Commuting - EEs & hams: Trigger for inductive loop traffic sensors?

On another thread, MarkS commented:


I've often wondered if it wouldn't be possible for some EE biking genius to design a broadcaster that would beam a signal down at the roadbed, making the loop act as if there was a semi up there. This would be especially useful at all those intersections where the cut lines have been paved over.

Interesting project idea. I have a lot of digital signal processing background, but not as much experience with antennas and analog circuits. Here's an article describing the basics of inductive loop traffic sensors:

Detection of Bicycles at Traffic Signals (http://www.humantransport.org/bicycledriving/library/signals/detection.htm)

Maybe there are patent applications that describe in detail how these sensors work, and especially how they detect the change in resonant frequency when a car is present. I don't know whether they would necessarily respond to any kind of injected signal. Maybe they would, and maybe not. If they did, it might be possible to sense the signal that excites the sensor and feed back a similar signal that's shifted in frequency by the right amount. A horizontally oriented loop would magnetically couple to the sensor.

Any technical ideas?

This thing claims to affect the induction loop and trigger the sensor.

http://www.greenlightstuff.com/trigger.html

This thing claims to affect the induction loop and trigger the sensor.

http://www.greenlightstuff.com/trigger.html

Thanks, I took a look at that page. The FAQ and the test summary have technical information. The Green Light Trigger is a permanent magnet that attaches to the bike. I don't think inductive loop sensors respond to static magnetic fields, but I can imagine how vibration from traveling over a road surface might couple in enough of the right kind of signal to trigger the sensor. I ordered two of them and I'll try them out. (The web page says that using two of them gives 50% better results.)

The Green Light FAQ says you need to be traveling 5-10 mph over the sensor to trigger it. I'm not sure I could come up on a red light or the gate to an apartment complex going over the sensor at say 10 mph and stop before getting into the intersection or crashing into the gate. Then there is the problem of, exactly where is the sensor if the pavement cuts are obscured.

I really, really, look forward to a world where any Tom, Dick and Harry can build one of these. It will make life sooo much safer for us all.

I just discovered that I can trip a sensor on my way to work by laying one of my bike wheels over it. It's where a parking lot the fourth direction in a four-way intersection. No traffic in the morning, so I just wait until I see the cross traffic light turn yellow and I pick up my bike and get ready.

Question: Is it the steel in my spokes or my aluminum frame doing the job. And how much steel do I need. Can I do it with my cleats?

It's the conductive material. Period. Steel, alu, it doesn't really matter. Search this forum for "inductive" and you should find some expert opinions on the subject.

These inductive loop car sensors seem to be nothing more than giant upside-down metal detectors.

I made a one turn copper wire loop about 14" in diameter. (I cut a multiconductor cable, stripped the conductors at the ends, formed the cable into a loop, and soldered all the wires together.) It triggers my apartment gate sensor when it's held horizontally about a foot above the sensor loop in the pavement.

I was hoping I could wear it like a belt and trigger the sensor by riding over it, but the magnetic coupling isn't strong enough when it's that high off the ground.

To give credit where it's due, this idea was from a long time EE friend I'll refer to only as Bob.

I was going to post this thread, but I guess I'll just bump it. If you lay your bike down and then pick it back up, it will trigger the sensor. Not only does it add metal to the magnetic core (changing the inductance), but moving your bike helps because it induces a current (something like an electric motor).

Has anyone tried this with an aluminum frame bike?

I was going to post this thread, but I guess I'll just bump it. If you lay your bike down and then pick it back up, it will trigger the sensor. Not only does it add metal to the magnetic core (changing the inductance), but moving your bike helps because it induces a current (something like an electric motor).

Has anyone tried this with an aluminum frame bike?

Why, oh why do we have this discussion again and again? The sensors are triggered by simple physics. A conductive object moving in a magnetic field induces a current which is pick up by a current detector. Very simple. Any conductive object moving in a magnetic field will induce a current. The only issue is whether or not the current detector has been set low enough to detect the rather small current the mass of metal the size of a bicycle induces. Riding directly over the loops will help with inducing enough current. No magnets or other fancy devices needed. And, yes Virginia, it works even with aluminum bicycles. It probably even works with carbon bicycles, because if you read the link, the wheels are probably the most important part of inducing a current.

I read somewhere in a discussion about composite bikes that people have had great results from putting a loop of copper wire (soldering the ends together) in their wheel beneath the tube. If your bike isn't able to trigger signals when laid down sideways on the sensor and you have metal rims, you might be out of luck at that particular signal.

I also read that the best strategy is to roll over directly on top of one of the pavement cuts/wires. This will give a much stronger reading than rolling over the middle of a loop sensor where you only cross two parts of the loop momentarily at a right angle.

You need a loop of conductive material, the nearer to the loop, the better. It's that simple. Stop with your wheel rims over one of the loops. That always triggers it for me. If not, you can try leaning the bike a bit into the coil.

It does not matter if the conductive material is ferrous, or even if it's actually metal; a loop of graphite or carbon thread should work, if it's conductive and a closed loop. All that's happening is that there's a coil in the road that's oscillating at a high frequency. When a conductive loop or large lump of conductive material comes near it, it changes the frequency (lowers it) and the circuitry detects that change.

If you can't see the coil in order to stop on it, ask the road commission to mark it. If it doesn't trigger for a bike, report it as defective.

Would a loop detect a trike, which is closer to the ground, more easily than a regular upright bicycle? It would seem that a trike places far more metal closer to the loops, and would have a better chance to trigger it.

Keep working hard with your local government body responsible for the sensors to get them re-tuned for bikes and applying paint markings to indicate where they are after repaving. Work with local cycling advocacy bodies to gain support to make changes to road maintenance policies for painting after paving, for standards on how they are tuned. It can be a long and hard process, but the results help all cyclists.

Even all 'carbon/plastic' bikes but with Aluminum wheels and/or cranks positioned right relative to sensor pattern can be enough to trigger a properly tuned inductive sensor.

Al

Another important consideration is that some traffic lights control systems are designed so that even after the loop is triggered, it may still take several minutes until the light changes* or it may be set up so that more than one vehicle must be in line to trigger the light. Both these cases are common for where side streets cross main streets.

What may appear to be a faulty triggering system to the impatient may actually be how the system is intended to work.

*Sometimes this is a built in delay, sometimes it is set up so that there must be a minimum time between triggering, for example perhaps the light was triggered and turned for a vehicle that was at the intersection 1min before you. Then you arrive and trigger, but the system waits a default 5min before turning again, so even though you triggered it, you still must wait 4min. This is done to help keep traffic flow of the main street instead of stopping it very frequently for scattered vehicles on the x-street.

Al