DIY low cost RF absorber for a 4' x' 4' x 4' anechoic chamber ???

[rhb]

I'd like to do some physical modeling and VNA measurements of a dipole over ground in the 200-1600 MHz region.  This naturally leads to the question of environmental effects.  I can't afford to buy commercial absorbing materials for this, so I'm looking for alternatives.  Iron and charcoal powder in a suitable matrix appears to be the way to go.  For ease of handling, I'd like for it to be lightweight, which implies using foam as the matrix material.

@cdev mentions having done this, but made no comment about the foam he used.  I need 80 sq ft and am at a loss to think of a material I could use to make foams with varying proportions of charcoal and iron.  Assuming 3" thick sides, that's 20 cu ft of foam.

Any suggestions?  At the moment the only thing I can think of is to mix the charcoal and iron powder with paint and then mix that with loose expanded polystyrene beads.

[RandallMcRee]

I don't know anything about RF chambers but I have worked with structural foam quite a bit, for creating super-strong carbon-fiber epoxy parts.

Structural foam is easy to work with but I can't recommend anything with my lack of knowledge of what you are trying to do.

A good vendor is Aircraft Spruce, of course, you have to wade through all the kit airplanes and trick aero stuff...
http://www.aircraftspruce.com/categories/building_materials/bm/menus/cm/foam.html

I would generally stay away from the liquid foam products--they always sound like the perfect solution to any problem but getting them to actually do what you want in the time available (liquid for a few minutes) is (for me) impossible.

[rhb]

Thank you for your suggestion.

RF absorption materials embed carbon and iron, or similar materials in a matrix.  A key requirement for suppressing reflections in an anechoic chamber is that the gradient of absorption be small.  Large gradients will produce reflections.  The square pyramid shape of commercial acoustic and EM absorption panels is motivated by this requirement.  So long as the pyramids are small relative to the wavelength, it has the same effect using a medium of constant properties as a layer with varying properties.

Until you brought it up, I'd not considered solid foam sheet because I could not see a way to impregnate the sheet.  However, one might well form wedges or pyramids of foam painted with an absorbing paint, though in this case, it might be difficult to make the dimensions small relative to the wavelength.  But it could certainly be useful  for suppressing corner reflections.

[edpalmer42]

Some black anti-static foam is semi-conductive.  The kind I'm thinking of is the stuff that you'd store through-hole ICs in by pushing the pins into the foam.  I did a quick test of some pieces that were handy.  Different types, thicknesses, etc.  Size around 10x10 cm.  Using two coins for contacts and pushing down on the coins, I got resistance values in the 1-100K range.  I would think that would be a good RF absorber.  The random nature of the fibers and spacing would ensure that there were no resonant effects.  You'd have to try some to make sure that it would work for your purpose unless you could find some that was actually spec'd for use as an RF absorber.

Ed

[rhb]

It's my understanding from reading product datasheets that you need a conductor for the E field and a magnetic material for the H field.  Hence the carbon and iron powder.  I suspect the good stuff uses powdered mumetal.

[2N3055]

There is a ready made graphite spray... Spray a layer, sprinkle with ground ferite. Dry, repeat.

[coppercone2]

There is a ready made graphite spray... Spray a layer, sprinkle with ground ferite. Dry, repeat.

I wonder if you used a spray paint gun and a bead blaster gun , how good a product you can make with very thin sprays. Both very cheap from harbor freight and you would not really care about the things you typically care about for paint job quality, and maybe a third spray paint gun with primer or some kind of filler compound if you need physical spacing. I can see myself building some kind of RF lasagna if I could wield two spray guns akimbo like a john woo movie, maybe with a foot switch on a heat lamp aimed at my mold to solidify it after each coat. It would take a long ass time but you can do it? Or maybe you can make a X/Y scanning laser to solidify it, some kind of UV cure foam binder maybe, i wonder if you can emulsify UV cure adhesive, it cures mad quick.

I wonder if you could grow trapezoids of the stuff using 3 spray guns and alot of time. Ferrite powder has to be cheap since you can buy inductors for so little. I don't think it makes sense to shield the really low frequencies at the foam level, you should just use mu-metal backing plates, they sell it alot for EMI crazy people that want to protect their houses from their junction box.

If you get the process down you could use solenoids to automatically trigger the guns, all taped together (like ripleys flame thrower pulse rifle), if you get the motion down, so aim it, and it gives you a few seconds of graphite loaded UV cure foam, then a few seconds of ferrite powder spray, then turns on a big UV led that you aim at your masterpiece before hand, all foot pedal operated, so you can do thousands of operation without destroying your hands, so long you get the rythem down. Or if the process dynamics are not too much of a pain in the ass, you might be able to mix it in the aresol phase by activating the sand blaster and the spray gun at the same time. It seems like you would need to try it to see if you can tune a good continuous distribution. Hell you can even easily do voice controlled solenoids, I imagine this would go on like 0.5mm at a time between cures, or slower... maybe throw in some stiffening strips of carbon fiber or some shit every so often so its strong.

You could actually make a big ass piece then cut it easily with a bandsaw to make your cones, rather then trying to grow it in cone molds.

Does anyone know the proper ingrediant? I can see something of very high quality being made this way without difficult shit like distributing everything in expandable foam to grow properly, it might be a pain in the balls if it starts sinking to form heavy layers. Unless you spray expandable foam real light, let it grow a bit and then spray it with the other stuff with layers...

I am convinced commercial pricing would hammer the hobbyist on buying this stuff. It might not even be that complicated.

I found that air-powder distributions are really good, even with large particles. If you are ever doing a garage epoxy floor from home depot, and you need to distribute those decorative flakes evenly (which will make it look great even if its all stained and fucked up from acid/solvent damage etc), I found the best way is to get a small leaf blower and put the flakes out in a cupped hand and just hit it with the leaf blower, works real nice compared to throwing them into the air, you get this nice evenly distributed cloud.

Maybe mixing UV cure adhesive with something like cabosil (fumed silica) would make a decent space filler that would kinda be foam like even if its not really foam. Cabosil is horribly voluminous for its weight. Or maybe aerogel powder of significant mesh can approximate foam where foam should not exist. I don't see commercial UV cure foam spray solutions.

Cabosil should be transparant to UV light some what, so maybe you can use a regular UV adhesive spray, then a dusting of cabosil mixed with the E/M conductors, if its not possible to spray the cabosil while mixed with UV goo.. might actually be more uniform then real foam without all the engineering crap to figure out foam behavior. Or maybe even using fiberglass 'powder' as a filler?

[rhb]

2N3055 and coppercone2  Thank you!!!!!!!!!!!!!

As it happens, I have a Marshalltown texture gun designed to shoot popcorn ceilings.  I hate them, but bought it to repair some damage while painting a house I bought 25 years ago.  Except for recently shooting orange peel in a bedroom and bath for my sister, I've never used it again.

Mix  aerogel beads to reduce weight and  carbon and iron powder with lightweight sheetrock  mud. Vary the layers from a lot of carbon and iron for the base coat and lesser amounts in each additional layer to provide a gradual impedance transition.   It's cheap, fireproof and easy to do.  If I shoot a light coat of latex paint between layers of texture it should be quite durable and easy to repair if damaged.

I *love* this forum!  I don't think I'd have ever thought of doing that, but it's the perfect solution for DIY RF absorber.  It will be interesting to compare it to open sky with the aluminum sheet on top of a fiberglass stepladder in an open field.  Figuring out how to make a box with no flat rectilinear sides will be interesting, but some 1/8" luan sheets with exterior formers should do the trick.

[coppercone2]

is iron powder the best tool for the job here?  Won't ferrite be better and cheaper and more corrosion resistant (I think the sulfate might react with the iron powder).


How much of a spacing do you need between layers on this stuff? I thought it would be like fractions of a milimeter.

Can you detail your work? I can't do any of it right now. Cabosil is dirt cheap, I used to use it for pyrotechnics, and it flows amazingly.

What is a commercial solution going to do better? (seriously, I want to know what makes that shit so fucking expensive it rages me when I tried to estimate the cost of making a RF wall for an antenna analyzer so you can spin an antenna infront of it to measure radiation pattern). There is a serious chance of some kind of fat cat business man making all our lives difficult with something very simple, I never saw it explained in good detail. Unless those additives they use are like the price of gold, that stuff must have INSANE margins for manufacture.

You can probably make some kind of spray gun assembly to run back and forth on a rail with the most simple parts if you can find a process time constant that works, using the most simple of electronics, Hell you can mount a RC car to a 2by4 to make something wide set on the floor so you can cut it up later, like drive it upside down on a ceiling made of some particle board over a partical board mold lined with some saran wrap, maybe like let it dry every so often and remove the build up on the sides to ensure there is uniformity and it does not curve over, and then make it too wide so you can cut the edges off later so you get a nice lasagna

The tolerances for making some kind of oversized foam structure just seem loose as hell

Your idea should also be less flammable, since it uses gypsium, and there is no air voids to melt away since its filled with some kind of inert light crap

[2N3055]

I also would like to know how a simple steel wool would perform.... Just attach it to surface and fluff it up ....

@coppercone2
I did suggest ferite... It shatters easy, and you shouldn't try to grind it too fine and uniform...

[coppercone2]

No I mean ferrite particles instead of iron particles, the stuff they sinter together to make inductors. Why would it have different performance in a matrix structure then iron powder? I think the domains would have a better frequency response then iron too. And it might be cheaper, more oxidation resistant, and lighter since its a alloy of zinc and nickle or whatever and its used in massive quantities, and I don't think its flamable like iron powder is.

I know if you use it solid its more fragile, but a powder is a powder?

[coppercone2]

can you sketch out a pyramid of what you have in mind to get me an idea of what the correct layer structure looks like (like food pyramid with distance of seperations etc)

[David Hess]

My solution when I needed to do this was to conduct the test in an empty lot on top of a hill.  The lot I used was not quite empty; I could measure the reflections from trees and lamp posts in the distance but they were small enough to ignore for my purposes.

[2N3055]

No I mean ferrite particles instead of iron particles, the stuff they sinter together to make inductors. Why would it have different performance in a matrix structure then iron powder? I think the domains would have a better frequency response then iron too. And it might be cheaper, more oxidation resistant, and lighter since its a alloy of zinc and nickle or whatever and its used in massive quantities, and I don't think its flamable like iron powder is.

I know if you use it solid its more fragile, but a powder is a powder?

I use word ferrite here for a ground magnetic cores.

Ferrite is actually a name for an allotrope of iron. It is iron, pure one.  Fine powder iron is mostly ferrite.
But, iron particles in cores are suspended in ceramic material. Which is good, because pure iron powder will become iron oxide very quickly.
Ceramic matrix in cores encapsulates and protects iron from oxidising. So it will keep it's magnetic properties. It also isolates particles from each other, preventing currents forming in core.

[mmagin]

I have sandwiched a layer of the silvery static shielding plastic bag material under the lid of a 10 GHz amplifier to stop it from oscillating (which it did, in other bands, with the bare aluminum box lid.  Not sure how it compares to proper RF absorber materials, possibly it was just good enough for these particular frequencies and the geometry of the box.

[2N3055]

My solution when I needed to do this was to conduct the test in an empty lot on top of a hill.  The lot I used was not quite empty; I could measure the reflections from trees and lamp posts in the distance but they were small enough to ignore for my purposes.

That is a valid technique and reason that I bought Signal Hound SA.. With a notebook, it is portable...

Problem is that where I live, it is increasingly hard to find a spot that has no RF activity...

[2N3055]

I have sandwiched a layer of the silvery static shielding plastic bag material under the lid of a 10 GHz amplifier to stop it from oscillating (which it did, in other bands, with the bare aluminum box lid.  Not sure how it compares to proper RF absorber materials, possibly it was just good enough for these particular frequencies and the geometry of the box.

Propper RF absorber is broadband, so it should work for (absorb) wide span of frequencies, preventing resonances and reflections..

[rhb]

All I know at this point is from reading datasheets for $$$ commercial material. In particular elastomer sheeting with carbon and iron powder in it.  For low frequencies the magnetic particles need to be larger.  So I  think I'll start saving the steel chips from my bandsaw.  It might well be that chopped up steel wool would be useful.  There should be quite a bit of literature, though I'm sure a lot of it is classified because of the application to radar reflection suppression.

I've got a couple of WiFi feedhorns, so when I get my xaVNA I'll point them at some fluffed up steel wool and compare that to a plain steel sheet and also to test various spray mixes.

[coppercone2]

I still don't understand why iron powder instead of a ferrite powder?

http://www.comcraft.co.jp/products/temex/data/tem01_ferrite.pdf

It looks like a MgMn ferrite would be good, look on page 11 of the document on the table that says FRQ range. I thought the dipole movement of iron is only good to maybe the low MHz range if that.

[rhb]

To be truthful, I have no idea.  In this application losses are desirable whereas in transformers they are not.

In the end though, for this project, materials which are easier to source are more desirable. It's crazy enough even without spending a lot of money and time.

Once I get my xaVNA and start measuring the reflection from an aluminum sheet with various materials on the face I might actually know something.  At present I'm drowning in a vast ocean of ignorance.  Hopefully, I'll learn to swim soon ;-)

[coppercone2]

yea actually I am thinking about it and I am not sure what material is the best/where its lossy

[T3sl4co1l]

Coppercone being coppercone...

The normal way it's done is radar absorbent foam pyramids on top of ferrite tiles.  I've seen hollow grid tiles and plain flat tiles.  I suppose, preferably, the one on top of the other in that order.

I think the ferrite is a particular grade, but it's hard to come by -- probably because you'll only ever order a pallet worth at a time, building test chambers.  Who needs just one or two, or a hundred pieces, right?...

You can probably do pretty well with a fairly large stockpile of NiZn or lowish mu MnZn plates, commonly sold for planar magnetics, or various EMI purposes (you can find perforated plates for connectors, and flat plates for near-field EMI).

Tim

[ogden]

To authors of "funny anechoic chamber build ideas": wheel was invented long ago 

https://emcfastpass.com/anechoic-chamber-guide/
Basic Rules for Anechoic Chamber Design

Im afraid that 4' x' 4' x 4' anechoic chamber is too small for 200-1600MHz antenna measurements. You would want to measure antenna quite into far field region.

My suggestion: drop anechoic chamber idea and consider open field tests (get portable AC generator and maybe tent as well).

[coppercone2]

Well I mean alot of people are interested if someone works out a specification it might be economical to buy a pallet and distribute it to a few people that have the means and will to use the stuff to build their own chambers

The only RF lab I worked in had cardboard pyramids with two sides only that had some ferrite on em and were stuck to the wall with double sided tape. I was told it was the most economical, but I suspect that if you do something like antenna directional testing, you will want full 3d pyramids made of foam. I don't think that chamber went past like 2GHz.

[rhb]

Nice link!   Thanks.

My goals are rather different than typical applications of an anechoic chamber.

My interest is in measuring the scattering parameters of a "fan antenna" as the element placement is varied, so a rather different case than typical.  If needed the xaVNA goes to 3.5 GHz,  so I can scale the antennae and/or build a larger box.  I plan to compare it to an open area site with some basic 1/2 wave dipoles so I know what the distortions are.  If they are low enough I should be able to correct for them.

Recording studios avoid rectilinear shapes to avoid resonant reinforcement of reflections from walls and corners.  And suppressing unwanted reflections is a common problem in reflection seismology.

But this is all hobby science.  Cheap rules and failure not important.  The original question, what to use for an absorber,  was answered.  Aerogel or foam beads, lamp black, powdered iron and drywall mud are cheap and readily available.  Ferrite powders are probably harder to source and likely pricey.  I should be able to get a good idea of how effective materials are using the xaVNA and a pair of antennae pointed at a sample at a short distance.

The chamber size is dictated by what I can conveniently store.  If I have to go much larger I'll just use cardboard and throw the thing away when I'm done.