Road loop troubleshooting device

Discussion in 'General Electronics Chat' started by ab-normal, Oct 13, 2011.

  1. ab-normal

    Thread Starter New Member

    Oct 13, 2011
    I work for a company that utilizes loops cut into the road to detect the presence of a vehicle. The devices are made by Federal APD. There are more than one loop in a lane.

    I am an old grey Electronics Technician from the US Navy. I know that these loops are the inductive element of a tuned tank circuit of an LC or LR oscillator. I am not privy to the schematics.

    So, the controller is operating, and the tank is operating, at about 200Khz.
    I don't want to know if the loop is working or not, the controller can tell me that.

    I want a simple probe that I can plug into my Fluke DMM to locate where the loop actually is under the pavement/concrete. It would also serve to quickly show if a loop was working or not.

    Preferably, it would have a focussed point on the end of a wand/handle. A hall-effect transistor could detect magetism. Since the loop in the ground is, basically, a metal-detector, How can I detect the coil's electromagnetic field?

    It only needs to be sensive enough so that when the sensor is within 2 or 3 inches of the coil wires, it produces a noticeable increase in DC or AC voltage on my DMM.

    Any ideas?

    I can make my own PCB's. This would be a very profitable device in the industry I work in. If I could buy all the parts for less than $40, I can make the circuit board, test it, Then sell these to the company I work for at $180 a piece. We can sell them for more to other interests.

    I can use a fox/hound device on a loop when the loop is not connected. Fox/hound devices cost > $280. Thus my efforts.

    Email: <snip>

    Last edited by a moderator: Oct 15, 2011
  2. crutschow


    Mar 14, 2008
    I would try wind a few hundred turns of fine wire on a ferrite rod and see what you pick up with that. A ferrite AM antenna loop might work well for that (not a tuned type). To significantly increase sensitivity you can add a capacitor across the antenna coil inductance of a value to create a resonant circuit at 200kHz (if all the loops are near that frequency).

    You will need a low voltage (mV) AC multimeter that goes to 200kHz, Alternately you could build a simple op amp type 200kHz AC amplifier and rectifier to boost the signal and give a DC output that any multimeter can read.
  3. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    Every embedded loop equiped intersection I've come across has thin saw lines in the concrete pavement showing where the loop has been embedded.

    You said that you don't need to know if the loop is working or not, just where it is, but in the very next sentence you say the simple probe you want to make for a Fluke meter could also serve to tell you if the loop was working or not.


    I have worked on traffic light control systems before and locating the embedded loop was a simple visual task, testing it for operation just requires a large piece of metal; like a car or motorcycle. The only proper place to be conducting such tests is from the controller box where all the devices are interconnected and controlled. If you don't have access to the equipment in the signal control box then I would think that you don't have proper authorization to be working on or around the traffic signal to begin with.

    Tell us more about your job and how it involves traffic signal control, and how such a locating ability would be beneficial to those whose build and service traffic controllers. I fail to see a benefit to such a device, if one has access to the control box and all the devices within.
  4. colinb

    Active Member

    Jun 15, 2011
    Well, many roads have been resurfaced or re-sealed, or perhaps the loops were installed when the road was paved.
  5. crutschow


    Mar 14, 2008
    I drive a reasonably large (1050cc) motorcycle and my experience has been that it is not of sufficient mass to be detected by the loop and activate the left-turn light. Eventually, if no other car shows up to trip the light, I just wait for traffic to clear and go through the red light.
  6. mcgyvr

    AAC Fanatic!

    Oct 15, 2009
    The actually passed a law in many US states (and probably other places I'm sure) declaring that as a LEGAL thing to do.
  7. #12


    Nov 30, 2010
    You should carry a photocopy of the law rather than expect a police officer to know the laws he was hired to enforce.
  8. CraigHB


    Aug 12, 2011
    I used to commute on a motorcycle and I can't tell you how many times I've had to do exactly that because the sensor would not detect the weight of my bike.
  9. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    All I can say in reply to those individuals with motorcycles is I'm sorry. The loop circuit is sufficiently sensitive to detect a motorcycle, but tuning must be done during the installation to achieve that maximum level of sensitivity.

    You individuals are unfortunately the victims of installations with incompetent installers.

    It is really just a tuned tank circuit, VERY similar in theory of operation to the common hand held metal detectors you see people using at the beach and parks.

    The company I used to work for built controllers and performed installs in far too many Texas towns to name easily, and we also had contracts with cities in other states as well. Some of the historic downtown centers of smaller Texas towns had brick pavement. Not wanting to destroy the brick to lay detectors, meant using timed controllers for the lights. OR the use of an optical or radar type detector. The company designed a radar based detector for such uses and was having some problems with them I was able to solve. One problem was a large number of transmitters seemed to be defective due to low power output. These simply needed two threaded metal slugs to be positioned to fine tune the cavity size to the actual operational freq. A second problem was 'lock up'. The system would digitize and store in memory the 'image' of an intersection that was empty of traffic.
    If the returned and digitized image differed from the one in memory, a 'car' was assumed to be waiting and the light would cycle. The natural base image of an empty intersection would vary slightly over the course of days and weeks, due to weather and humidity conditions, hence the need to constantly update the base memory image. Sometimes the unit would take an erroneous memory snapshot and then become unresponsive to cars. This was solved by a suggestion that all new memory images be compared to previous stored images and if the differences in the two digital snapshots exceeded a small threshold to repeat the process until an acceptable base memory image was obtained. I'm not 'hands on' familiar with the optical units, but understand they operate in a similar fashion.