A microcontroler is powered by a double power supply. Each of them is composed by a battery, a voltage regulator, and a diode to avoid current flowing from one side to the other. Here's a schematic. Microcontroler is connected to net "5V". enter image description here

Now I want to monitor all the elements of this chain for diagnostic purpose:

  • Batteries, not show on the drawing, are monitored with an voltage divider connected to an ADC
  • Voltage regulator is monitored the same way (VReg5V1 and VReg5V2 on the drawing)
  • I can not find a way to measure if each diode is "alive" or dead. I think the best way is to measure the voltage between each diode pads but I don't know how to interface this idea with the microcontroler.
  • 3
    \$\begingroup\$ Diodes are very reliable - its not really worth the effort. \$\endgroup\$ – Icy Jan 8 '16 at 13:55

Aside from the comment of Icy if can think of some ways to monitor the voltage drop on the diodes:

  1. you can do this with a simple differential OPA circuit and interface it's output to an ADC of the microcontroller. Or you use a differential ADC input (using two input lines) if the used ADCs frontend supports this feature.

  2. if you do not have a differential ADC input but the voltage drop on the diode is high enough compared to the resulution of the ADC (e.g. in your microcontroller) you just measure the voltage before and behind the diode with two ADC channels and evaluate the difference by software.

  3. you build a simple comparator and interface this to a logic input of your microcontroller (diode good / diode failed signal). Examples and a pretty good tutorial for comparator circuits is found here: Op-amp Comparator Tutorial

A completely different approach would be to monitor the current drawn from each power supply. Just use a reasonable dimensioned measurement shunt resistor and monitor the voltage drop over it. Again a differential OPA circuit or a differential ADC input channel can be used.

  • 1
    \$\begingroup\$ Note that if he intends on putting the micro into sleep or a zero-powered state, current may be able to flow through the divider and into the micro input, "phantom powering" it. See EEVblog#831 for an explanation. \$\endgroup\$ – rdtsc Jan 8 '16 at 17:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.