Head Light Alarm
This is the first project I put a soldering iron to. I have an old F150 pickup that I got tired of having a dead battery because I left the lights on. Yes, there are hundreds of circuits out there to do this (or at least tens of circuits.) The advantage of this one, is not so much the electronics of it, but where I get the signals. Most people tell you they sense when the headlights are on by monitoring the voltage on one of the light bulbs themselves. What scares me about that is that wiring is sealed up in plastic tubing and water-tight connectors to protect it from water and other bad stuff.
Not long after I bought the truck, I got a radio/tape deck from a buddy, and I bought a wiring harness from an auto parts store to wire the new radio where the old one was. The stock radio had a wire going to it that dimmed the LED display when the headlights were on. The new one did not have that feature, so that wire went unused. Later when I looked to put in a head light on alarm, this became the perfect signal source for the headlights. This also came on when only the parking lights were on as well, but I never use the parking lights alone anyway.
Of course the other signal you need is to know when the ignition is on. You can get this from one of the accessory slots in the fuse box.
Figure 1.
This is basically just an AND circuit that lets current flow through the buzzer if the lights are on, AND the ignition is not on. Notice transistor Q1 is an NPN and Q2 is a PNP. I found out (by letting the magic smoke out of several parts) that the diode is required to prevent trying the power the radio if it is left on, the lights are on, and the ignition is turned off. It seems that when that happens, power bypasses the accessory switch and goes through the buzzer, Q1 (because the lights are on,) through Q2 to the "accessory bus" to the radio. Those transistors are not big enough to handle that load so the diode is put there to prevent that.
I'm sorry I could not get good pictures. They all turned out too fuzzy to show to the public.
Some words of caution...
Make sure the wires going to your circuit are firmly secured in a way to avoid vibration. Mine were not, and the solder joints on a couple of them have vibrated themselves loose.
When I was first building this thing, I measured lights on signal to the radio, and got 6V. After building and installing the circuit, I found out that is adjustable, based on how bright or dim you have the instrument lights adjusted. I imagine it is possible to have them adjusted so dim, that the voltage signal to the "radio" is not high enough to turn on its transistor, but I haven't seen that yet.
Last year, this thing failed on me (thankfully my daughter noticed the lights were on and saved me.) I re-soldered the connections for the 3rd time, and it still did not work. After actually troubleshooting it I found out the buzzer failed. Who would have thunk it?