After my last article on resonant inductive coupling I received a question about how those interstate toll transponders work. I’ve never had one for my car. Even when I lived in Massachusetts I only really took the Pike from 495 out west and the toll was always exactly the same. So I had to look into how they work, but unfortunately, it’s not very exciting. They are basically RFID tags, but they are active, not passive. Each one comes with a permanently installed lithium battery expected to last between 3-5 years. When your vehicle approaches the toll plaza, transmitters in the plaza send a signal that “wakes up” your transponder, which then responds with a radio signal of its own. So there’s no induction happening, just good old radio communication.
Toll transponders may not be very exciting, but there is a pretty neat method commonly used for detecting cars stopped at traffic lights. You may notice that some intersections have saw mark patterns where you would wait at a red light. Sometimes it’s just a different chunk of pavement. This is the sensor that detects when a vehicle is waiting in order to more optimally control the signal lights. Did you think that it used some kind of a pressure sensor? A lot of people do, but actually, it’s far more clever and effective.
Inside the saw grooves they install a loop of wire called an induction loop. A controller box transmits electricity through the loop or loops at frequencies between 10kHz and 200kHz. Under normal conditions, the loop simply acts like an air-core inductor. We don’t often use air-core inductors because their inductive values tend to be low. Adding a ferromagnetic core (like iron) increases the inductance of inductors because of the higher magnetic permeability of a magnetic core (versus other materials like wood or air). This is exactly the principle behind an inductive loop vehicle detector. When the metal of your car passes over the loop, the inductance of the loop increases (its core is no longer just air). This change in inductance is picked up by the controller which then decides what do to with the signal lights.
You might guess that the change in inductance would depend on the type of vehicle passing over the loop, and you’d be correct. Smaller cars tend to produce larger increases because they are lower to the ground and therefore closer to the coil. The exact pattern of the signal generated by vehicles passing over the loops are also very unique. It’s almost possible to identify makes and models solely based on the different effects they have on the inductance of the loops.
As a bicyclist, those loops are my worst nightmare because they’re often not tuned to be sensitive enough to trigger based on a metal object as small as a bike. As I already said, the material plays a part as well. My bike is mostly aluminum which has a magnetic permeability roughly 1000x less than steel. Actually, the magnetic permeability of aluminum is almost the same as wood or air. For this reason, many traffic signals these days use optical motion detection cameras placed atop the signal poles. This technology is obviously more complex and wasn’t easily available until recently. Inductive loops are a very mature technology and are relatively failsafe. However, embedding the loop in the road means that if the road is resurfaced, the entire loop will need to come out and be reinstalled.
I was actually very surprised at how complicated the theory behind the operation of these sensors is. There is all kinds of documentation from transportation departments, and road construction firms online about how they work including lots of complicated math that I’ve left out here. As a possible topic for next time, did you notice how this is essentially a sensor to detect metal? Metal detectors work on a very similar principle but are more sensitive and able to do more than just detect YES/NO. As always, if you have any topic you’d like me to explore, send me an email.