Benjamin Franklin was the oldest signer of the Declaration of Independence, at 70. Highly inventive (credited with the lightning rod, bifocals, and the carriage odometer), he participated in many discussions with Priestley and Jefferson. At the Priestley house, I had watched the scientist make steel wool ignite and then place it glowing in an overturned glass. I realized that this would be the origin of a light bulb. It would take another 100 years for Thomas Edison to invent one.
So, I’m a little limited by the fact that the tail lights don’t work on my car. The fuse keeps blowing. I’ve gone through a box of them. Sometimes, it takes 20 minutes, sometimes 5. The short doesn’t appear permanent, but it occurs relatively quickly. I’ve tried putting in a 10A fuse instead of the 7.5A one, and this doesn’t stabilize the situation. It’s overloading by a significant bit. I don’t think the electrical problem occurs at the power source. There is a 80A fuse from the battery, and the fact that most of my other circuits don’t blow, rules out main fluctuations to the fuse box as the source of the problem. The switch itself normally controls both the head lights and tail lights. The head lights are stable, the tail lights aren’t. Since there are separate fuses on the fuse box for the head lights, and tail lights, the circuit splits from the switch prior to entering the box or at the box before the fuses. I could run a wire from the head light fuse inlet to the tail light fuse inlet, taking out the head light fuse, and keeping the tail light fuse in, and see if that blows a fuse, to definitively rule that out. It seems more probable to me that the short is downstream, but probability is never a really good argument for solving a specific problem. The above strategy would likely be negative proof by blowing a fuse, with alternatively a slight possibility of positive proof by lack of result (which could also take a while). Neither possibility being a psychologically appealing outcome, I’m going to put that experiment aside for the moment, in favor of another more constructive strategy.
(An aside: with a meter, one might simply assure connectivity between both high ends of the head and tail light fuse inlets at the box (fuses removed), and then again between both low ends of the head and tail light fuse inlets) with the switch off. I get no resistance between the tail light high and the right head light high (both highs measured off the low beam circuits), and 1 KOhm between the tail light low and and left head light low with the head lights off and all 3 fuses removed. In one scenario, it could very well be a jump between the lows to blow the fuse.)
As it turns out, inspecting the tail light assembly a little more carefully, there is at least a signal light, 2 tail lights, a reverse light, and a brake light in each housing. Removing one tail light from the housing (still wired), I was able to see it come on and off as I turned the lights on, and still have tail lights (at least temporarily) with a 5A fuse in there. I can also take the bulb out, and still have tail lights. So, I think the diagram below is probably more accurate:
Going through all of the wiring can be tricky (although it might just be one of the many possible wires that shorts causing the fuse to blow). Just finding the wires can be tough. I decide to set up my own little switch, with an LED indicator to let me know if something fails, and using an independent battery on the dashboard, run some wires to the tail lights on either side, splitting the circuit so that if one side fails, the other side will still be active. This should not, in principle, be too much more difficult than building a flashlight.
Removing the tail light bulb from the circuit, I put a 3A lamp switch inline on the high end, and ran it to a 6V battery. It lights up, but not very strongly. Putting 2 batteries in series will give me 12V, and then I could just cut and resplice the wires to the housing maintaining the original bulb, using terminators on the old wire ends. I still need to get the LED indicator for the dash, and then I could just use it as a backup system if the current system without the bulb actually fails. I could be lucky, and have correctly picked the circuit that is shorting with the one bulb removed. If no current is drawn through those wires (heating them up), then the short might not happen again with the bulb out.
To be really elegantly simple about the problem, one could take one of those led head flashlights, and duct tape it to the hole in the housing on each side, turning them on when one wants to drive at night. They take rechargeable batteries.