Differentiate between Inverter output and Mains

Thread Starter

keyboardcowboy

Joined Dec 31, 2012
25
greetings all

is there anyway to differentiate between the 220v mains and the 220v provided by an inverter (modified sine wave type) using a microcontroller or any other circuit which can be connected to a controller later. i want to switch a relay when the inverter is providing power instead of mains
 
Last edited:

t06afre

Joined May 11, 2009
5,934
I guess it can be done. By say looking for high frequency content. Also if you have a scope. Try looking at and compare the output then using a small 50/60 transformer. Just thinking out loud here.
 

ErnieM

Joined Apr 24, 2011
8,377
Just look for zero crossings, where the signal goes a bit above zero and a bit below zero.
Two comparators tripping on something like when the signal goes above a few volts (5 or 10). This will prevent false tripping.

Make these separate comparators and compare the time between the signal going to zero (comparator one) and going from zero to negative (comparator 2), and you can do the reverse just to be sure.

With a sine wave these times will be close to zero.

With a modified sine wave these times will be much larger.

Or you could do it with an A2D converter by taking a bunch of samples and doing about the same.
 

praondevou

Joined Jul 9, 2011
2,942
greetings all

is there anyway to differentiate between the 220v mains and the 220v provided by an inverter (modified sine wave type) using a microcontroller or any other circuit which can be connected to a controller later. i want to switch a relay when the inverter is providing power instead of mains
You could measure inverter input (DC) or output (AC) current or tap into the control signal to the static switch.
 

Thread Starter

keyboardcowboy

Joined Dec 31, 2012
25
Just look for zero crossings, where the signal goes a bit above zero and a bit below zero.
Two comparators tripping on something like when the signal goes above a few volts (5 or 10). This will prevent false tripping.

Make these separate comparators and compare the time between the signal going to zero (comparator one) and going from zero to negative (comparator 2), and you can do the reverse just to be sure.

With a sine wave these times will be close to zero.

With a modified sine wave these times will be much larger.

Or you could do it with an A2D converter by taking a bunch of samples and doing about the same.
I have decided to take the microcontroller route and i have managed to get zero crossing detection working using atmega8, now how should i measure the duration from the zero crossing to say 4v positive?
 
Top