AMC23C10: Zero-Cross detection circuit detecting peak instead of zero volts

Part Number: AMC23C10

Tool/software:

Hi, I'm testing the circuit below and it's detecting the AC peak instead of zero cross, unless I connect oscilloscope to isolated GND OUT2 then it works as expected.

Is connecting Neutral to INN necessary or can it be floating.

3 months ago

0 LA 3 months ago

Prodigy 20 points

Here are some screenshots:

I'm detecting falling edges only every 20ms for 50Hz AC

Here's the OUT2 compared to the AMC INP/INN:

Here's AMC INP/INN compared to the AC:

Here's the OUT2 compared to the AC:

0 Saleh Ahmad 3 months ago in reply to LA

TI__Expert 4530 points

Hi LA,

Neutral should be connected to INN/GND1.

Thanks.

0 LA 3 months ago in reply to Saleh Ahmad

Prodigy 20 points

Connecting Neutral to GND1 blows the circuit, the AMC23C10 high side is powered through an isolated SMPS.

0 Alexander Smith 3 months ago in reply to LA

TI__Guru 51271 points

Hi LA,

Which part of the circuit blows?

0 LA 3 months ago in reply to Alexander Smith

Prodigy 20 points

The initial schematic is incomplete, so here’s a more detailed version that includes the SMPS internals. I connected AC1 to GND1/INN, which caused the fuse to blow along with the bridge rectifier.

+1 Alexander Smith 3 months ago in reply to LA

TI__Guru 51271 points

Hi LA,

Yes, this makes sense.

I think you may need to consider a change in architecture in order for the circuit to work properly.

Building something like this is an option: https://www.ti.com/lit/an/snva735/snva735.pdf

Or using a standalone DC/DC converter to convert voltage from the low to the high voltage domain is another.

An example of such a circuit can be found on the EVM, we also have modules from TI: https://www.ti.com/power-management/acdc-dcdc-converters/overview.html

Amplifiers

Amplifiers forum

AMC23C10: Zero-Cross detection circuit detecting peak instead of zero volts