Opto feedback with opto biased with 4mA of current for better transient response?

[grizedale]

Hello,

To get a bit better transient response in your flyback, eg from no-load to full-load and viceversa......have the opto biased with about 4mA of current.

this is because the CTR of opto-couplers is terrible at currents below 1mA.....



JUST SEE PG 5 TO SEE HOW DREADFUL THE cny17 OPTO'S CTR BECOMES AT LOW CURRENTS.
(APOLOGIES FOR CAps)

opto datasheet
CNY17-2 Datasheet pdf - CNY17-2 · General Purpose Phototransistor Optocoupler - Agilent (Hewlett-Packard)

 

[leo_o2]

For better response, the flyback controller operating frequency should be increased.

 

[FvM]

I didn't hear a question from the original post. To add a comment on my own, adding a bias current seems reasonable, you are achieving this basically by defining the circuit on the primary (transistor) side.

 

[grizedale]

Sorry i forgot the question...it was actually an inquisitive statement.

I mean like R9 on the secondary side which does nothing other than pull current through the opto-diode.

...correpondingly, on the primary side . i have added R13, which is needed otherwise the smps will be constantly "braking" itself.......i.e. stopping itself from being able to transfer maximum power.

https://i55.tinypic.com/1197238.jpg

-so i take it we agree that putting this current through the opto is a good idea to improve transient response?

 

[FvM]

The optocoupler current in active operation is basically set in the circuit around the 3842. Adding a bias current for the LED mainly reduces the primary current, according to the current mode controller dimensioning. Avoiding the low CRT range of the optocoupler can be expected to improve transient behaviour, but I'm not sure if the effect is considerably.