# Inductive Loop Detectors at Hight Vehicle Speed

1. Sep 1, 2011

### Neyolight

Hello Everyone

I am currently studying the inductive loop detectors used in the traffic control department. I believe the data from the detectors is highly unreliable and noisy. A lot of research has been going on on the 'Fault detecting algorithm' and other means to improve the accuracy of an ILD(Inductive Loop Detector).

My question is, How do these ILD detect a fast moving vehicle. Lets say a vehicle going at 100 km/hr on a motorway. It would go over a normal (4m x 4m or even smaller) inductive loop within FRACTION of seconds.

IDL works in the following way : ILD have a magnetic field around them---> incoming vehicle disturbers the magnetic field --> the ILD induces eddy currents into the conductor part of the vehicle --> thereby reducing its own inductance -->if this decrease in inductance crosses a threshold value, a call is made to the electronic unit and the vehicle presence is identified.

This is quite a LONG process. I fail to understand how all this can be done within FRACTION of seconds. Surely the conductor(vehicle) needs to be in the magnetic field (over loop) for 'some' time to be able to disturb the magnetic field PROPERLY.

PLEASE PLEASE PLEASEEEEEEEE Can someone give me some tips on how ILD works in high speed environment ! Thanks !

Neyoligh

2. Sep 2, 2011

### Staff: Mentor

Welcome to the PF.

There are two techniques that are used that I am aware of for the detectors. The traditional way is to drive an AC waveform into the loop, and monitor the impedance of the loop. As a ferrous metal object moves over the loop, that alters the impedance, which the circuit detects as a vehicle.

For slow-moving vehicles or stopped vehicles (like the detectors at intersections), you could just use AC Mains frequency probably (50/60Hz). But as you say, for fast-moving vehicles on expressways, you would need to use a higher excitation frequency to be sure to get a few cycles of impedance change that your circuit could detect.

There is a newer detector technology that I learned about here on the PF. The sensors are just B-field sensors, and they detect the magnitude and direction of the Earth's magnetic field. That gets altered when a ferrous metal object is over them, which the circuit interprets as a vehicle. As long as the sensor and monitoring circuit have reasonable bandwidth, they can be used for fast-moving vehicles.

3. Sep 6, 2011

### es1

Unreliable in what sense? Do they say there is a car present when in fact there is none, or maybe they fail to register cars, perhaps both?

100km/hr = 27.8m/s. If I assume a car is more than 1m long then it will spend more than 36ms over the sensor. That could definitely be enough time to take a good measurement (that's more than the time of two 60Hz cycles) especially if you just want one bit (car present or not).

If two of these sensors were a known distance apart just one bit from each would be enough to make a speed measurement.

4. Sep 7, 2011

### Neyolight

Thanks berkeman and es1

The issue I have with the loop is that it just does not detect fast moving vehicle. I don't know why is that. I would be doing some experiments with a real inductive loop myself shortly and hopefully will figure out the reason behind it. :)