Thread by @mattblaze

Content warning: Nerdy thread ahead.

I redid my informal tests of various cellphone-sized Faraday pouches, to measure the amount of attenuation they actually provide. Tl;dr: the expensive commercial ones generally work well. Cheap makeshift ones generally don’t.

First, what’s a Faraday pouch and why would you need one? A Faraday cage severely attenuates radio signals going in or out. It can be used to assure that an untrustworthy device (like a cellphone) isn’t transmitting or receiving signals when it shouldn’t be. Paranoid? Yes.

A Faraday cage is simple in principle: solid conducive container that completely enclosed the signal source. But actually constructing one that works well can be challenging. Any opening can create a junction that acts as an RF feed. There are pricey (~$50) products for this.

So the question is, how well do they work, and do cheap alternatives work adequately well? To test this, I set up some equipment to reasonably accurately measure signal attenuation at the frequencies used by cellphones (low GHz).

My test setup consisted of a small signal generator (ERASynth Micro) with an antenna, and a measurement receiver (R&S PR100) hooked up to an antenna inside an RF test chamber (Ramsey STE3000B). The signal generator is just small enough to fit inside a cellphone pouch.

I generated a 10dBm (10mW) signal at 1, 2, 3, 4, 5 and 6 GHz, and positioned the feed point of the generator 30cm from the antenna inside the test chamber. (The purpose of the isolated test chamber was to reduce external noise that would affect the measurements).

As a representative sample, I’m giving the results for 4GHz here; results other frequencies were roughly similar.

First, the baseline, with just the generator at 10dBm, not in a bag. We measured 81.4 dBuV. So all our tests will be relative to that.

First up under test is on the cheap end: a recycled padded electrostatic bag that something was shipped to me in.

We measured 74dBuV, which is just over 6dB attenuation. Not good! You get what you pay for, I guess.

Next, I wrapped the generator in about 4 layers of heavy duty aluminum foil.

We measured 31.5 dBuV, which represents 50dB attenuation. Not half bad! But, alas, probably not good enough for assurance that no one is communicating with your device.

Now on to the commercial products. First was the EDEC brand window cellphone pouch. The generator only barely fit, which I think interfered with my ability to close it properly.

2.1 dBuV, which is about 79dB attenuation. This is excellent, especially considering the poor seal.

Finally, the “Mission Darkness” brand window pouch, which is slightly larger than the EDEC pouch and fit the generator comfortably well.

-17.2dBuV, which is a whopping 98dB attenuation. That’s an extremely impressive result.

The bottom line: mylar bags are useless. Aluminum foil is better than nothing, but not great. The Mission Darkness bag was great, and I’d expect similar results from the EDEC bags if it fits properly. But the fit and closure affects performance a LOT.

And measurements matter!

A couple notes on the equipment: The antennas used for the receiver (connected inside the chamber) and on the generator were just cheapo Bluetooth antennas with SMA connectors that I had around. Not particularly resonant, but efficiency didn’t matter for our purposes.

The test chamber is a Ramsey STE3000B, which is an RF-isolated “glove box” that lets you manipulate controls (though I didn’t use the gloves for this). It has passthrough jacks for sma, N, BNC, UHF and TNC connectors, as well as AC power. Really useful piece of bench equipment.

It provides about 50-60 dB isolation from the outside world, which is more than enough for most experiments and measurements/

The measurements were taken with a Rohde & Schwarz PR100, which is like a combination measurement receiver and spectrum analyzer. I really like it - it gives you a really good picture of what’s going on the spectrum.

Finally, the signal generator was an ERASynth Micro, which I got from Crowd Supply for about $300 a while back. It puts out a clean signal up to 6GHz at 18dBm, and is VERY handy for doing experiments like this.

Oh, I should also note that I think all these measurements should be regarded as accurate only to within +/- 10dB or so. The size and shape of the bags inside the chamber likely affected the signal polarization and reflections going on inside the box.

Doing this more accurately would require a larger RF anechoic chamber, which I don’t have easy access to. But this was good enough for a basic comparison.

Final note: if it wasn’t already clear, making accurate measurements of the attenuation provided by a Faraday bag involves expensive gear and fussy technique. But you probably don’t need accurate measurements for most purposes.

A quick and pretty reliable “go/no go test” can be done with an Apple AirTag and iPhone: drop the AirTag in the bag under test, and see if the phone can locate it and activate its alarm (beware of caching in the FindMy app).

This won’t tell you the exact attenuation level, of course, but it will tell you of the attenuation is sufficient for practical assurance. It can also detect whether an otherwise good bag has been damaged and compromised.

I suppose now I need to be ready for the Aluminum Foil Truthers.

And for completeness: a metal cookie tin.

53.5 dBuV, which is about 28 dB attenuation. Disappointing performance, but provided tasty snacks during testing.