Building an Inductive Loop Vehicle Detector

[Trax] was asked by a friend to build a device that could detect the presence of a car in front of his garage gate for it to open automatically. After searching the web for such a project and trying many of them, he decided to build his own detector based on an induction loop. As you may have guessed, this kind of detector works by detecting an inductance change in a wire loop (aka coil) buried in the road. Having a car pass several inches on top of it produces such an effect.

[Trax]‘s write-up shows a very well thought and professional design. All the detector parameters can be adjusted using DIP switches and buttons: detection type (presence/pulse), signal filtering, main frequency and sensitivity. The wire loop is isolated from the main sensor electronics using a 1:1 isolation transformer and a Colpitts oscillator is used to drive the latter. Moreover, gas discharge tubes are also used for lightning protection.

The change in inductance translates to a change in resonant frequency which is later detected by the main microcontroller. The board is 24V AC powered and a diode bridge + LM2596 SMPS step-down converter are in charge of generating the required +5V in an efficient way.

As if this was not enough, [Trax] also made a PC-based tool that can change other platform settings using a serial connection. All the resources can be downloaded from his website and a few videos are embedded after the break.

Comments

  1. Josh says:

    While I applaud those who build their own circuits, it’s a lot easier (to build AND maintain) to pick up a surplus loop detector used for traffic signals. They’re usually built well, don’t need a lot of tuning, and provide a relay contact that usually deals with 24V, which is generally what garage door openers use these days. It’s not hard to find these for well under $20. That being said, having this circuit powered BY the opener itself rather than another 120V plug is rather nice.

  2. fartface says:

    This works great for all cars, if you want selective add in a RasPi looking for bluetooth mac addresses. it can sense the car and the phones inside the car to only open for recognized mac addresses.

    • Fred (@Karnosbarmyarmy) says:

      If you can detect the *right* bluetooth then do you even need the inductive loop detector?

      • scienkoptic says:

        If you don’t want the gate to crush your car, you don’t. The inductive loop detects the presence of the vehicle. If the inductive loop is placed correctly, it will detect vehicle, open gate and keep it open until the vehicle clears the gate. Good practice dictates multiple loops and detectors in this situation.

        Some might ask why anyone would go to the the trouble of making a loop detector, but this builder is on to something: the serial interface is a welcome addition. I know of no other gate purpose detector that offers such an interface (I’m not talking about traffic loop detectors, but parking or gate loop detectors)

        Reno makes detectors with RFID or some sort of RFID built in. That would be interesting to see added a feature.

        Something I’d like to see: use of three axis compass as vehicle v detection. Banner m-gage does this. Rather novel product, but not without it’s own issues.

      • pff says:

        You don’t want the garage door opening if you walk past it with your phone.

  3. HC says:

    This is cool, but I’d need a much faster door opener motor than my current one for it to not be more annoyance than useful. I usually hit my opener 30-odd feet down the road from my driveway.

  4. bubba gump says:

    Loop detector + Bluetooth +garage door transmitter = a mighty fine “almost secure” garage door opener.

    • Greenaum says:

      99% of the work is in replacing the user pressing a button. The need for heavy lifting, and even getting out of the car, has now been removed by technology. We just haven’t quite cracked the last centimetre where the user has to actually press the button to start.

      I’d stick with a button and call it “solved”. It’s the bit that’s hard work and more likely to damage something when it goes wrong.

  5. AllenKll says:

    I was loving this project right up until, “by the main microcontroller” Bah… If you’re doing some good analog, why muck it up with a uC? He could have easily used some peak detection and RC circuit to drive a comparator to detect the change. Then a simple timed latch (I’d let him get away with a 555.) to run the relay.

    Am I being a bit picky? probably. I just feel it strange to do all that wonderful work and then cop out on the last stage with a uC. At that point, why not just generate the signals with uC, and skip the oscillator?

    • No Hack says:

      By using uC you get more flexibility and configuration parameters. With analog you would get only detect/undetect. Implementing all these parameters mentioned in this project with just analog would be ridiculous :-)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 94,048 other followers