I have several voltage dividers in my car as well. In theory FB is accurate to 0.005V. My experience in my car with all the noise and hoopla 0.015V, maybe 0.01V. Things like my speedometer I try to get close to 0.02V per unit of interest (i.e. 1 MPH= 0.0186V). Not the FB at issue, but everything else going on inside the car.
Here is a pic of my FB config when 12V regulated is powering the FB:
You can see in this pic that the FAN is enabled and the there is no LED showing the dig output is enabled. That port is always outputting ~11.2V when a regulated 12V is supplied to the FB.
Here is my config when USB is powering the pics and 12V for DO's:
The DO led is lighting up, and is outputting ~11.37V
You said it wouldn't be a good practice to power the FB with 12v and USB but don't they share the same ground (in my CarPC they do but not my test setup)?
The output ports are switched ground. This means there is always +12v (or whatever else you power it with) on the output pins, but the related pin will only become a ground (completing the circuit) when turned on. So showing the input voltage all the time is normal.
Originally Posted by rEegLer
As for not sharing the same ground, that is a valid point. Generally you should power it from a single source to prevent any problems. In most installations the grounds are shared (as with your CarPC). For testing, powering via USB but having the +12v connected to switch relays is the best method. It also prevents the FB from having to regulate the input voltage.
This is exactly correct. Getting past 10bit DAC resolution costs money, and isnt needed at the hobbyist level.
Originally Posted by ACCLR8N