Some cars have a single warning indicator that goes on whenever a light bulb is burned. Cars with multi function screens will also tell you exactly which one is burned.
How do these systems work?
When a bulb burns out the current stops flowing through the circuit. A sensor can detect that and tell you the bulb is burned out. It could be the circuit has been disrupted for a different reason but the vast majority of the time it’s the bulb.
More specifically the sensor probably measures the resistance in the circuit.
Infinite resistance = broken circuit.
“Light Bulb Burnout Indicator light burned out – turn on Light Bulb Burnout Indicator light”
But how exactly does it work? I can think of a voltage divider arrangement where one of the resistors is the lightbulb itself.
These circuits don’t measure the resistance in the circuit, they measure the current (although those two are related). Usually, there is a very small, known, resistance in the circuit, and a differential voltage measurement is taken across this resistor, which give the current in the circuit (Google "current sense circuit’). If this current is below some threshold value, the bulb is assumed to be burned out.
I’m thinking a Wheatstone bridge where a bulb is one leg of the resistance.
No idea how it’s done, but can’t you essentially do it with a NOT gate or something similar?
Here is a circuit.
That test is actually performed for the MIL (“check engine light”) during the start cycle. If the check engine light isn’t working, a particular fault code (P0650 - Malfunction Indicator Lamp Control Circuit Malfunction) is set.
This is so it will show up when the car emissions are checked via the OBDII port.
NOT is the what the output is… The question is the discriminator circuit, and its the same circuit for each for the same technology ( eg NMOS ) and on what to discriminate ?
Well, the issue was that the circuit is
12 volts <> bulb <> 0 volts.
The 12 volts can vary with battery voltage,charging voltage, etc so it can’t be finely tested. The 0 volts will remain 0 volts… So discrimination off voltage (directly) is not possible.
Well its a large current, 10 amps, and if can detect one volt, you run it through a 0.1 Ohms resister. But then that is 10 watts of power wasted, and 10 watts to dissipate (or overheat… ) and the resister has a habit of burning out.
Well they may have shifted to neater solution.
A small inductor and a Hall effect transistor. Hall effect is basically a magnetic field detector. The power draw would be tiny.
Easiest way I can think is that the bulb completes a circuit that powers a relay. If the relay isn’t energized the default completes another circuit that lights the panel.
Easy, yes. But not efficient. You’re burning up lots of energy in the relay for no good reason. (Not to mention there will be a significant voltage drop across the relay.)
Probably the best approach is to monitor the current using a low-value sense resistor (like this) or use a Hall Effect sensor.
I can only speak to one specific instance: we had a 97 Ford Explorer with a headlight that wouldn’t turn on. Replacing the bulb didn’t work, and the fuse was fine. So, I started doing some research, and found an electrical schematic.
As it turns out, before the power went to the lights, it first passed through a little black box under the center console computer called a “light-out module”. A multi-pin connecter passed the power through, and if it detected an open circuit, it would send a signal to the computer to display, “Head Lamp Out” on the computer screen.
So, I popped open the computer, pulled out the “light-out module”, and tested for continuity at the proper pins on the connecter. The left lamp, on pins 1 and 2 as I recall, showed continuity, but the right lamp, on pins 12 and 13, were open, meaning there was a broken connection somewhere in the module itself. So I eased a little fresh solder into the joints, checked for continuity again, and success! Put it back together, and it worked fine.