Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic 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.

Register Log in

Opto CTR degradation in SMPS is not really relevant?

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,602
Helped
560
Reputation
1,123
Reaction score
536
Trophy points
1,393
Location
cambridge
Activity points
76,158
Hi,

I am not sure why feedback optocouplers in SMPS are deemed as being wear-out components?
Do you know?

After all, if (as attached) one uses a low value of RLED resistor, then there is plenty of room for the opto CTR to degrade massively and yet still the SMPS will be in control.
For example, the attached offline flyback could suffer reduction of opto CTR down to just 4% and it would still be able to control itself.
However, Unfortunately the opto datasheets don’t depict age related CTR depreciation.

Maybe the attached would suffer opto CTR degrading down to less than 4% within a few years of 12 hours a day operation in 85degc ambient?

Opto datasheet (CNY17)
https://www.mouser.co.uk/datasheet/2/427/cny17-1767437.pdf

(pdf and LTspice sim attached...plus the TL431 model files for the sim if you wish)
 

Attachments


treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,602
Helped
560
Reputation
1,123
Reaction score
536
Trophy points
1,393
Location
cambridge
Activity points
76,158
Thanks, graph on page 3 says after 100000hrs (11.4 years) at 125degc and passing 60mA constantly , the CTR has only degraded by 30%. Wow.
 

Akanimo

Advanced Member level 2
Joined
Aug 30, 2016
Messages
625
Helped
111
Reputation
222
Reaction score
109
Trophy points
43
Activity points
4,382
Hi,
Hi,

I am not sure why feedback optocouplers in SMPS are deemed as being wear-out components?
Do you know?

After all, if (as attached) one uses a low value of RLED resistor, then there is plenty of room for the opto CTR to degrade massively and yet still the SMPS will be in control.
For example, the attached offline flyback could suffer reduction of opto CTR down to just 4% and it would still be able to control itself.
However, Unfortunately the opto datasheets don’t depict age related CTR depreciation.

Maybe the attached would suffer opto CTR degrading down to less than 4% within a few years of 12 hours a day operation in 85degc ambient?

Opto datasheet (CNY17)
https://www.mouser.co.uk/datasheet/2/427/cny17-1767437.pdf

(pdf and LTspice sim attached...plus the TL431 model files for the sim if you wish)
Normally, at the onset of the life of the power supply, The cathode to anode voltage of the TL431 is usually close to the power supply output voltage. This is so because the voltage drop across RLED is usually small. So if you divide V_RLED by RLED, you would have a current that is small. As the power supply ages (and of course the optocoupler) and CTR degrades, there is more demand for LED current. The feedback loop does the work of regulating the LED current to meet up the design current through the optocoupler phototransistor. At this time, because CTR has degraded, I_RLED is increased. So, evidently, the low value of RLED at the Beginning Of Life of the power supply does not mean that the current passing through it is high at that time. However, as time passes and the CTR degrades and the LED current increases, the increase in the LED current causes the CTR to degrade the more rapidly.

The End OF Life of the opto is when the CTR has degraded to 50% of the initial value. This is because when the CTR has degraded to 50%, the degradation becomes more and more rapid that in no time, the opto will completely wear out. The 50% mark is usually considered when designing the feedback loop, especially in situations where there is no RHPZ to detect where you crossover.
 
Last edited:

dick_freebird

Advanced Member level 5
Joined
Mar 4, 2008
Messages
6,766
Helped
1,992
Reputation
3,988
Reaction score
1,815
Trophy points
1,393
Location
USA
Activity points
54,257
For some environments (like space) the opto's
emitter efficiency, insulation and detector efficiency
all can suffer and CTR can really go in the tank.

If the opto is used as a gain element besides,
then gain drop could affect loop stability
especially if it was initially "aggressive" for
(say) load-step response optimization.
 
  • Like
Reactions: treez

    treez

    points: 2
    Helpful Answer Positive Rating

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,602
Helped
560
Reputation
1,123
Reaction score
536
Trophy points
1,393
Location
cambridge
Activity points
76,158
If the opto is used as a gain element besides,
then gain drop could affect loop stability
Thanks yes, this is especially a point for voltage mode SMPS's, where the gain bode plot has a dip at the LC resonance frequency, and the drop in gain of the opto CTR could take this dip below the 0dB line.....instability could then ensue. So yes this is a concern for some modern voltage mode controllers, eg HiperTFS
 

Akanimo

Advanced Member level 2
Joined
Aug 30, 2016
Messages
625
Helped
111
Reputation
222
Reaction score
109
Trophy points
43
Activity points
4,382
Realize that the TL431 cathode to anode voltage, Vka, is high at the beginning of life of the power supply. The phototransistor current is regulated. As the CTR degrades, the phototransistor current is maintained by means of an increased LED current that compensates for the degrading CTR and Vka is decreases. This continue to happen over the lifetime of the power supply and Vka continues to decreases until it gets to a point where the TL431 stops regulating, which is when Vka equals the reference voltage of the TL431 which is 2.5V. TLV431 or equivalent would be about 1.249V.

So, generally, the degradation in CTR doesn't constitute reduction in loop gain.
 

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,602
Helped
560
Reputation
1,123
Reaction score
536
Trophy points
1,393
Location
cambridge
Activity points
76,158
Thanks, as can be seen in equn 7.61, (Basso Book) the CTR is a feedback loop parameter in TL431/Opto based error amplifiers, so CTR degradation will in fact result in reduction in loop gain.
 

Attachments

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,602
Helped
560
Reputation
1,123
Reaction score
536
Trophy points
1,393
Location
cambridge
Activity points
76,158
reduction in loop gain is far better than the other way around ....!
Thanks, yes, and usually this is the case, but as you know, there are those few cases where reducing the loop gain can actually make things worse......as in Fig 5, page 4 of the attached.

..actually, these things are maybe not that rare...the Fig 5 is a voltage mode Buck, and with the power integration chips, voltage mode is getting more common.....voltage mode means that problems can occur when loop gain is reduced in certain cases....due to the phase margin getting poorer.
 

Attachments


Akanimo

Advanced Member level 2
Joined
Aug 30, 2016
Messages
625
Helped
111
Reputation
222
Reaction score
109
Trophy points
43
Activity points
4,382
Thanks, as can be seen in equn 7.61, (Basso Book) the CTR is a feedback loop parameter in TL431/Opto based error amplifiers, so CTR degradation will in fact result in reduction in loop gain.
Yes, confirmed. Loop gain does decrease as CTR degrades.

Actually I was picturing the midband gain as expressed in your photo, along side some other expressions used to determine the maximum RLED. So I had some mix up of beginning-of-life and end-of-life parameters. Now, I have confirmed it and you are correct. The point that the led current increases and the TL431 Vka decreases as CTR degrades are true though.
 
  • Like
Reactions: treez

    treez

    points: 2
    Helpful Answer Positive Rating

Easy peasy

Advanced Member level 5
Joined
Aug 15, 2015
Messages
2,496
Helped
913
Reputation
1,826
Reaction score
907
Trophy points
113
Location
Melbourne
Activity points
14,081
usually one designs a psu to account for a range of opto gains - if one assumes a 30% worst case reduction in CTR over 20 years at 55 deg C ambient - this can be factored in to the design at the outset - ideally use opto's with a very limited gain spread from the factory.
 
  • Like
Reactions: treez

    treez

    points: 2
    Helpful Answer Positive Rating
Toggle Sidebar

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top