quite work: for best receive performance it's common to under-illuminate the dish to avoid noise from overspill.

But if your feed in TX does not fully illuminate the dish, it would also not be sensitive to the entire dish, and thus you would still have a reciprocal system? Or am I missing something?

[dmills]

[quote author=TheUnnamedNewbie
But if your feed in TX does not fully illuminate the dish, it would also not be sensitive to the entire dish, and thus you would still have a reciprocal system? Or am I missing something?
[/quote]
The issue is the environmental noise added between the Tx and Rx. By under illuminating the Rx dish you improve the front/back ratio as less noise from behind the dish makes it to the feedhorn. On the transmit side having a few percent of the generated power miss the dish is often worthwhile because it gets you more power in the main beam and the larger effective aperture gives you a tighter main beam, the scatter is not (usually) important.

The antenna is reciprocal, the system is not.

I would note that at 70cm, that will be some dish to beat a yagi, but that you can make the thing out of mesh having openings no larger then about lambda/10 (so 7cm or less), it does not need to be solid as long as the mesh is electrically conductive (Be careful about joins). In fact at the sort of scale you need for a 70cm dish to be worthwhile, not solid is very much a feature not a bug as it reduces the windage and tendency to trap water.

Regards, Dan.

[TheUnnamedNewbie]

But if your feed in TX does not fully illuminate the dish, it would also not be sensitive to the entire dish, and thus you would still have a reciprocal system? Or am I missing something?

The issue is the environmental noise added between the Tx and Rx. By under illuminating the Rx dish you improve the front/back ratio as less noise from behind the dish makes it to the feedhorn. On the transmit side having a few percent of the generated power miss the dish is often worthwhile because it gets you more power in the main beam and the larger effective aperture gives you a tighter main beam, the scatter is not (usually) important.

The antenna is reciprocal, the system is not.

I see, I was thinking in terms of S-parameters being reciprocal (ie, \$S_{12} = S_{21}\$).

[vk6zgo]

70cm is quite a long wavelength compared to chicken wire (the small stuff used for baby chicks).
Surface perturbations on a dish can also be quite large, before there is any discernible effect.

[dazz1]

Hi
Unless your signal is circularly polarized, you don't need a surface.  A number of parallel rods that are formed in a parabolic dish shape will do the trick.   Cheaper, easier and much less windage that an actual dish.    This is a very old technique.  Still commonly seen used in large radar antennas
https://patents.google.com/patent/US3178713

[Howardlong]

70cm is quite a long wavelength compared to chicken wire (the small stuff used for baby chicks).
Surface perturbations on a dish can also be quite large, before there is any discernible effect.

Indeed, chain link fencing would easily work at 70cm. It's another engineering compromise, you need to take the weight into account. I'm not sure how much difference there is in wind loading between chain link fence and chicken wire. The aluminium mesh I've used in the past is light weight compared to chicken wire, but exhibits a relatively high wind load. The Flectron copper woven material I've also used is reasonably light weight, and handles far better than any metal mesh when folded, but it has a very high wind load not to mention being expensive.

[T3sl4co1l]

Yes, just like with transmission lines, reflection is about impedance mismatch.  The resistance of free space is ~377Ω; the resistivity of a metal is ~100nΩ·m, whatever that comes out to be, on the order of a wavelength and skin depth -- most likely a whole heck of a lot less than 377Ω, hence shorting out the wavefront and reflecting it, for the most part (the rest is absorbed, because of resistance, and if the shield is thin, some is transmitted as well).

I'd worry about mylar or foil getting kinked or fatigued.  Mesh might be nice.  Or regular fabric with a modest amount of metal thread integrated.  (Good luck finding some, I suppose, but it does exist -- linemen use it to work on live HV lines!  Maybe it's not so rare after all, because of the "EM is bad" crowd wanting tin-foil britches?)

Tim

[hamster_nz]

Yes, just like with transmission lines, reflection is about impedance mismatch.  The resistance of free space is ~377Ω; the resistivity of a metal is ~100nΩ·m, whatever that comes out to be, on the order of a wavelength and skin depth -- most likely a whole heck of a lot less than 377Ω, hence shorting out the wavefront and reflecting it, for the most part (the rest is absorbed, because of resistance, and if the shield is thin, some is transmitted as well).

So... when you have an antenna tuned to the frequency in question, it allows a standing wave to form, and this makes the resistivity appear to be higher than that of the material that the antenna is made out of, and removing the impedance mismatch - in exactly the same way as a speaker's impedance rises at the resonant frequency?

Oh..... (a dim light goes on in Sleepy Hollows)...

[T3sl4co1l]

Um... sort of?

More to the point, if the antenna were in a reflective box, the standing waves would be trapped, and your transmitter throws a fit.  It's equivalent to a coaxial resonator, or helical resonator or what have you.

Standing waves are bad.  Unless you need them specifically for filtering out undesired frequencies, but in that case, you only need them inside the filter, and not outside the filter at the frequency(ies) of interest.  That is, you might have a transmitter, feeding a filter of some sort, but the transmitter always sees a good match at the driven frequency, while the elements inside the filter handle additional reactive energy.

So, qualified in this way, as measured at a matched amplifier feedpoint, say, you don't want SWR.

So the antenna, then.  The sheer fact that it's open to free space, reduces its SWR, both within the elements, and at the feedline.  Less SWR on the elements means lower Q, wider bandwidth.  Bandwidth is generally considered A Good Thing in antenna design; but, see the above mention of filtering and desirability of frequencies.

So, qualified again, when you don't have a special interest in some frequency band: you should desire a wideband antenna, i.e., one with low SWR on its elements.

Still following? 

So then, a wideband antenna.  This is a structure which has similar geometry over a wide scale.  A dipole doesn't: it's a fixed length, and the element diameter doesn't scale along the length.  A conical dipole does, however.  Or a bowtie antenna.  Or a horn, or some fractals (but not all of them, indeed I might even say, not most of them, because there are a lot of curve and plane fractals that people seem to draw because they look cool, but which aren't chosen because of any electromagnetic intuition, or simulation results!).

These examples share a common trait: the free-space EM wave interacts almost seamlessly with the antenna structure.  It doesn't get trapped by the structure.  The waves don't, well, stand around.

I prefer to think of simpler antennas (like the dipole) as a worse version of a wideband antenna.  By removing material, you cause resonance (and anti-resonance) to occur at special frequencies, thus cutting off reception inbetween.  Instead of a wide open window, you have an opaque mask with holes punched in it.

Going back to material properties: very little of the wave enters the material.  After all, it would be a poor conductor if that were the case, and you'd have an absorber, not an antenna. The key is to direct the waves into your transmission line, from whatever directions and phases it's set up to do.  From the transmission line out to free space, the waves are shaped, guided by the conductors, residing in the space between them.  Currents aren't carried in wires, they are carried on wires, and the voltage is carried between them.

It's not that an antenna element has an impedance -- though it's still kind of fair* to say it does have one, given a size, length, shape and frequency -- it's that the materials have an impedance different enough from the space around them, to serve this wave-guiding function without excessive losses.

*Trouble is, impedance is a scalar -- just a number.  It's supposed to be measured at a point.  A resistor at DC, for example.  You measure the current into, and voltage across, a pointlike element.  At AC, we can still work in terms of impedance (complex), where the elements are much less than a wavelength -- still pointlike.  Or where the elements are comparable to a wavelength, but we measure at an RF "port" -- a pointlike transmission line connection.  If we're talking about elements, in and of themselves, then we have to allow that we're talking about voltages and currents in different locations, like how the impedance of a dipole's elements might correspond to the voltage at the tips divided by the current at the feedpoint.  We probably lose phase information in the process, or at least we must be very careful to add the phase shift back in (from distance between points / speed of light), and hope to come up with a still-reasonable number!

And even if the discontinuity is small, if it's low loss -- and you can afford to use more material -- you can still do all of this.  Fiber optics are just dielectric waveguides.  You can make practical microwave antennas with plastic shapes.  Mirrors can be made by stacking alternating dielectrics of particular thicknesses: a lowpass or bandpass filter, which, like an electronic filter, reflects incident (out-of-band) energy.

Tim

[cprobertson1]

Well, one way of mentally ''simulating'' it can be ... These moving charges result in the 'opposite' field moving away, resulting in your reflection...

In Pozar's book, example 1.4 studies this for a copper at 1 GHz, and we find that  \$\Gamma= 0.9999382\$ and \$T = 6.181\cdot 10^-5\$ (\$T\$ is the 'transmitted' component. I'm ignoring the phase shift in both). So in dB that would mean \$S_{11}\$ equals to -0.000536 dB, and \$S_{21} = -84\$ dB. (and then after that we lose another 8 dB per 2 micrometers of copper).

I think I'm beginning to get my head round it - I'm at that peculiar stage of proto-understanding; where you sort of get it but not entirely! What I mean is it's making sense but I can't apply it yet xD

For very thin materials (so when we have air - metal - air) it gets a bit more complicated because the propagated wave reflects again when we have the next interface between metal and air, and this second reflection will start interfering with the original wave. This is what gives very thin films nice rainbow colors.

Entirely by coincidence, I was discussing thin film interference with our NDT guy just yesterday - we were idly observing a S355 stainless workpiece during welding preheat and started discussing it - we had surmised that, the refractive index of the material being unknown (thus we were working on speculation), and with the colour being dark blue-purple, that the red side of the spectrum must have been eliminated - and in order for that to happen, we made the assumption that destructive interference was occuring; which would require a phase shift - so we assumed that the path taken by the reflected ray vs the incident ray (and thus the thickness of the film) must be on the order of 0.25 to 0.5 wavelengths of the colour observed - so we reckoned a couple of hundred nanometers would be a decent estimate.

We racked our brains trying to figure out how we could measure it to see if our suppositions were true, but alas, not with our available equipment! We resorted to asking our metallurgist who indicated that the thickness was correct, but wasn't convinced of the reasoning we used to come up with it: can you confirm our refute our reasoning per chance? [see attached figure for what I'm meaning with the phase difference/path length]

I will read over this again, I hope I didn't make any mistakes in the math If you have any questions feel free to ask, I don't know how familiar you are with the physics involved so.

I'm afraid I am the veriest tyro in terms of the physics - a handful of half-remembered equations (which I can occasionally still apply accurately - much to my surprise!). Most of my RF knowledge comes from studying for the advanced radio amateur license, and hobbyist electrical engineering.

A lot of my knowledge regarding RF is practical, which lends strength to application, but oftentimes leaves me weak on the theory, which can be unfortunate if you want to try something new!

A great free starter resource is Sophocles J. Orfanidis ''Electromagnetic Waves and Antennas'', which you can download from his website: http://eceweb1.rutgers.edu/~orfanidi/ewa/. Most RF people I know tend to say Pozar's "Microwave Engineering" is the go-to reference. But, as you said, it is expensive. I must say it is also the one book that I have on my desk all the time.

Auch, excellent! That's this month's bedtime reading sorted

How large are you expecting the reflector to be?

There's an practical engineering tradeoff between using, say, a yagi and/or phased array compared to a dish. The tradeoff is typically down to size and wind load, particularly if it needs to be re-oriented. Also the relative feed size will cause aperture blockage unless you use an offset feed.

...

See also http://www.g6lvb.com/brollydish.htm which used fly screen mesh. I wouldn't necessarily recommend it as despite being a mesh, it is fine and presents high wind load.

Aye, I've actually just finished building a 13-element yagi which is likely going to be far superior to any parabolic reflector I can make.

I have two designs of reflector - a small, 2m-aperture proof-of-concept (which would have a calculated gain of about 4dB - and that's a best case figure, I reckon a figure of 1-3dB would be a more conservative estimate).

This will consist of an umbrella-like design which uses fibreglass tent poles supporting a ring of material using gravity to provide tension on the fabric between them - which can then be augmented by using cord to pull them downwards, again using the tension of the fabric to stop the dish collapsing.

If successful, I have plans for a 5 or 6 meter diameter collapsible reflector - the design would be superficially the same, but with an added semirigid support around the perimeter of the parabola which, when coupled with a number of rigid struts would again use gravity to pull the fabric taught (at the expense of blocking some of the aperture). The larger dish would also be suspended from a point just above its focus, and would subsequently hang like a pendulum; this allows me to tilt it a little while still using gravity to provide rigidity to the structure.

The larger dish would obviously be quite a big project - and a lot of care would be needed in selecting the materials for low weight and wind load!

Small dish first though

70cm is quite a long wavelength compared to chicken wire (the small stuff used for baby chicks).
Surface perturbations on a dish can also be quite large, before there is any discernible effect.

Indeed, chain link fencing would easily work at 70cm. It's another engineering compromise, you need to take the weight into account. I'm not sure how much difference there is in wind loading between chain link fence and chicken wire. The aluminium mesh I've used in the past is light weight compared to chicken wire, but exhibits a relatively high wind load. The Flectron copper woven material I've also used is reasonably light weight, and handles far better than any metal mesh when folded, but it has a very high wind load not to mention being expensive.

Chickenwire is a very good candidate actually - though I wonder if the weight will be of concern! On the other hand, the wind loading will be great with it (relative to other "closed" materials).

I'm also looking at using what is basically a net woven from copper wire: one can weave it with relative ease using a breadboard (the kind for cutting bread ) with a number of equally spaced nails around the edges and then looming it across that with a knitting needle. (You would put a simple knot where the weft meets the warp - nothing fancy, though it is stronger in one axis than the other. I've actually made clothes from nettle strands using pretty much this exact technique in the past - it's a little on the time consuming side, but it's not difficult.

An A4 sized panel will probably take an hour to make - so let's see, for a 2-meter dish we're looking at a bit over pi square meters, lets call it 3.5 square meters - at 0.075 square meters per hour thats... okay that's 46.667 panels total - so I'd need 48 panels including the perimeter... which is 48 hours of weaving or so...

Never mind that's a terrible idea.

As a side note, I'm also looking at getting started in microwave/mm bands within the next three years - building reflectors for them will probably be a much greater payoff than for the 70cm band!

Bonus: here I am with Skylab & Shuttle astronaut Owen Garriott http://www.g6lvb.com/dayton_demo_2002.htm. Coincidentally I gave his son Richard Garriott (space tourist, Soyuz & ISS) a lift from London to Warwick and back a few years ago for a talk shortly after his trip to the ISS. Never have I been so delighted to be stuck in a two hour traffic jam.

I'm totally jealous xD

I managed to score a chat with the systems engineer for the Saturn V Instrument Unit when I was visiting the Kennedy Space Centre a few years back - the guy had also worked on Mercury-Atlas, Agena and Gemini - but I can't for the life of me remember his name! I'm certain it was "Jack" but I've been searching for a good half hour now and I can't find him online =/ Wonder where he's got to!


So the main contenders for material just now are perforated foils, wire netting/chicken wire, and very light metallic fabrics. I'm going to try making a section of copper weave/netting at some point to see how long it takes - I can drastically increase the size of the weft/weave (up to a tenth of the wavelength, I think dmills was saying) to increase production time - maybe I can cut that 48 hours down to 24 hours or less - even then, that's a LOT of work (I would do maybe an hour a night so you'd be looking at a month just to get enough material together.)

I'll try other methods first before properly attempting to weave my own

[TheUnnamedNewbie]

Entirely by coincidence, I was discussing thin film interference with our NDT guy just yesterday - we were idly observing a S355 stainless workpiece during welding preheat and started discussing it - we had surmised that, the refractive index of the material being unknown (thus we were working on speculation), and with the colour being dark blue-purple, that the red side of the spectrum must have been eliminated - and in order for that to happen, we made the assumption that destructive interference was occuring; which would require a phase shift - so we assumed that the path taken by the reflected ray vs the incident ray (and thus the thickness of the film) must be on the order of 0.25 to 0.5 wavelengths of the colour observed - so we reckoned a couple of hundred nanometers would be a decent estimate.

We racked our brains trying to figure out how we could measure it to see if our suppositions were true, but alas, not with our available equipment! We resorted to asking our metallurgist who indicated that the thickness was correct, but wasn't convinced of the reasoning we used to come up with it: can you confirm our refute our reasoning per chance? [see attached figure for what I'm meaning with the phase difference/path length]

That figure is indeed what is going on and gives those rainbow colors depending on thickness.

[vk6zgo]

The ARRL Handbook, or their Antanna book had such a demountable dish back in the 1970s.

Alternatively, if you don't require the maximum gain you can get, maybe a "corner reflector" design would be adequate, as those are easy to make in a form that may be easier to break down.

Most 70cm antennas are either Yagis, Quads, Phased arrays (both in line or broadside) or helicals.
All of these are easier to make in demountable form.

[cprobertson1]

A corner reflector design is an interesting one - relatedly, I was also considering a cylindrical parabolic antenna which greatly reduces the size in one dimension (at the expense of gain/directionality, giving us a fan shaped lobe rather than a relatively thin one)

Helicals are an interesting one actually - one of my future plans is for a helical antenna made from microbore plumbing; I could probably do with a VNA if I were to go ahead with that though (but then again, I could probably do with a VNA for prototyping the parabolic reflector as well - and my lack of a VNA isn't stopping me with that )

Of course, just building a long-boomed yagi is probably easier than either of these options. I will give the corner reflector a go though - I was considering using it to direct the driven element of the parabolic antenna towards the dish - so building a small one would be a remarkably useful exercise - thanks for the suggestion!

[Neganur]

Orfanidis’ books are nice, with MATLAB examples and functions available online.

[Howardlong]

Axial mode helicals are practically speaking quite easy to make as they are pretty wide band so are tolerant of dimensional aberrations. The caveat here though is that you’ll need a matching method of some sort, which might not be wide band.

Other comments from my experiences are that any supporting boom, whether conductive or not, should be in the centre. Non conductive booms have a dielectric effect which can signifcantly shift the boresight and reduce the gain.

A similar caveat applies to using a non-conductive tube as a permanent former: the tube again has a dielectric effect, detuning the antenna, and may well be lossy too.

The way to minimise the support boom’s effect is to have it going through centre, and use standoffs, say every 3/4 or 1 1/4 turns.

[cprobertson1]

So I've finished testing some materials for their physical properties (not their electrical properties - as I don't really have a way of testing how radio-reflective they are! I have based my materials on the assumption that they are conductive across their surface - but I've not looked any further into their electrical properties than that.))

Testing consisted of making a  45 degree cone out of the material, putting a rigid ring around the bottom, and suspending it from the vertice with a pole in my garden.

I'm a HAM so the neighbours are used to weird metal things in my garden

Manual copper weave:
-Time consuming (about two hours per square meter)
-Would probably work for very small antennas but too time consuming to be useful for a large dish unless you built a loom of some sort.
-Difficult to weatherproof properly
-Cupric/cuprous oxides aren't very conductive - and so weatherproofing would be required - BUT weatherproofing may reduce the conductivity between the individual strands. Bit of a catch-22!

Prefabricated aluminium weave:
-Managed to find a source of aluminium filter-mesh (I think it was destined for coffee filters?) Would probably work but reasonably expensive.
-Construction has to be spot-on at the seams; otherwise it tends to fray.
-Weighs a lot when wet (water clings to it - so if you poured a glass of water on it, it will spread out a bit until it becomes saturated, and then it will start passing through) - probably unsuitable for Scottish weather!

Aluminium foil:
-Very cheap
-High wind load
-Tears easily
-Dishes will require drainage to prevent water pooling
-After discovering it could be torn easily, birds stole most of it.

Aluminium foil backed with cloth
-Heavy
-High wind load
-Cloth would have to be waterproofed with wax or oil, making it even heavier. Unless you like moldy antennas.
-Dishes will require drainage to prevent water pooling.
-Birds pecked holes in it (most likely some insects were trying to make their home in it)

Aluminium wire spiral
-Fairly light
-Low wind load
-Difficult to construct (but not as time consuming as a weave)
-Birds like to use it as a chair.


And so it was that I gave up and built a yagi.

Still, it was fun to experiment! I'm looking forward to my textbooks arriving so I can start to really understand how reflectivity comes about - but in the meantime I shall settle with my yagi-uda antennas (nobody ever seems to mention Mr Uda... poor guy always seems to get forgotten about!)

Thanks for the info, folks!