How Loop Detectors Work
A frequently asked question in the newsgroups is about how the vehicle detectors used in traffic lights and in car parks work. The author designed such a detector too many years ago for comfort, and has frequently posted replies to these questions.
Here is one of those replies. It is not a definitive treatment of the subject, but should give you an overview of how loop detectors work. When you approach a traffic signal or car park boom gate, you may notice a rectangular “scar” where the road surface has been cut with a saw and then re-sealed.
This is the sensor loop. The loop consists of several turns or wire buried in the roadway, and connected to an electronic circuit which can detect a car passing over it. Normally the loop is about 1m x 1.5m in size. How do the electronics detect the passage of the vehicle over the loop?
There are three methods I know of:
“Damped Oscillator” The loop is part of an oscillator circuit. The oscillator is held just on the verge of oscillation via AGC. When a car enters the loop it “dampens” the loop by absorbing energy, and the oscillation decreases or stops entirely.
The circuit needs a very long time constant in the AGC loop in order to detect slowly moving vehicles, because whatever time constant you start out with is reduced by a factor equal to the effective loop gain of the AGC. Eagle Signals had a product of this type in the ’60’s. It used a partially charged NiCd battery as a huge capacitor to get the long time constant required. The main disadvantage is that every loop will have a slightly different natural frequency of oscillation, and the interactions between adjacent loops are nasty due to beat frequencies (imagine a straight-through lane right next to a slip (green arrow) lane). “Fixed Drive” A fixed frequency oscillator and the driver drive the loop from high drive source impedance.
When a car enters the loop it “dampens” the loop by absorbing energy, and the oscillation decreases or stops entirely. The circuit needs a very long time constant in the AGC loop in order to detect slowly moving vehicles, because whatever time constant you start out with is reduced by a factor equal to the effective loop gain of the AGC.
Eagle Signals had a product of this type in the ’60’s. It used a partially charged NiCd battery as a huge capacitor to get the long time constant required.
The main disadvantage is that every loop will have a slightly different natural frequency of oscillation, and the interactions between adjacent loops are nasty due to beat frequencies (imagine a straight-through lane right next to a slip (green arrow) lane).
“Fixed Drive” A fixed frequency oscillator and the driver drive the loop from a high drive source impedance. When a car enters the loop the loop voltage drops by a very small percentage. The drop is detected. The circuit requires a differentiator amplifier which can pick up loop voltage changes as low as .5% with a time constant of several seconds.