By Kris Magri, engineering intern
Thanks again to everyone who entered the Alex Rider Dream Gadget Contest! As promised, our team at Make: Labs has built the winning gadget: the super-stealth Listening Cup designed by Grand Prize winner Nic. Check it out!
Amid a raft of great entries that were extremely creative, the Listening Cup was deemed the overall winner because it’s stealthy and high-tech, but still buildable. It came with detailed hand-drawn plans, even showing what type of electronic parts would be needed. The original idea was a drinking cup with a false bottom and electronics hidden beneath — a microphone, an amplifier, and a speaker — so that a person could put the cup to their ear and eavesdrop on conversations from a distance, or listen through walls.
Using electronics available to anyone, we found that the Listening Cup can easily pick up faint nearby sounds and make them louder, though it couldn’t listen though walls unless they were paper-thin. Of course, we figure Alex Rider’s employer MI6 could afford some awesome miniaturized circuits, like those in expensive hearing aids, that would boost the Listening Cup’s performance tremendously.
Overall, the Listening Cup was a pleasure to design and build. It really put us in the shoes of Smithers, the gadget maker for Alex Rider (though we are envious of his lab).
Building the Listening Cup
After judging all the entries on three criteria (creativity of idea, cool factor, and technical realism), tabulating the results, and choosing Listening Cup as the ultimate winner, our troubles were just beginning. Now, how to build one?
We’ve published quite a few amplifier circuits in MAKE magazine, and built several more that weren’t published, so we had a pretty good stash of circuits to try. We went through the magazines and identified 3 potential circuits that might work. On the first day, engineering intern Eric Chu and I reached for the quick-build to test what it would be like to have a amplifier with its microphone inside a cup. We stole the amplifier module from an existing project sitting on our shelf: the “Covert Wireless Listening” device disguised as a book, from MAKE Volume 16, the “Spy Tech” issue. We cut a hole in an ordinary red plastic Dixie cup and shoved the mic in. This became our tester unit for the next few days…
Test Circuit 1
After cramming in the stolen electronics and taping batteries onto the Dixie cup with black tape, we soldered a microphone onto 2 long wires so we could try different microphone configurations. There were so many dangling wires that one person had to hold the cup, aim it, and fiddle with the switches, while the other would tag along with the earphones on. It was like being tied together. We discovered that it was very easy to point the mic the wrong way and get head-splitting feedback! We quickly learned to not put the earphones in our ears.
There was much congregating going on at the MAKE offices, with the entire staff in big meetings all day. Roped together by the headphone cord, Eric and I tried to eavesdrop on their meetings. It was a total fail. We couldn’t hear a thing. We tried placing the cup against the glass door to the conference room. People inside stared, then looked away, with a sort of, “Oh, it’s those crazy interns with their odd contraptions again.” I guess they’re all pretty used to seeing these sort of things.
After a while, engineering intern Tyler Moskowite joined us, and we got more scientific. He became our sound source. We traveled to various rooms and sealed him inside, talking to himself. Then we fiddled and reconfigured the mic, and hurt our eardrums more, but nothing worked.
Someone had a good idea that maybe we needed a funnel-shape to direct the sound into the mic. Eric made a paper cone and fitted the mic inside the tip and snooped on meetings. That didn’t work either. How about a parabolic dish inside the cup? I went home dreaming of ways to bend wires, cut cardboard, or mold plastic into a parabolic shape. The next few days were spent in similar tests. Our amp finally stopped working due to all the things we’d tried soldering to it by that point.
Feedback? Headphone Jack
We had big concerns about feedback the entire time; in fact I had raised the concern the first time we looked at Nic’s rough circuit, which placed the microphone and the speaker back to back. At the time, I thought feedback was an electrical phenomenon, so I thought that sharing the same space on the circuit board would cause problems. Eric figured out that feedback is really a physical phenomenon: the sound waves emitted by the speaker get picked up by the microphone. It all depends on where the mic is pointing. Eric got pretty good at knowing just where feedback would occur, and being very careful not to point the cup the wrong way. But we decided to separate the mic and speaker even further; we would build a hidden headphone jack into the cup. That would be better for stealth listening anyway!
At this point Eric and I both sat down and did some internet research. I came across “shotgun mics,” named for their long shotgun-barrel shape. We quickly made our own, by lengthening the mic’s wires and placing it into a Â½” PVC pipe about 1Â½ feet long. We couldn’t hear any improvements.
Finally, I sat down and learned about making parabolic dishes. I went through the calculations and found out that due to physical wavelength of sound waves, the dish would have to be at least 1 foot in diameter to focus the sound at the mic. A miniature one that would fit inside a cup was against the laws of physics. I felt quite scientific and engineery at this moment. Furthermore, we happened to be studying sound in my college physics class, and that very week I had learned that sound is a longitudinal pressure wave. The human ear can detect these tiny changes in pressure over an amazing range, down to a displacement as small as the spacing between atoms. It dawned on me how little we in the lab really knew about acoustics.
Test Circuit 2
But by the next day, Eric and I had arrived at the same conclusion. We realized that what we needed wasn’t a large amount of amplification. Instead, what we needed was the ability to pick up very faint sounds, and amplify them just a little, enough to drive a headphone speaker.
Eric found a circuit that did this in the online archive redcircuits.com. He ordered all the parts and we each built the circuit on breadboards. Two circuits — neither worked. I checked his over and found several errors. Mine didn’t work either so I emailed the author of the circuit and got a lovely reply back from Flavio Dellepiane. The next day when I came in Eric had got his circuit to work (I never did get mine to work!). It seemed very promising — so promising, we didn’t even try the Dixie cup. We went right for the real thing.
Keith Hammond, the editor in charge of the project, decided to house our amplifier circuit inside a collapsible travel cup to make it even more spy-gadget-esque. Eric ordered this great stainless steel collapsible cup, so my task was to try to fit the circuit onto a small round PCB about 1Â¾” in diameter. We were using all through-hole parts, and the two 1uF capacitors were huge — each was 1″ long! This simply would not do. I scoured the Digi-Key catalog for smaller caps and found some tantalum caps I thought might work. Before they arrived, however, in a stroke of luck, we found some electrolytic 1uF caps in a neglected box in the lab. We threw them into Eric’s breadboarded circuit. They seemed to work and were significantly smaller, so I went ahead with the build.
Arranging all the parts into a small space while still being able to follow the circuit flow was a pleasure. I really got into the Zen of it, in the late evening in the lab by myself, and there was an enjoyable element of craftmanship in the task. I had the whole circuit in my head and knew where each lead on the bottom side (as opposed to the component side) needed to be attached. I thought the finished circuit was quite nice, and, it worked — always a bonus.
The next day was our deadline to have the circuit in the cup and working. Eric drilled a hole into the cup for the audio jack and we mounted the circuit. The way he did it was really slick, nicely disguised, and still allowed the cup to collapse. We turned it on and it worked. We ran around the office with the working Listening Cup, trying it out. I was quite pleased to have met our deadline, with 45 minutes or so to spare before my bus came!
But then, a problem. When we showed it to Keith, he pointed out that we had failed to disguise the circuit. The project was supposed to have a false bottom, so you couldn’t see the circuitry inside the cup. ARGH. I quickly cut a false bottom out of metal and sticky-taped it to the circuitry. I knew I wasn’t coming in the next day so I really wanted to make our deadline and be done that night. But the false bottom stuck up too high because our capacitors were too tall.
I tried to rush and desolder-resolder the capacitors. They’re cylinder-shaped, so I tried to lay them over instead of having them stand up. I even found a capacitor that was slightly shorter and substituted that, and it fit quite nicely. Then I replaced everything and tried to get the false bottom to fit. I had about 10 minutes left. The false bottom was a little too big. I sanded and sanded and kept checking the fit. I thought it would work, and the bus-time was looming, so I took a gamble and “used a bigger hammer.” That’s my name for the judicious use of force. It fit, but it busted the jack out of the very nice mounting hole Eric had drilled. Pity. I plugged it in anyway and listened. Thrum-thrum-thrum-thrum. The circuit made a curious throbbing noise. It was broken. My gamble had failed. Eric was not happy with me and started shooting lazor beams out of his mouth. I ran away on the bus.
Victory at Last
The next day Eric troubleshot the circuit and found out that in my haste, I had soldered in a 47uF cap instead of a 470uF cap. That explained that problem. We also forgot about how to fit our current power supply, an AA battery, inside the cup! We decided to use a 3V button-cell lithium battery as our power source. We found that the circuit could handle above 1.5V for a short time and was noticeably louder, but we didn’t leave it at 3V for risk of frying the components. We used Ohm’s Law and calculated the necessary resistor value (which turned out to be about 260 ohms) to drop the voltage down to the required 1.5V.
Eric decided to build another circuit since the first one didn’t fit with the false-bottom plate. The tantalum 1uF capacitors arrived and he tested them on his breadboarded circuit. They worked perfectly so he laid out all the components on a new protoboard, keeping in mind to make it as thin as possible. We decided to use an old technique called wire wrapping to connect the components instead of soldering, since it was easier to undo any mistakes. A notch was cut out of the board to fit around the audio jack and the jack’s threads were secured with a bit of super glue to keep the screw-on ring in place. After another round of troubleshooting, the Listening Cup was finally complete and ready for some spying!