Magnets could work but they work far differently than a conductive metal area. They could work by inducing a current in the coil and to do that they have to be moving. Some people who claim they work say you have to move at up to 20 mph, not exactly what you want to do when approaching a red light! But I agree that they cannot work while stopped and they are not a good idea. For one thing what if the sensor circuitry is time gated and only looks at the loop part of the time and you don't happen to cross the coil wires during the right time window? A static mass of metal is there whenever the sensor circuit happens to look for it.

IFF the sensor coils in your area are visible or marked you can park your bike over the wires as shown. If you have a CF bike and rims you can wind light gauge wire inside the rims to create a coil that the sensor can detect, just be sure to solder the two ends together to form a complete circuit. In my area the coils are often buried under a featureless asphalt surface, you can only guess where they might be. Illinois does not use the more advanced detector loops as far as I can tell. Of course if the sensitivity is not set high enough to detect a bike you are out of luck.

The loops are designed to detect horizontal metal surfaces by inducing eddy currents into them. So laying a bicycle flat over the loop may make it detectable when the sensitivity is set too low. Not terribly convenient but some report that it works. Again it helps if you can see the coil but even when it is hidden you have a pretty good idea where it is and the positioning accuracy required is less for a bike laying flat. Now, what if you carried a loop of wire that you could flip down onto the pavement where the detector is likely to be and then retrieve with a string? I don't know how big it would have to be or if the size could be reduced with multiple turns but it would be worth experimenting with this. As with the loops in a CF rim you have to solder or connect the two ends of the loop together to form a complete circuit. I've been meaning to try this myself but a crash a few weeks ago had me laid up and I only started bike commuting for the year yesterday.

I only have one light on my route. I don't blow through lights, ever, but I will carefully cross against the light when it is safe. This particular light is on a very busy four lane road that runs by my employer's campus. I enter the property by going straight across this street from a side street which becomes the entrance to the campus on the other side. There is very little straight through traffic either morning or evening so I would wait a long time if I did not cross against the light. There is always plenty of traffic turning left onto the busy street however. So I wait for their green arrows, then I wait for the oncoming left turns to clear, then I cross while my left turn arrow is still on. Now yesterday I managed to get a green light in the morning without the aid of any straight through traffic so I must have just happened to stop on the detector wires. If only I had been carrying a can of spray paint I could have marked the street where I stopped....

That is another strategy that you could use if your lights are sensitive enough but the coils are unmarked. Spend the time to determine where the loops are by trial and error and then mark the locations. It might work if you don't have too many lights you would have to mark. In some localities if you call the streets department they will mark the coil locations and turn up the sensitivity so don't ignore that possibility out of hand, try calling them. Even in Illinois the state DOT hand books have the usual high sensitivity coil designs and the recommended pavement marking patterns. We just don't bother with them, grrrr.

Ken