Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

opto coupler CTR degradation

Status
Not open for further replies.

hamidmoallemi

Full Member level 2
Joined
Jan 28, 2010
Messages
148
Helped
41
Reputation
78
Reaction score
39
Trophy points
1,308
Activity points
2,135
Hi

we are using this circuit for getting a feedback of AC mains to MCU. this is a mass produced circuit.
the problem is that sometimes the opto-coupler CTR decreases so fast in couple of months and makes the circuit malfunction
I have searched and know about CTR degradation with time and its speedup with high temperature but this degradation rate is so unusual

?anyone has experience with this
do you have any comments about the circuit
thanks in advance opto.png
 

Hi,

Where do you have this schematic from?

I doubt that this is a recommended circuit for generating AC voltage feedback. (At least I don´t recommend it)

Klaus
 

this is a schematic that has been designed early , its a typical one
why don't you recommend that Klaus ?
 

Hi,

because the CTR has 1:1 influence on the feedback.

**
When you look at flyback optocoupler feedback circuits, then there usually is a zener that generates a threshold and thus the CTR has less (but still has minor) influence on the feedback.

Klaus
 

This is just a AC mains monitor circuit, when ac mains is in circuit makes optocoupler turn on half in halfwave mode and it makes the output collector to get Low ,
there is no feedback loop

the problem is that the Current transfer Ratio of the opto coupler degrades with lifetime
then the circuit doesn't work well ,
 

The peak current is only 2mA if there are no voltage spikes on the AC. Voltage spikes will also arc across a little resistor that has a 250V absolute maximum voltage rating causing very high current in the LED.
I think you should rectify the AC, it feeds a series resistor with a capacitor to ground as a filter then another series resistor to feed the LED. Then the peak voltage of the AC is filtered and is shared by the two series resistors.
 

Attachments

  • AC detector.png
    AC detector.png
    8 KB · Views: 65
I agree with Audioguru that mains surges might be a problem if the resistor hasn't sufficient voltage rating. The risk to see actual flashover at a 0.25 W or larger resistor is limited, but I don't know about your operation environment. Adding a filter or overvoltage limiter is a good idea. I don't see a reason for fast degradation in normal operation. Of course the circuit should be dimensioned for expectable component aging and e.g. still work with 50 % of initial CTR.
 
Hi,

your text of post#1:


now:


What is true now?

Klaus

Sorry for that, maybe its because of my poor English

- - - Updated - - -

Thank you Audioguru and FvM !
I think that is the answer for my question
I observed some spikes on AC , and also my resistor is a small 1206 SMD ,

just another question : can I solve this problem by using more resistors in series? for example 4*33k
 

Resistors in series will prevent a high voltage spike from arcing over one resistor. With series resistors a high voltage spike can still cause a higher current in the LED.
The capacitor I added is needed to reduce the level of the spikes.
 
Yes, use several series resistors, some 1206 resistors are only rated to 50V. You need to have enough to make it at least 310V plus extra for safety margin. I would also consider dropping the total resistance to no more than 100K to allow more LED current to flow.

A better solution is to use several resistors, choose values so the peak current is adequate for the LED, but wire the LED across a bridge of four diodes. If you connect the bridge at the LED side of the resistors, the voltage will be very low so you can use cheap silicon signal diodes (1N914 for example) but the transistor will then be made to conduct on both positive and negative AC cycles so you can also reduce the time constant of R8/C4 to make the circuit react faster to AC presence. For example drop C4 to 1uF.

Brian.
 
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top