I just put my phone in the refrigerator and called it -- works perfectly. Then I tried the same thing with a microwave oven (following pilsetnieks's suggestions) -- didn't even lose one bar on the signal indicator. It did kill the Wi-Fi reception, so clearly the oven is blocking the 2.4 GHz frequency as it should.
Edit: Also tried a cookie can. (And anti-static bags -- see below.) Faraday cage my ass.
Clarification: I use a Nokia. Maybe the technique works only on iPhones? I've heard they have some reception issues.
Yeah but "can't hear you" is not the same as "sophisticated signal processing techniques will be thwarted;" putting the phone in another room might not be good enough. Leaving the phones in the car might work, though I suspected that the risk of the phones being removed from the car while everyone is inside would be pretty high in Hong Kong.
All in all, I think the fridge's insulation probably acts as a good sound barrier, and the hum from the compressor probably adds quite a bit of noise. Probably a smart thing for someone in Snowden's position to demand.
Or being removed/replaced and outfitted with an independent bug complete with radio and power supply. Perhaps it exists as part of a covert, replacement spy battery that powers the phone!
The "refrigerator" had a false back. When the phones were placed inside and attention directed elsewhere, NSA agents opened the back of the refrigerator and installed bugging devices (complete with reserve power cells) in the phones. The goal has always been to identify and entrap counter-intelligence targets.
Snowden is a false flag exercise, which would be obvious to anyone who read Catch-22.
You can never be paranoid enough, because they have a million people with TS clearance thinking of ways to get into your heads. That's just in the US.
What is safer in this scenario? Sealed fridge or coffee table in another room?
Car wise, well, that could be impracticable:
1) does one have a car in HK? Or did one fly in and hire a cab?
2) Has one just walked up a tower block to get to Snowden? All go down, and come up again? That would be annoying.
3) If I take all the phones and put them in a box, I know I have the phones. I have control.
In a normal apartment, I reckon a fridge is possibly the best place to put turned off phones to isolate them as best you can at short notice with out super spy resources. Not prefect, but looking round my place, I don't see something better than a fridge for the purpose.
And a freezer acts as a surefire way to not catch some conversation, if you end up walking around the apartment or into the room where the cell phone was tucked away.
Had visitors not brought cell phones in the first place, it wouldn't be an issue.
A solid box of metal is going to be a pretty effective Faraday cage regardless of what (sane^) frequency you're using. Microwaves on the other hand are specifically designed to be Faraday cages, so they're tailored specifically to the frequency they're blocking.
(How exactly they do this I'm not quite sure, given that the frequency of a phone and the frequency of phones is quite similar.)
> the frequency of a phone and the frequency of phones is quite similar
(I'm assuming you meant frequency of a phone and microwave oven.) That's not quite accurate. Cell phones use many frequency bands [1]. I only see one frequency on the list (2500 MHz) that is close to the frequency used by consumer microwave ovens (2.45 GHz) [2]. I don't think 2100 MHz is close enough to be attenuated by a microwave oven, but of course I'm just speculating because it depends on the design of the oven's shielding.
Don't forget that phones switch to a frequency on which they have signal. If you don't have 4G coverage, your phone will use 3G, etc. These operate at different frequencies. So attenuating one or two of the frequencies on the list is probably not good enough.
> I don't think 2100 MHz is close enough to be attenuated by a microwave oven, but of course I'm just speculating because it depends on the design of the oven's shielding.
I assumed that the design was that of a Faraday cage, which blocks the frequency it's designed for and all lower frequencies. anotherhue noted however that it's a tuned RF choke, which will only block a certain frequency.
For reference, I'm in Australia with a 3G phone on Optus, so the frequencies here are a little less scattered [1].
Microwaves aren't Faraday cages as it's too difficult to manufacture a perfect seal, instead the size of the holes in the door act as a tuned RF choke.
People sometimes like to watch the food being cooked, for one. At KW levels of power a few dB is still enough to do harm. Lots of nasty reports in the medical literature of this type of exposure, typically unlucky repairmen.
The eye has very poor heat dissipation mechanisms, and they showed in the 70's the RF exposure could cause cataracts in rabbits. (I think dogs were exposed too -- it was a new field, so they didn't know what would happen)
One of my old professors developed radiation-hardened orbital sensors for NASA. From what I remember of his lectures, I think a refrigerator is probably a terrible Faraday cage.
Here are some excerpts from a nifty Analog Devices tutorial about RF shielding[1]. This is probably the most relevant part:
The longest dimension (not the total area) of an opening is used
to evaluate the ability of external fields to enter the enclosure,
because the openings behave as slot antennas.
Remember those big rubber seals on the doors of most refrigerators?
It then goes on to describe an equation for approximating the shielding effectiveness at a given frequency:
Let λ = wavelength of the interference
Let L = maximum dimension of the opening
Shielding Effectiveness (dB) = 20 * log10(λ/2L)
And here's a nice rule of thumb:
A rule-of-thumb is to keep the longest dimension less than 1/20
wavelength of the interference signal, as this provides 20 dB
shielding effectiveness.
There's other stuff in there, such as how to properly shield cables entering or exiting the box. (Unshielded wires penetrating the box can also act as antennas.)
An anti-static bag (even the metallized type) probably won't be the best idea. Faraday cages depend on the conductivity of the outer shield to exclude electric fields from the inside, but anti-static bags aren't very conductive (100k-10G ohm square resistivity[1]).
You might even be better off wrapping the phone in aluminium foil, but the problem is then going to be ensuring that all seams in the "cage" are actually conductive.
I've recently tried the static bag using a samsung tablet - wifi and bluetooth both showed no attenuation with two different bags. A "farraday cloth" that was suggested to me is priced at about $24 a linear foot (8'x1') so I wouldn't expect these cheaper solutions to be very effective. I would expect a microwave oven cavity to effectively reduce signals in 2.4ghz, though not signals in the lower gsm and related bands.
I just repeated this, while streaming (with no buffer) over wifi. The wifi connection dropped out almost instantly, and the phone signal dropped out after about a minute. My guess on the phone signal is that it assumed it was still connected momentarily while trying to reconnect. I know from experience this happens when between towers, and appears to be a sort of UI smoothing to avoid showing "no signal" frequently.
Also, from experiments I've conducted, a cookie can will work for certain frequencies; for example, the RF transmitters in anti-theft devices. I suspect the material is too thin to stop phone transmission though.
Indeed. Might be something to do with the differing frequencies in different countries; I'm in Australia, so my 3G would either have been on the 900MHz or 2.1GHz bands (with GSM on 900MHz/1.8GHz).
A couple of years back a friend and I tried the same thing with my stainless steel colored LG fridge. It totally blocked the signals on our phones. He had a blackberry and I had an el cheapo LG smartphone.
A fridge will act as a fantastic acoustic shield but probably not a functioning Faraday cage. I think this blog post has taken a few artistic liberties in suggesting its use was as a Faraday cage, especially as the article she links to doesn't mention the fridge's purpose as a Faraday cage.
Assuming this story isn't stretched and exaggerated (I have a hard time believing it isn't) and the fridge was on, this could be a fast way to trigger condensation to form inside the phone when the device is removed back into a warmer environment, depending on the ambient humidity. I just wouldn't dream of putting my phone in a cold fridge for anything other than the briefest time for this reason.
Maybe phones are not so sensitive to condensation. At least here in Finland phones seem to work fine, despite outside temperatures being similar or below to what is found in a fridge most of the year. People move phones in and out of buildings all the time with large temperature differences between. I don't think I have heard anyone breaking a phone because of condensation here.
Why not on? Wouldn't the compressor improve the situation by adding background noise? Also if condensation is a concern you can put the phones in a plastic bag.
When a fridge is on it usually means it's cold! If it's off (and has been for a while) it could be expected to be around room temperature, eliminating the issue of condensation.
The compressor isn't on all the time. It turns on intermittently when the temperature inside the fridge exceeds a certain threshold. It would be unlikely of much use most of the time.
Putting the phones in a bag does nothing to solve the issue of condensation. What happens is the phones would cool down substantially in the fridge. When they're removed, the cold surfaces (inside and out) would cause moisture in the warmer air to condense into water droplets, possibly causing issues with the battery and the other circuitry. Hong Kong is quite a humid place and it appears that he was at an acquaintance's home at the time of the frigid phone incident. It's unclear how well air conditioned this acquaintance's home was, which could lead to some especially nasty condensation issues. It's also unlikely this acquaintance's fridge would have been brought up to room temperature just for the sole purpose of storing phones for a couple of hours. This is why I'm inclined to believe the anecdote is BS.
"The compressor isn't on all the time. It turns on intermittently when the temperature inside the fridge exceeds a certain threshold."
Modern, high-efficiency fridges have compressors that operate continuously. It's more efficient than cycling. My fridge draws about 60 watts continuously.
It's still thermostatically controlled, of course, so if the doors are left open, it will kick up higher.
I'm not exactly a fridge nerd but I had no idea such fridges existed and I'm somewhat curious as to why it's more efficient than cycling. Perhaps it's the ease of which to maintain a constant temperature, rather than a fluctuating one?
I was introduced to refrigeration nerdery by talking with a friend who designs the cryocoolers for JWST (http://www.jwst.nasa.gov/cryocooler.html) who nodded sagely and said "more efficient that way" when I mentioned my new fridge did not cycle on/off. Now that's a fridge nerd!
The problem with cycling is that powering up is the least efficient part of the cycle. So a fridge (or an air conditioner for that matter) that cycles frequently is spending more time running in its least efficient mode.
I'm not a fridge nerd either, but my understanding is that adding thermal mass to the fridge will reduce the heating spikes from things like opening the door which will reduce the frequency at which it cycles up. So you can make your fridge more efficient if you have the space to put big jugs of water in the fridge and big blocks of solid ice in the freezer.
You might only save a buck that way, but if the space isn't being used for anything else you have nothing to lose.
Slightly tangential story. My first assignment on my first job was building a knockoff TEMPEST enclosure (http://www.jammed.com/~jwa/tempest.html) to isolate spectrum measurement equipment. The original use of the enclosures was (somewhat ironically given the article) the NSA's paranoia that foreign intelligence would spy on government computers by picking up their EM radiation.
So if you're worried about the government spying on your computers via RF and want something more convenient than a fridge, drop me a line...
A Faraday box might be more fun for higher trust environments. Starting your staff meeting? Family dinner? Watching a movie? Everyone puts their phone in the Faraday box and they get it when it's over. Similar to "The Phone Stack" game, but enforced. A nicely designed box in bright colors with a padded, felt-lined interior with separators (or even chargers).
If you leave your phone on and put it in a faraday cage, the battery will drain at the maximum rate as the phone goes into "find a tower" mode which is basically to broadcast at maximum power over and over trying to make contact with any towers that might be within its maximum range.
I used to develop nintendo DS games and once received a very memorable bug report - something to the effect of "got error [error message] when playing the game in a microwave".
We weren't handling lost wifi connections gracefully so the user would get the wrong error message if their wifi dropped. The tester was simulating this by placing the DS in a microwave.
Someone once came to us and asked us to design a wireless thermometer for a freezer. We told them it wouldn't work, because the freezer would act like a Faraday cage. Then they showed us a similar device that worked just fine. Embarrassing.
We test a production device here that makes a LOT of noise, and we put it inside a mini-fridge to dampen the noise so it doesn't drive the testers crazy. Works quite well. I doubt you'd pick up any outside conversation if you were trying to eavesdrop (especially if the freezer was humming).
Shouldn't a microwave oven be the perfect household Faraday cage? It certainly has the shielding to contain 2.4 GHz waves, so the 800, 900, 1800, 1900 MHz cannot be that far off.
Good question. The mesh size of the Faraday cage should be less than 1/2 the wave length of the RF radiation. At 2.4 GHz, the wavelength is ~125mm, and the mesh size of your typical microwave is a few mm. With the lower frequency of the cellular bands, you'd expect the microwave to work pretty well.
Yeah, that's not going to work as a Faraday Cage. I guess you'd have to put the cellphone in the microwave, then put the microwave in the refrigerator. That's going to look weird. :)
I used to do a bunch of micro-sites -- small websites targeted around specific topics. One I did a few years back was http://gps-cell-phone-tracking.net/
I really learned a lot setting that site up, and it was one of the reasons I became such a privacy freak. Our cell phones are basically tracking devices the government monitors which we happily adorn to our bodies each day. Weird.
To get to my point, one of the questions on the site was whether your phone could be tracked, either location or having the audio monitored, without your knowing it. Location is easy -- yes. Audio was a bit weirder. Obviously you'd have to install an app, and who would want to do that?
Then I learned that cell providers will "help" governments by putting tracking code inside the auto-update of the phone. They can even push it at night and have it install without your having any knowledge at all.
As far as I know, pulling the battery still kills things, but then somebody said most phones have a dual battery to keep the memory refreshed. Somebody even said something about the possibility of "illuminating" phones from a distance and picking up audio that way. Of course, "somebody said" doesn't count for much, and with all the secrecy in place good luck figuring out what is what.
Fun stuff. Since the micro-site business didn't take off, maybe there's a future in selling tin-foil hats?
I won't speculate on eavesdropping since I haven't tested this, but what I hear when I heard his method is more of the security risk that he could possibly be pinpointed. A red flag would go up for me, hearing multiple people (the lawyers) clustering to one location.
Smartphones are a beacon of constant info...GPS and signal pinging can get your coordinates (at least general area), and Wifi can triangulate the position [0]. Cameras and microphones are just icing and still require a device to be compromised, whereas signals, require carrier network access...which is what PRISM revealed [1].
Verizon uses an 700MHZ and 1700/2100MHZ signal on their 4G network which will give you incredible penetration. People brag about how you can be in a basement of an office building (lots more concrete and steel than a refrigerator) and still get great reception.
If they were using a carrier in Hong Kong, most of the carriers use a higher spectrum, usually around the 1800-2100MHZ range. While it doesn't give you great penetration in buildings like the lower frequencies Verizon use, I'm guessing the overall density of the towers is significant which would help this issue.
If he said they want into the basement of a building THEN put the phones in the refrigerator, it would be a little more believable. Just putting it in the refrigerator probably won't do a lot to block the signal.
Its not the mass of concrete and steel, but the faraday cage effect of a continuous closed loop of conductive material.
Though aas has been pointed out, a fridge probably isn't electrically continuous at the door gasket. Depending on the construction, you could replace the rubber gasket witha rubber core metal mesh gasket.
I used to work as a phone tester and sometimes had to test things while the phone was in a unreceivable location. I tried many things, like a safe, fridge, cookie jar etc, but the only thing that worked was wrapping the phone in 2 or 3 layers of tin foil.
Wrap the phones in tinfoil, put the wrapped phones inside the stainless-steel martini shaker (see other thread), put the martini shaker inside a microwave, put the microwave inside a refrigerator, and put the refrigerator in the bathtub.
Now turn everything on "high", including the shower ("hot & steamy"), and run.
Snowden's guests will have done that a while ago of course...
We need to crank our paranoic creativity up to 11, people.
The fridge itself had a false back. Snowden's associates in SPECTRE/Wikileaks then used their covert physical access to the phones, for the duration of the dinner, to copy the phones' contents and install new malware.
Now, they know what the lawyers know -- including whether any are double/triple/etc-agents.
Jeez, I missed this when I posted my reply. It looks as if we're the only two paranoid enough to see the danger and stupid enough to point it out and draw attention to ourselves.
Actually I thought the article put him in a better light than most of the other pieces about him. It makes him out to be more in tune with the ways the gov could track him. If they wanted to smear him then NYT could just come out with a short piece saying "Snowden made his Lawyers put their Phones in the Fridge", and then just spend the rest of the article speculating about his growing paranoia and questioning the decision.
Instead they explained the logic behind the seemingly unreasonable request. How is that being malicious?
I'm from the UK. We have tabloids. They will take a fact like (made up example) Snowden wanted all phones turned off, and simply make up the rest. 1) to demonize the subject, and 2) to put a more sexy sounding story in the paper. So, my natural default is to hope to believe the simplest "fact" in the story, if that.
Maybe you have faith in your newspapers, and to be totally honest I don't actually know how Americans feel about their news papers. But you do have Murdoch floating around, so Americans are being manipulated somewhere. But I sure as hell dont take them at their word. For me, newspapers are as accurate as "the internet".
But to me, that read like a slightly more sophisticated that a UK tabloid smear job.
I do how ever accept that my UK perspective on the press might well be very different to a US one.
Comments on this story illustrate the difference between Hacker News and other news sites. Most people say "That's cool" or maybe share links related to the research. People on HN said, "Okay, now that all our phones are locked in the freezer, let's compare data." :)
The article linked to above, plus several other sources I've come across suggest you are incorrect. I don't pretend to understand it, but there is some technique that makes your phone into a remote listening device even when off. The article above says "[US District Judge Lewis] Kaplan's opinion said that the eavesdropping technique 'functioned whether the phone was powered on or off.'"
Also -- that's the whole point -- otherwise Snowden would just say "Please turn your phones off".
> The article linked to above, plus several other sources I've come across suggest you are incorrect.
They are wrong. Which is not unusual, news outlets get technical information wrong all the bloody time. Hell, they get things wrong more often than they get them right.
> I don't pretend to understand it, but there is some technique that makes your phone into a remote listening device even when off.
Nope.
> Also -- that's the whole point -- otherwise Snowden would just say "Please turn your phones off".
Snowden is a dramatic source, not an accurate one. I'm glad he blew the whistle because it brings the NSA into the media's focus, but technical accuracy is clearly not his strong suit.
This advice is wrong. Try turning off an older feature phone's regular operating system, and then plugging it in to charge. Many of them will immediately come alive with a "charging" indicator on the screen, even when supposedly powered off. This pre-boot mode is perfectly capable of turning on the mic and recording what's going on.
Combine this kind of thing with a database of phone remote expoits and it's easy for the NSA to "upgrade" your phone's software and turn it into a bug. This is what Snowden means when he talks about "poor endpoint security." It doesn't matter how secure your fancy encryption applications are when the phone itself can easily be hacked.
Sometimes they don't even have to remotely push software. There was a recent scandal about Carrier IQ, a system on smartphones (Android players and iPhone) that by default was recording every tap, every virtual key pressed, every incoming/outgoing SMS, and more. Anyone with the device connected to a PC could take the forensic log off the phone and see basically everything, even in the past. With per IP logging and billing on mobile plans, as Snowden's Verizon leaks showed, the NSA can just have your phone upload that log anytime they like, without a trace on your bill or data usage.
> This advice is wrong. Try turning off an older feature phone's regular operating system, and then plugging it in to charge. Many of them will immediately come alive with a "charging" indicator on the screen, even when supposedly powered off. This pre-boot mode is perfectly capable of turning on the mic and recording what's going on.
Smartphones do the same thing. But that's triggered by a hardware interrupt (you physically connecting power to an input). You CAN NOT do that remotely, as there must be power on the receiver to receive the signal in the first place.
Or here, let's simplify it. Power off your cell phone, then have someone call it. Does it ring? No, of course it doesn't. Your argument is that by changing the caller from your friend to the NSA that somehow the phone would now be able to react to that signal. Which is utterly idiotic.
I'd love to see a definitive source one way or the other... But your point seems to be "all sources are wrong" so I'm not sure what to do with that. I certainly have no reason to believe some guy on hackernews (i.e., you) more than the findings of a federal judge and his staff. So, at this point, I have something like "accepted opinion" and "doubt" -- i.e., unless you can cite or explain something.
In this case, the ideal source would explain why this is so frequently misreported, not simply assert that it is always misreported. Also, it would explain why it was an impossibility.
Definitive source is easy as shit - power off your phone, then see if it's drawing any power from the battery. If it's not, then it can't be remotely activated. Period, end of story. That's just flat out not possible. If we had the technology to receive wireless signals without using any power our cell phones would have batteries that last for weeks instead of a day.
Again, a $20 multimeter wouldn't measure shit with a draw that small. And there are devices which use GSM radios for communication and last half a year on a 9V battery, for remote monitoring stations and such. So a virus implanted in a bootloader could very easily keep everything off and stay ready to spy. Your inability to accept this as a technical possibility is very childish.
There's a thing called a bootloader , and if something is implanted there the entire phone can be off, except for the radio "checking in" every few minutes or so. And you are very,very naive if you think you can measure a 0.1A draw using a $20 multimeter, good luck with that.
If you want something NSA scale, you can do better than that. I used to use Copper-Beryllium mesh and gaskets (http://www.sealingdevices.com/metal-emi-gaskets-and-becu-ber...). Choose mesh size based on the frequencies you expect to deal with (and harmonics).
If you want to keep your computer in there you also have to be careful designing the ports for wires coming in/out.
Every wire is also an antenna that is attenuated to the appropriate frequencies related to the length of the wire.
A 2 meter cable also receives on 144 MHz any spurious signals (amateur radio band). If that was an EMP, it would send a voltage spike up the port and blow out whatever it hit.
I bought some EM radiation blocking cloth from LessEMF and stitched it to the inside pocket of my jacket which I rarely use. If I'm feeling paranoid (or if I'm not arsed to turn on airplane mode to save battery) I just put it in that pocket instead of the normal one and zero communications can get in (especially my mother)
curious. Do you notice decreased battery life if you do this? I would think you would chew through battery like a mofo if you block carrier, but leave the phone on. It's gonna try hard to hit a tower non-stop, I expect.
My initial reaction was that the phones-in-the-fridge rule was meant to muffle sounds. Can anyone comment on whether or not a phone in a fridge can pick up sounds from the room the fridge is in?
Isn't just wrapping the phone in aluminium foil (as many layers as needed to make you feel comfortable) enough to act as a faraday cage? And leaves the cellphone still portable, you can unwrap if/when needed. And then bury it in a pile of clothes in another room (as the article suggests somewhere) for the soundproofing. The refrigerator I feel serves neither of these purposes well (as mentioned in this discussion), and ruins everything when it's time to make a sandwich.
I just tested this, aluminum foil does seem to block signals coming to my phone. At least, the wrapped cell phone doesn't ring when I call it with my house phone. The wifi signals may still be getting through, I don't know.
Why would it confirm it? "Correlation does not imply causation".
He may be pulled in by the same rumours as everyone else and being overly cautious. Just because he does this doesn't indicate he has confirmed or denied the existence of a slave microphone bug.
Just that he believes strongly enough in the possibility to be cautious.
The actual cases I've seen (some linked on this page) are of specifically targeted individuals having their phone receive a command to turn into a bugged phone. I don't think all phones are bugged by default. This fits with Snowden wanting people around him to sound-proof their phones as they might be targeted.
You would think instead of stashing them in the fridge, he would ask visitors to not carry a GPS receiver right into his hideout. We complain about the gov possibly bugging and tracking us, yet we do it to ourselves willfully almost 24/7.
Couldn't he have just put the phones in airsafe mode? Plus if they were tracking the lawyers then they'd probably have seen them arriving at the location even if the signals went off when they arrived.
They address this at the bottom of the article (read the whole thing).
Battery removal can be equally deceptive. Even once one figures out how to extract the primary battery, there may be additional power sources within the apparatus. “Some phones use an additional battery for memory management; it’s unclear whether this battery could be used by logging and/or tracking systems such as Carrier IQ,” Mr. Harvey explained, referring to software that monitors mobile phone users.
Which is stupid because of how obviously wrong it is. Batteries that can sustain the cell radio, CPU, and audio DSP are not small and easily concealed, and more importantly tear downs of phones happen all the time and no such reserve power is ever found.
Sure, but those greeting cards typically don't have quad core CPUs in them, either.
You'd have to expand the claim to not only include hidden reserve power, but also hidden reserve memory, CPU, and audio system. Not to mention somehow convincing phone OEMs to actually spend the money adding this system to their motherboards and building it - there's no way in hell you are randomly sneaking that in at the factory without the OEMs knowledge. So now you also have to somehow keep the OEMs quiet about it.
Oh I totally agree that they're not doing it - it falls at the fact that as much power as the NSA has in America, there's no way they're convincing Korean and Japanese phone manufacturers to add a single penny to their BOM. I was just refuting the idea that it would be impossible to do.
I suspect the OP is claiming that Carrier IQ tracking could still be active, in low-power/sleep modes, that still could represent a privacy risk if the battery is removed. When the battery is re-inserted and/or the phone is turned on, then the data is transmitted to the cell tower, then Carrier IQ servers.
What would that be collecting when the main application processor and baseband processors are not powered? The battery referenced in the article is likely to backup a RAM chip and isn't even used in most modern phone/smartphone designs, which use Flash EEPROMs instead. There have however been phones which have a small internal battery powered by the external battery which has to be charged to operate the radio.
Sure, this would block the incoming/outgoing signal. But last time I checked you can still hear your surroundings when inside a Faraday cage/bag. So unless you plan on permanently storing that phone in a bag I can't see how this would prevent a potential recording from eventually getting out.
Some people are saying that mobile phones have a second secret battery, and that battery is used to power the phone to record anything the microphone can hear, and then later when the phone has a signal and normal battery power that recording is secretly sent out to some secret spy agency?
Battery removal can be equally deceptive. Even once one figures out how to extract the primary battery, there may be additional power sources within the apparatus. “Some phones use an additional battery for memory management; it’s unclear whether this battery could be used by logging and/or tracking systems such as Carrier IQ,” Mr. Harvey explained, referring to software that monitors mobile phone users.
Edit: Also tried a cookie can. (And anti-static bags -- see below.) Faraday cage my ass.
Clarification: I use a Nokia. Maybe the technique works only on iPhones? I've heard they have some reception issues.