Building a Hardware Store Faraday Cage

Most Hackaday readers are no doubt familiar with the Faraday cage, at least in name, and nearly everyone owns one: if you’ve ever stood watching a bag of popcorn slowly revolve inside of a microwave, you’be seen Michael Faraday’s 1836 invention in action. Yet despite being such a well known device, the average hacker still doesn’t have one in their arsenal. But why?

It could be that there’s a certain mystique about Faraday cages, an assumption that their construction requires techniques or materials outside the realm of the home hacker. While it’s true that building a perfect Faraday cage for a given frequency involves math and careful attention to detail, putting together a simple model for general purpose use and experimentation turns out to be quick and easy.

As an exercise in minimalist hacking I recently built a basic Faraday cage out of materials sourced from Home Depot, and thought it would be interesting to not only describe its construction but give some ideas as to how one can put it to practical use in the home lab. While it’s hardly a perfect specimen, it clearly works, and it didn’t take anything that can’t be sourced locally pretty much anywhere in the world.

Faraday Cage Design

At the most basic level, a Faraday cage is an enclosure made of a conductive material that blocks electromagnetic fields. In comparison to a Faraday shield, the cage variant is not a solid object, but rather a metallic mesh. Among other advantages, this allows observation of the subject inside of the cage. Put simply: if you simply want to protect a device from interference (or prevent it from causing interference) then it’s enough to enclose it in a metal box; but if you want something that you can experiment with, you’ll probably want a cage.

The trick is to make sure the holes in the cage material are smaller than the wavelength you wish to block. It’s the same principle that allows you to use standard chicken wire as a RF reflector as long as you’re working with relatively low frequencies. But as your target frequency increases the wavelength gets small enough that it can sneak through chicken wire, so you need to use something tighter. But how small is small enough?

To start, we need to find the wavelength for the frequency we want to block. This can be found by dividing the wave’s speed in meters per second by its frequency in hertz. As we’re dealing with a radio wave we know it will be traveling at the speed of light, and for the frequency let’s say we want to block 2.4GHz. So the math will look like:

The rule of thumb for a Faraday cage is that the openings should be no larger than 1/10th of the wavelength, which in our case is 12.5 mm (approximately 1/2 inch). As luck would have it, steel “hardware cloth” with mesh sizes of 1/2″ and 1/4″ is widely available. On paper either should work, but I did end up going with the 1/4″ to be safe.


With the mesh in hand, the next step is to build some kind of frame for it. As it so happens, Home Depot has wooden crates in their storage section which are strong and relatively cheap. You could also construct a frame from pieces of wood or PVC pipe, which may end up being cheaper if you don’t mind taking the time to build it.

I popped a couple of the slats out of one side of the crate so it would be easier to see inside, but beyond that, the construction simply consists of wrapping the crate with the hardware cloth. I did one long piece that started at the front and wrapped all the way to the back, and then two smaller pieces to “cap” the sides. In the end, it’s not entirely unlike gift wrapping; if the gift wrap was metal and had a nasty tendency to cut you, anyway.

One thing to pay close attention to is where pieces of the mesh overlap. You want to maintain a good electrical connection and avoid any gaps, so you should overlap the pieces by at least a couple of inches to be safe. I attached the hardware cloth to the crate with a power stapler, so I also made sure to drive a few extra staples through the areas where the mesh overlapped to ensure they were held together tightly.


As a simple test, I set my phone up on the bench running the signal strength function of the “WiFi Analyzer” application for Android, with an Access Point that’s one floor above selected as the target. In the two pictures below, the only thing that has changed from one shot to the other is placing the DIY Faraday cage over the phone.

As you can see, the phone had a signal strength of approximately -55dBm originally, and it dropped down to nearly -80dBm when inside the cage. Weaker Wi-Fi networks were rendered undetectable when the phone was inside the cage, and there was a clearly detrimental effect on the phone’s LTE reception.

Flawed Design

I should say, before the commenters below get a chance to do it for me, that this is admittedly not a very good Faraday cage. For one, it’s not fully enclosed. Since there’s no bottom signals are still able to enter from below, greatly reducing its effectiveness. The steel mesh is also not an ideal conductor, and copper would likely work better. But finding a local source of tightly wound copper mesh proved tricky.

There’s also some debate about whether a Faraday cage must be grounded or not. For what it’s worth, during my testing there was no observable change in performance when the cage was grounded. Though it’s possible a more conventional Faraday cage may perform differently.

Having said that, I still feel this design blocks RF enough to be useful. It will never completely isolate the device inside from electromagnetic interference (or vice versa), but it attenuates signals significantly enough to be clearly observable. That was the extent of my ambition to begin with, so I’m happy with the results.

Possible Applications

Imagine you are developing or testing a remotely controlled device, and want to see how it behaves when signal strength is poor. Tossing the cage over it would allow you to induce a drop in signal strength instantaneously. Or perhaps you’re observing the RF emissions of a device, but want to cut-down on superfluous background noise. Putting the piece of gear under test and your SDR hardware inside the cage would be an easy way to study it in a less “noisy” environment.

Those are just two possibilities. This demonstration shows there’s really no good reason not to have a simple RF blocking device at your disposal. It’s cheap, it works, and it can be your next weekend project. What you do with it is up to you, just make sure you drop us a line when you figure it out. What would/do you use a Faraday cage for?

66 thoughts on “Building a Hardware Store Faraday Cage

  1. If you’re looking to isolate a small RF device, you can go to the hardware store and get a gallon paint can with lid and use this. With the lid sealed there’s NO RF getting out. I’ve used this to test or isolate a lot of RF AMI endpoints with terrific success. You can even put RF connectors through the lid and run a coax or antenna on the inside for measurements or to ensure that the signal you want is the only thing that gets out, not any digital noise or spurious emissions. And it’s CHEAP!

    1. I was working on a bluetooth beacon project. We had around 50 of the darn things we had to setup, test for a site. There was SO much noise in the office from all of them broadcasting. The design was such that to turn them off required popping the batteries out of each. We resorted to putting them in the microwave in the break room. I should have thought about an unused paint can, that would have been a much better solution.

      As it was, “Daniel, can you swap these beacons out for some others…and while you are there, can I get a coffee?” :)

      1. I’m working one a BLE beacon project next week. I’ll be finding me an old paint can (as most new ones seem to be plastic!) and building a bad faraday cage to simulate RSSI changes more easily than throwing them down the corridor.

        1. Home Depot sells new/blank empty metal paint cans in the paint dept. Great for storing handheld radios and batteries. A retired kitchen microwave (appliance) is excellent, too. It’s basically a Faraday cage with a moderate storage cavity for your goodies. Stick it in the attic or garage rafters so as not to bother your wife. :)

    1. I own many microwaves. I can’t really do much with them anymore though as they have traveled out into the stars where i have no way to retrieve them. Fortunately I can make more whenever I need to.

        1. It’s a bit like being the owner of a fart. Using the possessive makes sense to me for anything created at a source point that radiates outward. I remember in grade-school: “Yes Mrs. Blaskovich, that was Jimmy’s.” (Those wooden-seated school chairs could resonate like a Stradivarius and often did. :-) )

    2. Everybody owns a microwave oven, and if you just want a quick isolation put your gadget inside the oven. (Don’t turn it on….)

      Good scroungers can find dead microwaves, and you can gut it and wire the bulb to illuminate the interior if you need to see what’s going on inside.

      Instead of using steel hardware cloth, you will probably get better performance out of aluminum window screen. And it’s probably cheaper. But you can’t easily solder to it, so you’d have to make mechanical connections.

      You can get even better performance by making a double-wall cage. If you get 25 dB out of one layer, you’ll get 50 dB out of 2 layers. I’ve sat inside such a cage and ran an oscilloscope while a megawatt-level pulse transmitter banged away just feet away. It was built with copper screen.

      1. You can use old washing machine cases.

        At lower frequencies I’m guessing an anechoic chamber design is required for the audible ranges and since if not thick enough metal; the metal can be resonated to a certain extent to create other combinations, undertones and overtones depending on design, tension I guess or just, resonance of the structure considering overtones if solid metal versus a screen. Due to overtones of drum membranes causing more signals since are more than just typical other designs fundamental frequency harmonics.

        Bury an inter-module shipping container 4ft underground at least and assume baffling and screening the intake and exhausts as well as opto coupling or isolation transformer coupling with monitoring of spectrum at those points. Maybe lining like a anechoic chamber would be the next best thing.

        One gentleman who owns a submarine mentioned similar for the low EMF that can get through underwater if say did an underwater system if needed. From what I’ve read that isn’t required. I think he uses the submarine for other purposes that are recreational I guess.

        I use 325 stainless steel mesh screen cut to fit the inside of the NIDO powdered milk can lid to complete the metal lining of my milk can I’ve used for close range determinations when radio directional finding.

        I also made sure to use aluminum screen for the intakes and exhausts on my farm building that is steel lined though still have room to improve where more elaborate designs use, special I think, copper screen lined windows or maybe there are tints that are effective in more than the AM range I’ve researched.

        At higher frequencies in microwaves and higher… welded seams are required and special reflective materials don’t hurt either, i.e. say like certain oxides when you hit the IR range is more commonly known.

          1. I didn’t realize the above comment was going to be approved. My fault for the redundancy in thoughts.

            I’m a victim of interstate domestic violence and gang stalking by those that not only have poisoned me using chemicals and potentially biologicals… they use sound, body and mind control assault weapons and tactics and I’ve even been home invaded, armed robbed and maimed including being shot by more than the malicious lost their jobs and pension police.

  2. Tom: I think a good test for comparison would be to just stick your same phone in the microwave oven. Obviously, you do not want to turn it on, ha ha. Just see what your signals look like when inside it. It would also be good if you carried your oven to the same location that you tested your cage, or bring your cage near where your oven is. I have seen tests of microwave ovens where a cell phone was placed inside 5 different models and then called. If I recall, the phone rang in 3 of the 5 ovens showing some pretty good leakage. Pretty scary.

    1. And yet three out of five microwave owners don’t display signs of exposure. The level of radiation it takes to make a phone ring is enormously different from what it takes to cause injury. They aren’t designed to block cell signals, after all.

      1. That’s almost like microdoses with chronic exposure don’t cause issues unless diagnosed, if even disclosed as causation, at a later date… like banned pesticides for example.

        Using just microwaves from a household oven, I demonstrated in my undergraduate research different synthetic products compared to the same classical reactions hot plate heating methods. Seems I demonstrated that pissed off the mafia also.

        I sometimes wonder if that caused more surveillance on me as I wanted to try using different theoretical and known microwave rotational bond and IR vibrational bond activity ranges also as a catalyst and see what the effects on the products were thinking that the bonds energies resonated would catalyze reactions to cause specific products to be formed in some calculated way that correlated. Seemed the same with NMR (magnetic transition resonance) and UV-Vis/X/Gamma (electronic transition resonance).

          1. Read into microwaves causing immunosuppression and hormonal changes. From what I understand the amounts aren’t at the levels where “heating” occurs… whatever you’re considering to be “heating”. Technically, any energy unless sympathetically resonating in tune will be converted to heat. Could be lower intensity pulse train effects that, like cause microwave heating, can cause cavitation or pressure issues in delicate structures like capillaries making them burst or even mast cell issues that could cause scarring over time maiming the integrity of the organ structure and function.

          2. Worked around RF for many years. VLF to X-band. As of the early 90s the only recorded case of a person getting cancer from RF (and that is a maybe) was a guy working at the top of the Empire State building in NYC. Transmitter room. If there was a problem with RF exposure causing disease, I would expect airport workers to suffer the most. They are exposed to everything from HF radiation to microwaves from radar. All at significantly higher levels than most of the public.

          3. I’m educated guessing, not from an EE or RF Engineering major background, that the main stream consumer systems aren’t going to be dangerous COTS… or at least as dangerous if used safely. The systems that are dangerous are those that are designed to be dangerous… say more like collimated to fine point systems like a maser/laser (theoretically now days can be technically nanoscopic) that can ablate or mutate to cause scaring coupled with immonocompramisation can lead to spread of cells and tissues not supposed to be where are, RF cautery tools, Cyber/Gamma Knives, Corrupted signals AED systems (like reset the heart or EEG’s in ill ways), unreasonable used stun guns, Medical equipment gone missing, electron beam guns, etc. Bob Iannini had some interesting designs from the 1980’s and John William’s of Lonestarconsulting Inc. demonstrates some challenging to argue with logically systems. I mean… for $14 you can buy an RF Cautery tool now days.

            I like to describe the technology like starting with the telegraph. Most don’t even realize the modulation and multiplexing used back in the 1800’s for that system of communication. Now there are more advance Direct Current and Alternating Current communications systems methods that aren’t only wired like our telephones, or back in the old days ones that were cordless on a hardline, there are cell phones (like sleeper cells too though that is another topic) that have coverage almost anywhere now days with advanced modulation and multiplexing that are even satellite based that most aren’t aware of in the planet.

            So, like the advanced tech most aren’t aware of that exists and now with satellite based systems coupled with the land, sea, air and cyber battlefields… who knows what space warrior operations are going on that can affect and assault without consent the minds of our youth in worse than MK Ultra Projects and other horrific mind control projects. This going on in the new frontier of the Electromagnetic Spectrum (EMS) Weapons systems to perform more than just sound assaults with intent to maim and murder… they can perform body and mind control assaults with intent to maim and murder operations.

            Check out first post and other posts.

          4. I wonder a few issues like:
            What is the life expectancy of those workers around RF/Microwaves/Ionizing Radiation?
            Why aren’t the fertility rates high like prior to wireless systems implemented?
            Why isn’t the life expectancy increasing for those families that had longer life expectancies?
            Why are, or at least were, there alarming rates of Alzheimer’s and in some communities the elderly are dead and aren’t there anymore?

            I know from my Chemistry background the issues are due to chemical and biological reasons if not kinetic… however, I am confident electro/magnetostatic and electromagnetic issues are a causation also that is being neglected other than a few cases that made main stream media, i.e. sonic assaults, satellite dish harassment/injury and X-Ray/laser used in a premeditation to murder.

          5. I forgot to note another source that I’ve read causing an alarm… gamma ray sources used for inspections and I’d guess maybe irradiating to disinfect… though I’d think the disinfecting systems are lower power. Not sure how those can be lased either… though I’m thinking isn’t very complicated if I read into.

            Thanks to Kathy Loves Physics latest post for reminding me how easy to make X-Rays and electron beams technically (those got banned for causing radium poisoning like issues when were COTS):

    1. For the wavelength, yes. But you then divide by 10 to get the ideal mesh size, so 125 mm becomes 12.5 mm. Looking at it again, I see how that might not be very clear the way I worded it.

  3. Leaving an entire side off of the cage is pretty silly — you could have achieved similar results by setting your device in a metal bucket. Since the box is larger than half a wavelength at the frequencies you’re using, there will be standing waves in the box that will make performance sensitive to exactly where the device is sitting.

    Just wrapping the screen around the bottom edges of the side walls and setting the box on top of another piece of screen would be a good start. That’s likely to only make contact at a few points, and ideally the gaps between contact points should be just as small as the openings in the screen. Weighing it down will encourage better contact, but for really good performance you want some sort of springy contact arrangement — copper finger stock or foam wrapped in fine metal mesh are the usual approaches, but you’re not likely to find those at Home Depot.

  4. Next time try metal window-screen mesh. It is a fine mesh, intended to keep most insects out.

    It is usually in the same areas of the store as doors & windows, or door-sweeps and door insulation.

  5. When I play Skylanders on the WiiU, I like using the old Wii controllers, but as long as the WiiU can see it’s own controller, it refuses to connect to the older ones. Once I figured out what was going on, I stuck the original controller in the microwave and had no more troubles out of it. Of course, you do have to remember to take it out before heating up your coffee!

  6. For those of us *not* in the know on such things – if we go with a smaller-than-calculated mesh, is the cage *more* effective? And/or if we overlap (and offset) layers of mesh, will that also make the cage more effective?

    1. Finer mesh may help a bit, but that’s usually not the weak link if you’ve followed the “holes smaller than 1/10th of a wavelength” rule of thumb. Leakage normally happens where the layers of mesh don’t quite make electrical contact, even though they may be overlapping (either they’re only touching in a few places or oxides and stuff are preventing metal-to-metal contact). Large overlaps provide some capacitive coupling, which helps even when conductive contact is absent or spotty.

      Once you’ve established good contact around all the edges of your mesh and are looking to get still better shielding, then moving to a finer mesh or solid walls becomes a relevant consideration.

  7. Suppose you are using tin foil and grounding the cage, and your circuit under test is a resonant circuit. (In particular, I have an LC circuit and drive it with 100 watts of power.)

    The capacitance of the cage will affect the resonance of the circuit, yes? If the C in your LC system is 0.5 pF or so, the cage will add significantly to that and change the frequency of oscillation.

    Also, if you make the cage bigger, the area goes as the square of the radius, so a bigger box makes for a bigger parasitic capacitance.

    Is all that the right analysis?

    1. A larger cage does have greater area, but it allows you to get the surfaces further apart. If you’re keeping the device approximately in the middle of the box, capacitance between parts of your device and the box walls will be smaller for a larger box.

      Whether the capacitance between the box and the relevant bit of your circuit is any significant fraction of 0.5pF depends completely on what you’ve built and how. A device built following good RF design habits shouldn’t notice the box in the slightest unless you’ve set an exposed part of the device very close the metal (or if it has an antenna that you’re also putting in the box), because those habits are built around preventing that kind of uncontrolled parasitic interaction.

    2. > The capacitance of the cage will affect the resonance of the circuit, yes?

      But with equal on the “far side” & there being an infinite number of equal near and far sides, I thought the issue was moot.

  8. I’m guessing that grounding does NOT matter because the RF radiation already has an extremely easy path to ground (i.e. through the air right from the transmit antenna). Your shield wouldn’t really be giving the RF any better path to ground.

    1. Grounding at higher frequencies is a pain in the ass; grounding this cage with a good solid wire is about as effective as letting it float. wide braid would be good; like for example the shielding on coax cable. Strip the insulation off some coax; use it as a gasket around the removable side and get it tight and well connected; and bury the other end of the cable deep and long as you can.

      Beware absolutely goofy things happening in the presence of lightning.

    2. I think grounding may become more important if there are cables going out of the box. Though if the cables are shielded and shield is properly connected to box, the cage is effectively extended for the length of the cable.

  9. I did this to my house when building it, only partially intentionally. I grew up in an area where Adobe-style construction is very popular and somewhat mandatory in some communities. To get the stucco to adhere to the outside of the house, I used a very fine (1/16th inch) metal mesh with the borders folded over (So both pieces are sandwiching each other). For the windows, I used some that were specifically designed to block out UV and bright lighting by sandwiching a metallic plastic in between the glass layers (The plastic is very similar to what is used in anti-static bags). The windows also had a metal frame that I was able to use some conductive glue to attach to the mesh of the walls.

    The entry way was designed as an airlock / mud room type design to prevent cool air from escaping, but also causes the cage to remain unbroken even with one of the doors open. Both the front and back doors are designed like this. Really, the only holes in the cage are the galvanized air exhaust vents in the roof and where the power lines enter the house (Although I am working on combining a pair of construction generators to build a power conditioner / flywheel. So might enclose it into its own cage in the garage). I get my internet via fiber optics and elected not to have any other form of communication line installed (there really isn’t any point in this day and age). I have cell service via a micro-cell I have on my network.

    I originally built my house this way just because I noticed it was possible without too much effort and money, but now I’ve been getting into tinkering with RF and its really coming in handy.

    1. I use 325 stainless steel mesh screen cut to fit the inside of the NIDO powdered milk can lid to complete the metal lining of my milk can I’ve used for close range determinations when radio directional finding.

      I also made sure to use aluminum screen for the intakes and exhausts on my farm building that is steel lined though still have room to improve where more elaborate designs use, special I think, copper screen lined windows or maybe there are tints that are effective in more than the AM range I’ve researched.

      At higher frequencies in microwaves and higher… solid metal welded seams are required and special reflective materials don’t hurt either, i.e. say like certain oxides when you hit the IR range and lead in the ionizing range is more commonly known.

      1. At lower frequencies an anechoic chamber design is required for the audible ranges and since if not thick enough metal; the metal can be resonated to a certain extent to create other combinations, undertones and overtones depending on design, tension I guess or just basically, resonance of the structure considering overtones if solid metal versus a screen. Due to overtones of drum membranes causing more signals since are more than just typical other designs fundamental frequency harmonics.

        I’ve been informed that an intermodule shipping container buried underground a certain depth is a good idea. Couple with the above said and other thoughts. Figure how to baffle and screen the intakes and exhaust as well as couple signals in and out with monitoring for background scanning subtraction if required if you’re really serious is a must.

        1. Burying an intermodal container sounds great but is actually not a good idea. Cargo containers are built to support weight only on specific contact points (mainly corners) and can collapse from pressure loads on the midpoints. An example might be 4 feet of wet earth piled on the roof. Also, cargo containers shed rain but are not designed to be waterproof.

          1. Correct, I can add on a code footing, foundation, with walls and poured roof. With proper drainage considered for the geological conditions. Probably is going to be 8″ cinder block filled walls and thick poured perimeter in general using rebar. Yeah… burying a intermodule shipping container is a bad idea for sure without considering code requirements. Even the flooring is bad to be exposed to and that is something that should be replaced.

            Technically, the point of the intermodule shipping container is an easier to make commercial off the shelf was a welded sealed container is better than screen only Faraday Cage. I’ve even thought about spray foam lining maybe wouldn’t hurt outside the poured structure of which when pouring will need support beams inside for 30 days so the cement can cure enough. That would somewhat act like an outer Anechoic Chamber.

            Even better is finding geological formations that are better at shielding like I’m guessing higher iron deposit areas and other minerals kind of like Cheyenne Mountain and other deep mine/cave systems operations. I’ve wondered about real time actuator controls also on the springs used in between the footing and Faraday Cage to inverse lower frequency vibration that could get in… sort of like placing the whole system on a dampening table.

      2. I was thinking about a thought yesterday using conductive paint and forgot to post. A more cost effective ferro-organic or even salvaged ground, to a powder if wanting, metal paint might be an option in resin on cloth kind of like using cloth instead of fiberglass… especially if certain shapes were desired. Seems conductive polymers are an area of opportunity maybe also. Seems the ideal or effective material in silicone would be more insulating as well as flexible, mobile and I guess even wearable if on a layer of EMI/RFI shielding fabric.

    2. The comment section for this video is priceless. So many people are so scared and insecure today….makes me scared to be around them. Like I need to go buy a gun and put it on my wall and then build a metal box to sleep inside of.

  10. Watch this whole presentation regarding multi-layer (like multi rooms) Faraday cages noted at the Green Bank. This is a great Defcon 21 presentation that provides some insight (I thought this point was interesting also… like was hacked even audibly that to me associates observing audio correlated with RFI/EMI to study in more details):

  11. I don’t have one because I don’t need one. If I needed one I would figure out how to get or make one. Lots of hackers don’t have lots of things. Right no even if I needed one I would not have indoor space for it.

  12. So I wonder if it would be cheaper or easier to mass produce steel wire mesh plated with copper? You could use electrolysis. You can buy grounding rods like that at the hardware store, so it does work.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.