Hi everyone,


I’m working on an automotive project where I need to detect ignition and seatbelt signals using a tracker device, but I want to avoid cutting or disturbing the original wiring.


I was exploring the idea of building a non-contact voltage detector using a transistor-based high impedance circuit (similar to how AC voltage tester pens work). The idea is to place a sensing wire near the target cable and detect whether voltage is present without physically connecting to it.


However, I’m unsure if this approach is practical for automotive 12V DC systems.


My questions are:

1. Is it feasible to detect 12V DC in a car wire using a non-contact method (capacitive or electric field sensing)?
2. Can a simple transistor-based circuit (high impedance input) reliably detect such signals?
3. Has anyone successfully implemented a non-contact DC voltage detector for automotive use?
4. Would noise and shielding in vehicle wiring make this approach unreliable?
5. Should I instead use a direct but non-destructive method like back-probing or an optocoupler-based voltage sensing circuit?

If possible, I would really appreciate if someone could share a practical schematic or suggest a robust approach for this use case.


Thanks in advance!

 

Sensing ignition system pulsing by capacitive coupling to a wire wrapped around the swiitched ignition coil primary wire should be easy, although it may require a bit of amplification. Sensing the status of the seatbelt switch wil be more complex, and probably require "back probing" one of the wires to the seatbelt-in-use switch. In all of the cars that I have worked on, back-probing the electrical connectors has been rather simple and easy, with adding tape over to keep the probe wire in place.

 

require "back probing"

What is "back probing"?

 

Back probing is forcing a skinned section of wire into the part of the connector where the wire to thepin enters the plastic housing. It unfortunately does not work with moded on connectors.
In mosr cases it is important to wrap some tape around the wires behind the connector to keep the inserted wire from coming loose.

 

Hall effect sensors come to mind.

 

Many of the wires that would be monitored do not carry much current, and so any magnetic field will be rather small. AND hall sensors require amplification because their output is low. AND they require external power. I would sayHalleffect sensors are a poor choice for MOST current sensing applications. Not ALL , but most.

 

A wire wrapped a few times around a spark plug wire is a common and simple approach as noted by @MisterBill2 . I have one on my boat engine, and it turns on an hour meter, so I have an accurate measure of engine on time.

What exactly do you want to measure?

 

A wire wrapped a few times around a spark plug wire is a common and simple approach as noted by @MisterBill2 . I have one on my boat engine, and it turns on an hour meter, so I have an accurate measure of engine on time.

What exactly do you want to measure?

I was not thinking aboutsparkplug wires. Consider that if the installation was pperfect, still, loss of the spark plug loading would put a lot of voltage capacitivly coupled into that wrapped wire. For those folks who don't believe that, disconnect one sparl plug and grab the wire well away from the open end. Grab the wire while the engine is running, of course.
Certainly capacitors block the DC voltages, but that fast rie-time pulse couples a whole lot of enery thru the capacitance. In many 12 volt wires in the car there is plenty of lower level noise to operate a simple sensor circuit.

 

We (or the TS, anyway) needs to distinguish between sensing that the ignition is on (key turned to the ON position, but engine not running), and detecting that the engine is running.

There are numerous IDC connectors intended to tap into an existing wire. I used a bunch of them when I added things to my old Bronco. At first I was concerned that the need to nick the wire would cause them to break, but I well over a dozen (possibly two dozen) when all was said and I never had a single failure. Small-sample, so YMMV.