Opinions on AC main sensing circuit

Thread Starter

Dariodario

Joined Apr 6, 2019
14
Hi, I'm using this circuit and it works well (i splitted the 390Kohm resistor in the two circuit legs with 2x180Kohm 200V rated resistor). He senses the presence of the main AC at the Signal pin.

1578688411755.png

I have seen a similar circuit in which the 390K resistor is placed before the bridge, on the AC input side. I made it and as for my knowledge it works well too.

I would like to know if there are differences between the two circuits (in terms of reliability or other things), in case I should made a professional PCB with it.


Thanks a lot.
 

MaxHeadRoom

Joined Jul 18, 2013
20,428
I would think a small cap placed after the resistor would be better, athough see the alternative method used in Fairchild AN-3006 fig-3.
Max.
 

OBW0549

Joined Mar 2, 2015
3,304
I have seen a similar circuit in which the 390K resistor is placed before the bridge, on the AC input side. I made it and as for my knowledge it works well too.

I would like to know if there are differences between the two circuits (in terms of reliability or other things), in case I should made a professional PCB with it.
Placing the resistor(s) ahead of the bridge rectifier, in series with the AC input, would allow you to use a bridge with a lower PIV rating (50V or even 25V) since none of the bridge diodes will ever see more than a couple of diode forward drops in reverse. With the resistor(s) after the bridge, the rectifier has to withstand the full peak voltage on the AC line.
 

AnalogKid

Joined Aug 1, 2013
8,377
A common method of preventing a single component failure from blowing up everything else is to have two identical resistors in series. This is required for some surface mount parts because the leads (pads) are too close together for the high peak voltages involved, but it is a *very* good idea for through-hole parts also.

You have not said what kind of output signal you want: a pulse for every cycle, a pulse for every half-cycle, a DC level, etc.

ak
 

Thread Starter

Dariodario

Joined Apr 6, 2019
14
Thanks for your answers. So in definitive I should place the resistor before the bridge.
Another option as rosimpson said, I could design the circuit using the FOD814 and in this case I should have a little semplification as pcb space and a cheaper solution, if the circuits are equivalent.

@AnalogKid: I need a DC fixed level to the output, to sense the presence of the main. I mounted 2 trhrough-hole resistors one the Line side, one on the Neutral side. I did that in order to divide the voltage across the resistance and separate them. At the moment I used this one:

https://it.rs-online.com/web/p/resistori-montaggio-a-foro-passante/0149048/

@MaxHeadRoom I have seen the AN spec, I appreciate your suggestion. I only don't understand the role of the zener D5.

About the small cap you suggest, do you mean in parallel? I made a zero crossing detector like this in the follow, and I was asking me if the role of C2 was as RC filter togheter with R1.

1578745259596.png

Thank you.
 
Last edited:

MisterBill2

Joined Jan 23, 2018
5,784
My suggestion is to put the resistance before the bridge, and to split it into two resistors, one on each side of the line. Thus you would use two 180K resistors, or whatever. The reasons are, first if anything beyond the resistors short circuits or becomes grounded, a resistor will limit the current or perhaps burn up, protecting the rest of the assembly, and second, it will indeed reduce the voltage stress on the rectifier bridge.
Avoiding the special opto-isolator with the two anti-parallel ddiodes will allow you a much wider choice of models, in case one part number becomes unavailable.
AND, be sure to install those resistors spaced above the PCB a bit, because they will run slightly warm.
 

BobaMosfet

Joined Jul 1, 2009
1,063
When working with A/C, for current-limiting purposes, particularly when dealing with high voltages, I recommend using capacitors before the bridge. Reactance is an effective method, eliminating thermal issues altogether. Use resistors on the DC side.
 

Reloadron

Joined Jan 15, 2015
5,359

AnalogKid

Joined Aug 1, 2013
8,377
In #8, C2 is a small noise filter with a cutoff frequency of 360 Hz. That is over 2 octaves above 60 Hz, so it will introduce a *very* small phase shift (time delay) between the actual zero crossing and the output signal.

ak
 

MisterBill2

Joined Jan 23, 2018
5,784
In post #1 the TS sort of implies creating this circuit as a commercial product. In that case it is a far better choice to design the product to not be locked in to a specific part available from only one manufacturer. Hobby builders may not have that requirement, but in a production setting unavailable parts can be an unfortunate show-stopper.
 

Reloadron

Joined Jan 15, 2015
5,359
Depends. What exactly in detail do you want the optocoupler to do under what parameters? How fast do you want it to switch. This is the data sheet for your link. The Vishay unit is designed for an AC input which I assume is what you want? That based on your first post. Read the data sheet and decide if it fills your design needs. That is how it is done.

Ron
 

Thread Starter

Dariodario

Joined Apr 6, 2019
14
Yes Ron,
my question was related to the first post, namely the use of the optocoupler as 230 VAC main sensing. I thought maybe it was understood.
 

Reloadron

Joined Jan 15, 2015
5,359
Yes Ron,
my question was related to the first post, namely the use of the optocoupler as 230 VAC main sensing. I thought maybe it was understood.
With that in mind there were a few good suggestions as to couplers made to do exactly what you are looking to do. Personally I would not be looking to use external full wave rectification, I see no reason. Pretty much expressed what I would use in post #11.

Ron
 

MisterBill2

Joined Jan 23, 2018
5,784
Many LEDs, including in some optocouplers, have lower reverse voltage tolerance, and thus using external rectification is a simple way to avoid grief.
 

Reloadron

Joined Jan 15, 2015
5,359
Many LEDs, including in some optocouplers, have lower reverse voltage tolerance, and thus using external rectification is a simple way to avoid grief.
Yes, I agree but when you have opto-coup[lers designed for a specific application why noy use one like a few of those mentioned?

Ron
 
Top