How to Build a Non-Invasive Current Sensor for Car Battery Amps With PIC18F27K42

Hi all!!!

I'm building a data logger with a PIC18F27K42 and almost everything is complete.
What I'm a bit stuck is with the sensor. I bought a Dodge Journey 2012 and as I
understand in there posts regarding a problem with the battery draining in two days,
is that it may come from something either in the PCM or else. So to help me
finding the ELSE, I must use a current sensor that wild be around the positive
wire of battery to be able to measure the amperage while not running.
So far so good but what I like to know is how to build that sensor.

1: No "shunt" resistor.
2: I can't disconnect the battery post for the test as it will reset the computers.
3: The amps can go over 500amps on starting (150A on charging).
4: For the test, the preferred window is around 10A to .05A.
5: The current will change direction (Charging while running and discharging while starting)

I thought of using a kind of toroid, around the positive wire ,
with lots of small wire turns, going to an op amp with an input clamp.

Is there a simple way to build a sensor capable to withstand the starter inrush
and being able to measure from 10A to .05A when the car is not running.

Any suggestions on parts or diagrams will be appreciated.
Thanks all.

A toroid with some wire wrapped around it will not work at DC.
If you are looking at a long term problem, then surely you can disconnect your measuring equipment from the sensing network if you wish to start or run the engine, since you have stated that you wish to measure the current drain while not running.
Modern cars often have a current sensing resistor in the battery lead, not sure if it applies to your vehicle though. They can be retro fitted,https://www.oreillyauto.com/shop/b/lighting---electrical-16777/sensors-16480/battery-current-sensor-20010/47a469757973 for example. Don't know how good these are temperature coefficient wise. Additionally, at the current levels you are talking about, you wold probably have to do some low drift high gain amplification.
If that is not possible, you could measure the voltage drop across a length of the cable, it is terminated at two points that should be accessible. The disadvantage of that is that you would need to be able to calibrate it, which means knowing what the resistance of the cable length is. Additionally, copper has a temperature coefficient that would further hinder the measurement.
The best possibility may be to use a Hall Effect sensor, these are manufactured by a number of companies, and they work down to DC. They are available at the chip level, which would give you the best flexibility as far as getting it into your system is concerned. To get better sensitivity at lower currents, it may mean splitting a toroid, and putting the sensor in the gap. TI https://www.ti.com/sensors/magnetic-sensors/overview.html have some info, there is plenty more out there if you do a search.
Cheers,
Richard

Thanks for the reply.
I ordered some toroid and linear hall effect sensors
to do some test as how it will react on low amp and
on the 500a on start. I look at the split core, as
per your info and I will try that.

I don't know what would be the voltage at around 10a
down to 0.1a so I must make some test.

Just a quick question: What will happen when the
sensor will receive the 500a. Will it burn ?

Its the magnetic field you have to consider, look at the data sheets, but I'm guessing all it will do is saturate.
If you use a core that may well saturate as well, you have some maths to do.
For starters, what field do you get around the cable at the currents of interest (the ones you want to measure at the low end of the current range) The Hall effect sensors have a sensitivity that will typically be so many mV output/mTesla of field. The purpose of the toroid is to concentrate the field locally to get a better sensitivity from the sensor. The toroid core will saturate at some point, at which time no further concentration of the field is possible.
Have a look at this posting, https://hackaday.com/2018/01/27/make-your-own-current-clamp-probe/
By the way, by using feedback, you can get a toroid to measure DC current(zero flux transformer), and to improve linearity, you can also use feed back with a hall effect sensor, both use the idea that you drive the flux to zero in the core, the feedback signal to do that will be in some way way proportional to the flux, and hence the current in the conductor.
Cheers,
Richard

Hi
I have found this link.
It's in Italiam, but you can translate.
Cheers.
Giovanni

https://www.digikey.it/it/blog/smart-current-sensors-for-isolated-high-current-applications