How to detect 3phase...

[yusuf]

it's an opto-coupler, not an auto-coupler. Pin 1 is diode anode, pin 2 is diode cathode, pin 3 is emitter, pin 4 is collector.
You're shouting again, unnecessarily.
See the component changes in the dotted-line rectangle in the attached revised drawing. The led will light only when all three phases are present and in the correct sequence (a,b,c).
View attachment 67745

can you give the name of the software you're using for designing above circuits..

 

[yusuf]

and friend if suppose i want to drive or take any pins of cd4093 high from the output of opto-coupler then what should i will have to do..

 

[alec_t]

can you give the name of the software you're using for designing above circuits..

I use LTSpice (a free download from Linear Technology) to simulate the circuits I design, then use its Print Preview feature to show the circuit in black and white and capture the screen.

suppose i want to drive or take any pins of cd4093 high from the output of opto-coupler then what should i will have to do

Connect the top end of R4 or R5 or R6 to the CD4093 inputs.

 

[yusuf]

Connect the top end of R4 or R5 or R6 to the CD4093 inputs.

So , i have connected the circuit....

I have attached the image in attached..
please have look...
is the connection correct...

and what is the use of 10k resistor at the output pin of opto-coupler...

 

[alec_t]

That circuit should work. The 10k is the emitter load across which a voltage is developed when the opto-transistor conducts, raising the input pin voltage. It pulls the input pin voltage down when there is no current flowing in the opto-transistor.

Edit: BTW, if you find that the output pulses from the opto-isolators haven't got steep enough edges to clock the CD4013 latches in my suggested circuit reliably you could use CD40106 Schmitt inverters between the optos and the latches to sharpen the pulse edges.

 

[yusuf]

alec !

I also want to add led at every phase....
I have added led in the circuit diagram and attach in my attachment..
please have a look there...

is that correction correct !

well others are working excellent !


or like this .......... in other attachment .........

 

[alec_t]

Neither of those two arrangements in post #46 would be suitable . I'll post a further revised circuit.
Why do you need a LED for each phase? I don't see how knowing which phase is missing would help you to cope with a power failure. Or is this a display to impress your mates?

Edit:
Here's the circuit with extra LEDs and their associated components necessary to provide sufficient current.
View attachment 67765

 

[yusuf]

Alec thank youuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu veryyyyyyyyyyyyyyy much once again..

I have tested it's working...
Next.....
Alec i want give some delay...
Means if i remove the phase and power supply (which is connected on pin. 4 of opto-coupler )
the led turns off.
I want to give some delay before turing off..
Eg : 3sec..

Can using capacitor will solve this..

Or any thing else we will have to do...

 

[alec_t]

If you remove the power supply you would need a large capacitor to delay turn off of the LED for 3 sec. I'll have a think and get back to you.

Edit:
If all 4 LEDs are on their current is ~20mA. A capacitor of ~20000uF would be needed to keep their current > 20mA for 3 secs in the absence of the +V power supply, and additional components may be necessary.
You need to think about what you want to happen if, say, one phase goes off for 1/2 sec then returns. Should that state be latched or ignored?
Please be sure exactly which functions the circuit should have. I can't keep on re-designing the circuit when you add new features!

 

[yusuf]

If you remove the power supply you would need a large capacitor to delay turn off of the LED for 3 sec. I'll have a think and get back to you.

Edit:
If all 4 LEDs are on their current is ~20mA. A capacitor of ~20000uF would be needed to keep their current > 20mA for 3 secs in the absence of the +V power supply, and additional components may be necessary.
You need to think about what you want to happen if, say, one phase goes off for 1/2 sec then returns. Should that state be latched or ignored?
Please be sure exactly which functions the circuit should have. I can't keep on re-designing the circuit when you add new features!

OK , alec just guide me...
i will create circuit...
so, where should I connect large capacitor in the circuit...

 

[alec_t]

In order to guide you it would be helpful if you answered these questions:-
1) why do you need a separate LED for each phase?
2) what do you want to happen if, say, one phase goes off for 1/2 sec then returns. Should that temporary off state be latched for indication or ignored? Bear in mind that if you are controlling a motor it could be dangerous to re-start it automatically after a temporary loss of one or more phases.
3) why do you want to delay turn-off of the LEDs? Normally you would want the earliest indication of power loss.

 

[yusuf]

in order to guide you it would be helpful if you answered these questions:-
1) why do you need a separate led for each phase?
2) what do you want to happen if, say, one phase goes off for 1/2 sec then returns. Should that temporary off state be latched for indication or ignored? Bear in mind that if you are controlling a motor it could be dangerous to re-start it automatically after a temporary loss of one or more phases.
3) why do you want to delay turn-off of the leds? Normally you would want the earliest indication of power loss.

I AM JUST SORTING OUT MY CONFUSION...

friend i am just learning on your circuit by doing practical step by step....

So friend for giving delay where should i connect capacitor..

 

[alec_t]

You haven't answered my questions. Your answers will help determine how best to use the capacitor. As I said previously, additional components may be needed; simply connecting a cap won't work.

 

[yusuf]

1) why do you need a separate led for each phase?

for indication at every phase...

2) what do you want to happen if, say, one phase goes off for 1/2 sec then returns. Should that temporary off state be latched for indication or ignored?

it should latched for indication

3) why do you want to delay turn-off of the leds? Normally you would want the earliest indication of power loss.

this is my doubt..
I want to learn how should we give delay..

 

[alec_t]

I want to learn how should we give delay.

Personally I wouldn't delay turn-off. However, if that's what you want I suggest using additional CMOS gates to make delay timers. They and the LEDs and an indication latch would need to be supplied from a very large reservoir capacitor normally kept charged when mains power is available.

 

[yusuf]

Personally I wouldn't delay turn-off. However, if that's what you want I suggest using additional CMOS gates to make delay timers. They and the LEDs and an indication latch would need to be supplied from a very large reservoir capacitor normally kept charged when mains power is available.

I have added a 330 uf capacitor ...
please have a look in my attachment...

will this give a 2-3 sec delay to stay stay led on .. if the power supply fails....

 

[alec_t]

That circuit is ok in principle to slow the turn-off of one LED. The LED current will decay to about half its initial value in ~0.6 secs with the cap value = 330uF. According to my simulation you'd need about 3000uF for a 2s delay, or 5000uF for a 3s delay, to the half-current point. Of course you would need a similar cap for each LED.

 

[yusuf]

hello alec...
well all is working fine...
but as i said previously on taking pin high..
your answer was :

Connect the top end of R4 or R5 or R6 to the CD4093 inputs.

well I want to take pin high of 4012B ( Dual 4-input NAND gates )..

I AM USING 5.6V DC POWER SUPPLY...
I HAVE CONNECTED THE CIRCUIT AS YOU SAID BUT IT IS NOT TAKING PIN HIGH...
I AM GETTING OUTPUT VOLATGE 2.11V ON PIN 3 OF OPTO-COUPLER[ PC817 ] ..
IF I REMOVE THE OPTO-COUPLER CIRCUIT AND GIVE MORE THEN 3V TO INPUT PINS THEN THE 4012B PINS GOES HIGH...
ELSE DOESN'T GOES HIGH..
IS THERE ANY WAY TO TAKE PIN HIGH OF 4012 FROM OPTO-COUPLER OR INCREASE THE OUTPUT VOLATGE OF OPTO-COUPLER TO TAKE PIN HIGH..

CAN WE USE PNP TRANSISTOR TO TAKE PINS HIGH...

THANKS ...

 

[alec_t]

I AM GETTING OUTPUT VOLATGE 2.11V ON PIN 3 OF OPTO-COUPLER[ PC817 ] .

You're SHOUTING again. Please don't use upper case (capital) letters unless essential; it's considered rude and won't encourage replies to your questions.
2.11V is about what you would expect if measuring the 'average' output voltage from the opto-coupler using a cheap DMM set to measure DC volts. The peak voltage (probably ~3.5-4V) is the important voltage and is obviously high enough for the latch circuit, since that is working ok, so it will take the 4012 pin high.

CAN WE USE PNP TRANSISTOR TO TAKE PINS HIGH...

Not necessary.

 

[alec_t]

Oops. My apologies . I've re-simulated the 3-phase delta-connected circuit in posts #36, #39 and #47, which has two phase inputs to each opto-coupler, and realised it won't detect a phase dropping to zero Volts (although it should still detect a disconnected phase or the wrong phase sequence). So don't build it! If one phase drops to zero volts the other phase is still enough to turn on the opto. The earlier star-connected circuit, with one phase plus neutral as inputs to the opto, doesn't have that problem.