# Comparator Problem - Help Checking Circuit

#### Ant_F

Joined Nov 2, 2022
1
Hello, first post sorry if in the wrong place.

Could anyone give me advice on this zero cross detection circuit I'm working on to allow for phase angle detection? Just to add I'm only a student so I'm very much still learning. My background is electrical by nature, electronics is relatively new.

I've built the circuit and have included pictures of the output waveforms, yellow is voltage, current is blue. The AC ground 0V and DC ground 0V are separated in the physical build, that's why the ground references are different in the schematic. When they were connected the diodes failed so they have been separated now.

When the circuit is connected as per the circuit diagram, the output square waves are as the output 1 picture. The voltage wave form is operating as expected but the current waveform is constant regardless of the current transformer output. It always has that pulse near the falling edge of the voltage waveform regardless of the load sensed through the current transformer/sensor.

When I invert the current transformer inputs to the comparator, the waveform is as the output 2 picture. It seems to operate right but I think that I might have just inverted the waveform by 180 degrees by putting the larger input into the inverting pin of the comparator.

The only difference in the circuit diagram and the physical circuit is the current transformer in the physical circuit isn't grounded to a specific 0V like it shows in the circuit diagram. I'm not sure what to do with that, if it should be connected to the same 0V as the voltage transformer or not.

Any help would be greatly appreciated. Sorry for the long post, just giving all the information i have.

#### Attachments

• 75.2 KB Views: 20
• 3 KB Views: 17
• 4.3 KB Views: 13

#### crutschow

Joined Mar 14, 2008
34,680
I think the current transformer ground should be gref2, not the transformer ground.

#### tonyStewart

Joined May 8, 2012
132
There are at least 2 simple discrete ways to create a ZCS pulse at logic levels.

1. Use a FW diode bridge without C to drive NPN with >1M on base and collector to 0V = Neutral. With pull-up coming from a separate PS or Bridge for logic DC . THis is the non isolated version. THen use an optocoupler for isolation on output or transformer for input.

2. Use CD4xxx Logic inverter AC coupled and series R with 1Meg feedback to make a 60 or 50 Hz square wave. USe XOR gate to create pulse on each edge using square wave and RC delayed wave.

use a 1kHz LPF on the sine wave To filter out glitches. I prefer method 2.

#### ericgibbs

Joined Jan 29, 2010
18,988
hi,
There are a number of ways the ZCD can be realized, but the TS has asked.

Could anyone give me advice on this zero cross detection circuit I'm working on to allow for phase angle detection?

I'm not sure what to do with that, if it should be connected to the same 0V as the voltage transformer or not.

E