This is a super cool blog post. However I think it alludes to a much larger problem. For many products today, it’s very difficult or impossible for most people to verify the product works.

For a network switch, there are free and open source tools like iPerf to test and verify speeds are as advertised. For a faraday box, you have to go through all these steps (and knowledge learned) just to be able to test these. What about for vitamins, or pet food, or any other “durable” products that are supposed to last for X years?

In an age of snake oil salesmen, paid for reviewers, and fraudulent products on Amazon, there is a real opportunity for creating systems that enable individuals to verify products do what the sellers claim they do.

I really like Snowden's recommended Faraday cage for phones or other devices of similar size: two drink-shaker cups[1].

Drink shakers are cheap and widely available. The average hotel room in any medium/large hotel chain probably includes a shaker as a standard item as part of the minibar. The simple two-cup style (as seen in [1]) is preferred over the fancier "strainer top" style because you can press the two cups together; this should cause the inner cup to slightly cut into and/or deform the outer cup along the circle where they join together. The seal between the cups should work sort of like the knife-edge seals used in vacuum chambers.

A metal box or conductive bag is only a Faraday cage if it is fully closed/sealed. Any imperfection in the seal or hole[2] might allow the radio signal to leak out. Most improvised items (freezer, random metal box, etc) have poor seals. Making a high quality cage that actually block a modern phone can be done without much trouble, but the drink shaker method is the only method I know of that will do the job using widely available (free) or very cheap ($10-ish?) parts.

[1] https://ae01.alicdn.com/kf/HLB1yRyhXZTxK1Rjy0Fgq6yovpXaZ/Win...

[2] Holes of sufficient size. How big are the waves you are trying to block?

At my former job we had a rather large faraday cage for testing telco equipment. Two things surprised me: 1) It is more fiddly to ensure a room-sized faraday cage doesn't leak RF than I thought. It is really easy to make mistakes that compromises experiments. 2) There was a bath-towel sized piece of cloth that came with one of the expensive Rohde & Schwarz instruments that was astonishingly effective at blocking signal.

So I mainly used the room because that's where the magic blanket was - which was easier to use than ensuring the room was set up correctly.

> While the cookie tin performed much better than the mylar bags, and could provide a modicum of useful attenuation under some circumstances, it was not sufficient to provide meaningful assurance of signal isolation at any frequency. However, it was unique among the containers tested in providing tasty snacks during measurements

Back in the day working for Symbian I used to regularly have to run a load of automated Bluetooth tests. Unfortunately there were 200+ engineers in the close vicinity of my test setup and they all had smartphones and all had Bluetooth turned on, causing loads of the test to timeout. We were very pleased with ourselves when we thought of using a biscuit tin as a Faraday cage to try and improve things. Didn't work as well as we hoped, but got the tests running within their timeouts. I always thought it'd be interesting to try using an old microwave.

Best thing was that we got to eat all the biscuits first :)

I was hoping they would test chip bags, because it was a plot device in the most recent Terminator movie.

After searching, I found that Kaspersky has a blog post on the topic: https://usa.kaspersky.com/blog/terminator-dark-fate-chips/18...

It surprised me that chip bags were more effective in their test than cookie tins, but only when they double-bagged it

The pink bag shown is actually a static dissipative bag, not a static shielding bag, so would not be expected to work. Though the article mentions trying other bags so no doubt some of those where static shielding (metallic appearance) bags.

I've had a lot of fun in the past trying to shield Bluetooth devices for testing... metal filing cabinet? Practically no attenuation. Biscuit tin? Minimal attenuation like in the article.

Modern radio receivers are ridiculously sensitive.

> "Wrap it in tin foil" is perhaps the most common advice [...] Unfortunately the results were extremely inconsistent [...] 90 dB attenuation in one test would produce only 50 dB the next time [...] You would have no way to tell whether you've managed to fold and seal it adequately well

On the other hand, aluminium foil is incredibly commonly available, if not in a kitchen draw, it will be at the nearest shop. Although unreliable, the _potentially_ high attenuation the author shared is pretty good (we don't know his worst reading), better than all other improvised solutions tested, and similar to commercial products. I wonder if a more reliable construction method could be found.

I'm not sure what movie scene scenario I'm envisioning, but in the case you needed a Faraday cage and don't constantly carry one around, perhaps simply taking an entire kitchen roll of foil and wrapping the device into a giant unsightly ball of it would be a reliable enough process for an "emergency" (if not pocket sized). i.e rather than trying to make a neatly folded minimal version, just resort to sheer number of layers of material - unless RF doesn't care and even 1000 layers with tiny gaps is no good?

[edit] Similarly, I wonder how well (or poorly) common household appliances work as a Faraday cage, e.g a fridge, microwave-oven - From what the author described, it seems they would all have too many gaps, however they are also constructed from higher gauge metals... i'm particularly interested in a microwave-oven which is specifically designed to reflect and retain microwave frequency fields.

I wish it covered (so to speak) aluminum foil more thoroughly.

In particular, foil is cheap, so rolling up in several layers (3? 5? 10?) and twisting the ends cracker-style ought to be more reliable and practical than trying to do it with one layer and fiddly seams. A suggested minimum number of layers and number of twists would be useful.

I built something in college that needed UL testing for RF emissions, and visited a facility in Long Island with a TEMPEST room about the size of my house. Was very memorable.

I also read an article this year by a guy that made his house radio-wave proof, because, he claimed, he was very sensitive to radio waves. It was WAY more involved than I would have guessed - for the same reasons discussed in this article and in this HN discussion. Nails and screws are a problem. Windows are a big problem. But still - how cool to be cut off from the onslaught of EM radiation.

I actually have a little cookie box, which is lined with heavy aluminum foil, which I put in a steel lunch box.

Perhaps it sounds like overkill, but it’s the only way I seem to be able to fully isolate the devices.

That said, I’d probably just buy a bag... but I’ve been trying to keep it with what I have lying around the house.

Cheap science fiction book idea: imagine having to choose between living without your smartphone or living with every company and government agency willy-nilly spying on you. Oh, wait...

Please reduce the image size, they seem to be 3 Mb each. Some people (especially if their phone is in a pouch) won't be able to get to the interesting parts of your article!

That was really interesting. I can’t believe how poorly the metal tin box did compared to real faraday pouches.

If all phones were mandated to have a physically isolating means to disable their wireless chips, we could avoid this problem all together.

Where I live, all I have to do to block 4G is to stop standing in the far corner of the yard.

Interesting results. The article states that Faraday cages are no directional. My physics 101 understanding of them was otherwise. You show they work by integrating over the surface of the cage and saying that's equal to the charge within the surface. But you don't say it's equal to the charge outside. Indeed it couldn't possible, since that charge could be light-years away and steady state would never be achieved. Now that was for electrostatic which isn't really what we care about, if I remember correctly.

Wikipedia also states they are more effective at blocking incoming than outgoing signals. Can anyone clarify the situation?

If you are sufficiently concerned, why not mod the phone to add a physical battery disconnect? Would this not provide 100% protection?

That red plastic "mylar bag" isn't what should have been tested.

Red plastic == electrostatic dissipative. It is not good at conducting electricity -- it is still very good insulator. It just has ability to dissipate electric charge fast enough so that it will not accumulate in normal use.

What you have wanted to check is a clear bag with metalized layer (which is silver or gray looking). It should, per my understanding, act as much better faraday cage than an electrostatic dissipative bag. And I would be really interested in seeing that measurement.

This article was a fun read, and it really highlights how we trust the words on the tin to be accurate and not mistaken marketing gibberish.

However, the part about the cost ranging from $40 - $80 is simply not true. I know this, because I purchased for $20 yesterday a faraday box to hold my auto key fobs and my cell phone. [1] There are very many items for sale at this price point, fabric, pouches and boxes.

[1] https://smile.amazon.com/s?k=faraday+box

What's the use case for these products?

Even if the radio is blocked, how does that stop a device from collecting data anyway, and exfiltrating it the moment it finds a network? Won't it look for a network the moment you take it out of the bag? And don't you have to remove it from the bag to use it practically?

Ultimately, if your device is compromised or untrustworthy to the extent that it has to be policed this way, is it not safer to just... get rid of it?

I think they could have done a better job with the aluminum foil. It looks rather not well done. He could have made a pouch and folded the sides over a few times.

I made a pouch for my car's key fob with aluminum foil and metal tape. It isn't as pretty as the ones you can buy, but it wasn't $25 either and It has worked really well.

Glad to see these tested. I've been using the EDEC bags for probably a decade to have on hand for incident response/forensics in security consulting, but I never had the gear to test the bags, I just trusted their supplier. Good to know about millimeter wave / 5g though.

One thing that most people should use a Faraday pouch for (in my opinion) is storing your key fobs if your vehicle uses keyless ignition with a fob. This prevents a very common relay attack which is to stand with a laptop by someone's front door and relay the keyfob signal between the keys (somewhere inside the person's house) and the car. For people who live in cities where you park on the street (as I do) it's extra-convenient for thieves in that most cars will flash some lights to say the alarm is disabled so the thieves know which car they just unlocked.

I'd be curious if anyone finds it worthwhile to have a smartphone while being so paranoid that you have to hide it in a Faraday bag. I understand the problem. I just don't understand the practical use case: every time you use the phone, you have to take it out of the Faraday bag, and it can be used to spy on you or report information about you. Assuming you're keeping it in the Faraday bag all day, when and why would you find it useful and worth the risk to take it out? And with that level of use, wouldn't it be more practical to have a dumb phone or a laptop?

Interesting experiment however his experience with tinfoil and metal boxes doesn't match mine.

I placed my phone wrapped in a towel inside a metal box: as soon as I close the lid I can't ping the phone on local WiFi and can't reach it with call (it goes direct to voicemail as if the phone was switched off). The GPS doesn't record any position while in the box. I didn't test the Bluetooth though.

Same thing for my car keys inside a paper towel and aluminum foil: impossible to open the car even at touching distance.

I wonder if he wrapped the device before putting it in the box or aluminum.

> The average hotel room in any medium/large hotel chain probably includes a shaker as a standard item as part of the minibar.

Maybe the hotels you stay in, Bond, James Bond.

> The average hotel room in any medium/large hotel chain probably includes a shaker as a standard item as part of the minibar

I've stayed in dozens of hotels all over the world, and never once have I seen a cocktail shaker in the minibar?!

It is hard to believe this would perform better than a Lazzaroni tin.

And I don't recall ever finding one of these things in a hotel room, even in Europe.

Now would be a great time to sell a ton of mints that just happens to fit an iPhone max with a case inside.

“Security Mints - sweet & secure”.

Maybe a kickstarter?

If you're interested in simple cages similar in effectiveness to the EDEC Window Pouch (they drop off above 3GHz)... try a Microwave Oven. They are highly available, reasonably convenient and well designed, with pretty good quality control. You can even still see your phone inside has zero bars. I do, however, recommend unplugging the microwave (or even cutting the power cord) as a fire would likely result from cooking your hardware.

To answer [2] How big are the waves you're trying to block? - it's actually crack length not width that usually determines the effective re-radiating antenna length. They can be very narrow, but a long (a few cm) crack will allow one polarization through almost completely.

I've run this test with a GSM base station and confirmed the effectiveness. If you want to test communication with a device to a base station, but have many other local antennas interfering and want to fit many devices inside... it's a decent choice. The alternatives are less convenient and usually thousands of dollars to construct.

What's the physics behind being able to make a container light-tight with a couple of right angle corners yet radio waves seem to leak out like water?

I think the issue is then you have to wait until you are already inside the hotel to use those shakers, which as others have said don’t actually exist in most hotels, and at that point you have been tracked to your exact location. Then it would be trivial to surveil you with other methods without needing your phone.

At one company I worked at we had an 'out of coverage bucket'. It was a zinc plated steel bucket. Worked almost every time. Some dude had found at an arts and craft store a very small version of the same thing. It fit over phones perfectly. Basically put device on ground. Put bucket over it. If you had any gaps or any cables hanging out it was sort of iffy and you would get coverage sometimes.

Biscuit tins and aluminium foil have layers of insulation on them, the varnish of the tin and the Al2O3 oxide layer for the foil. It is true that the gap in each case is small, but when you look for 10 orders of magnitude in attenuation - a little leakage goes a long way. Silver has the unusual property in that both the metal and it's oxide are both conductors = true silver foil made of thin metallic silver will make a good wrap around Faraday bag. Even better would be true gold metal foil - note they often use gold plated shields in space uses, even though gold foil is like fly paper in space = contact welds readily, much more so than most other metals (many of which have also contact weld problems that have to be mitigated). Those Ramsey boxes are excellent, far better than many costlier, but crappier boxes, obviously made by someone who knows an ohm from a volt...

The simplest test you can make to verify that the microwave oven & biscuit tin can are good Faraday cages is to put your phone in one and try to call it - every time I did that in the past they failed to function as a Faraday cage.

Edit: clarification.

When I was in jail I had a cellie who imported drugs from China. They were always shipped inside defunct microwave ovens. I wondered if it was just because it looked like a harmless kitchen appliance or because the microwave provided shielding from an x-ray scan?

How many different biscuit tins did you try? :-)

Microwave ovens make for poor Faraday cages. The problem is that you only need 30-60dB to "keep the power in" while you probably want at least 100dB to "keep the signals out."

You can test this: put your phone in the microwave and try to call it. Over the years I have had about a 75% success rate.

Other household appliances tend to be even worse; any gaps in metallic seals effectively turn into slot antennas and let signals through. You need a conductive mesh gasket to stop this, and I generally don't see mesh gaskets on consumer appliances.

Your "make an al-foil ball" strategy is what I would resort to in a pinch (I have some nice Anritsu shield boxes, but those are cheating). Quantity is a quality of its own. My only addition would be making sure that the gaps tend to not line up. You want to make it difficult for the "slot antennas" modeling the gaps in each layer to couple to each other.

Microwave-ovens definitely don't leak below 2.4 GHz. Any frequency above that is a non-issue for microwave-oven operator safety. It also wouldn't be an issue for any communication device transmitting below 2.4 GHz. If you want to shut down 5 GHz communication then a microwave-oven would be leaky. 60 GHz: it shouldn't matter since it needs LoS anyway.

Also, remember to ground your faraday cages.

I'm not an expert on Faraday cages, would the effects stack if you wrapped your phone in aluminum foil, then put it in a cookie tin?