A shielded box protects small electronics from one specific thing. It does nothing for the grid, your wiring or anything with a wire attached to it.
What it really protects →Learning how to make a Faraday cage takes about twenty minutes and under forty dollars. Almost every homemade one fails at the same single point, and it is not the material, the thickness or the grounding. It is the lid. Get that right and a galvanised trash can from the hardware store performs respectably; get it wrong and you have an expensive bin.
What You Need
- A galvanised steel trash can with a tight metal lid, around $25-35 at the time of writing. Bare galvanised steel, not painted or powder-coated, because coatings are insulators.
- Conductive foil tape, aluminium or copper, ideally with conductive adhesive. Roughly $10-15 a roll. Ordinary duct tape is useless here; it is not a conductor.
- Cardboard, closed-cell foam or bubble wrap to line the interior. Free.
- Heavy-duty aluminium foil and plastic bags for wrapping each item. A few dollars.
- A cheap AM radio for testing. If you do not own one, buy one. It is also the thing you most want inside the can.
Notice what is absent: grounding rods, wire braid, special alloys. A closed conductive shell shields its interior whether or not it is connected to earth, and steel that is thick enough to hold a shape is far thicker than the depth a fast pulse penetrates. Nothing about this build is exotic.
The Build, Step by Step
- 1. Check the metal is bare. Run a multimeter on continuity from the can body to the inside of the lid rim while they are seated together. If you do not get continuity, sand the mating surfaces down to bright metal.
- 2. Line the inside completely. Cardboard on the base and around the walls, taped in place. Nothing you store should ever touch the metal.
- 3. Wrap each item individually. Plastic bag first as an insulator, then two layers of heavy foil, sealed with foil tape so there are no open edges. This gives you a second independent shield inside the first.
- 4. Load it, leaving room at the top so the lid seats fully. A can packed to the brim cannot close properly, and a lid that does not close is not a cage.
- 5. Seal the lid seam. This is the whole build. See below.
- 6. Test it before you trust it. Also below, and it takes five minutes.
The Lid Seam Is the Whole Build
Here is the physics, briefly. Any gap in a conductive enclosure behaves as a slot antenna, and how much energy passes through it depends on the gap’s longest dimension compared with the wavelength. A slot approaching half a wavelength radiates almost freely. At 1 GHz, half a wavelength is about 15 centimetres.
A lid resting on a can leaves a gap perhaps two millimetres wide running the entire circumference, maybe 90 centimetres long. It is the length that matters, not the width. That is a large slot antenna with your radio sitting inside it, and it is why “I put it in a metal can” so often produces disappointing test results.
The fix is to run conductive foil tape all the way around the joint, overlapping generously onto both the lid and the body, pressed down hard so the adhesive makes contact through to metal. Do it in one continuous pass with a good overlap where you finish, not in short pieces with gaps between them. The whole seam, every time you close it. It is tedious, it is the only step that really matters, and it is why some people keep a foil-lined cardboard box inside the can and treat the can as a rough outer layer, so the fiddly sealing happens on the box instead.
Proving It Works
Never assume. Two tests, both fast:
The radio test. Tune a cheap receiver to the strongest AM station you can find, turn the volume up, put it inside on the cardboard liner and seal the lid. You should get silence or a faint hiss. AM broadcast is low frequency, which is the hardest case for any shield, so passing this is a strong result. Repeat on FM, then try a shortwave band if the radio has one.
The phone test. Wrap a phone in cloth so it does not touch anything, seal it in, and call it from another phone. A pass is straight to voicemail with no ring at all. Also start Bluetooth audio before sealing, which checks 2.4 GHz specifically. This is the same protocol we use on commercial pouches in faraday bags: which work and how to test, and it works identically here.
If the radio still plays, the seam is your problem in roughly nine cases out of ten. Retape it properly and test again.
Three Builds, Compared
| Build | Rough cost | Effort | Best for |
|---|---|---|---|
| Galvanised can, taped seam | $35-50 | 20 min | Bulk storage you rarely open |
| Foil-wrapped box inside a can | $40-55 | 40 min | Two layers, easier resealing |
| Steel ammo can, gasket bridged | $25-45 | 30 min | Small items, stackable, durable |
| Foil-wrapped cardboard box alone | Under $10 | 15 min | Spares, or testing the idea first |
The ammo can carries a trap worth naming. The rubber gasket that makes it watertight is an insulator, so the lid never makes electrical contact with the body no matter how hard you clamp it. Either strip the gasket out or bridge it with copper tape or short lengths of braid at four or five points around the rim, then test.
Mistakes People Make
- Running a cable in. Even a charging lead through a small hole carries energy straight past the shield and defeats the entire enclosure. Nothing goes in or out. Nothing.
- Using a painted or plastic-coated can. The coating stops the lid making contact. A plastic bin with foil taped inside works, but only if the foil layer itself is continuous and sealed.
- Letting contents touch the wall. Insulate everything. It costs nothing and removes an easy coupling path.
- Worrying about grounding. It is not required and it will not rescue a leaky seam. Spend the effort on the tape.
- Storing batteries inside for years. Alkaline cells leak and will destroy the equipment they were meant to power. Keep cells separate and rotate them.
What I’d Put In It
Cheap, hard-to-replace spares that are useless in singles: an AM/FM/shortwave receiver, a handheld transceiver if you hold a licence, a spare solar charge controller, a small inverter board, a USB power bank, and an old phone loaded with offline maps. Perhaps a hundred dollars of gear inside forty dollars of can.
Then close it and go back to the ordinary problems, which are far more likely to happen and much better documented in what breaks first when the grid goes down. A sealed can full of spares is a sensible cheap hedge. It is not a substitute for stored water, a carbon monoxide detector or a way to stay warm, and if you have the can but not those, you built the wrong thing first.
Amazon affiliate links follow. As an Amazon Associate I earn from qualifying purchases, at no extra cost to you. Details
Aluminium foil tape, 2 in x 65 ft, on Amazon →
AM/FM/shortwave emergency radio on Amazon →
Frequently Asked Questions
Does a metal trash can work as a Faraday cage?
Yes, provided the metal is bare and the lid seam is sealed with conductive tape every time you close it. Untaped, the gap around the rim behaves as a long slot antenna and performance drops sharply. The can is fine; the lid is the problem.
How many layers of aluminium foil do I need?
Two or three, wrapped so there are no open edges and sealed with foil tape. Household foil is roughly 16 micrometres thick, which is already several skin depths at the frequencies that matter, so continuity beats thickness every time.
Do I have to ground it?
No. The field inside a closed conductive shell is cancelled regardless of whether the shell is earthed. Grounding is worth doing on a large permanent installation for other reasons, but it makes no difference to the shielding of a trash can on a garage floor.
Can I use a microwave oven instead?
Poorly. It is built to contain one frequency around 2.45 GHz, and the door seal, vents and power cord penetration are not designed for broadband shielding. Test one with an AM radio and you will usually be disappointed.
Some of the links above are Amazon affiliate links. As an Amazon Associate I earn from qualifying purchases, at no extra cost to you. Details in our affiliate disclosure.