Do we really need a multi function flyback SMPS control chip?

[treez]

We are doing a 60W offline flyback.

We realise that all offline SMPS’s need to have overload protection and short circuit protection.
The UCC28C45 PWM controller IC does not have these functions built in, but the L6566B PWM controller does have these functions built in. However, the L6566B is a far more complex chip to get through prototype test..

However, we have a microcontroller on the board, which is supplied by a separate 5V output “housekeeping” flyback, and surely that microcontroller can do all the short circuit and overload protection?, meaning we don’t need a pwm controller IC with in-built overload and short circuit protection? -Because this microcontroller can simply sense the output voltage of the flyback and shut the flyback down if there is a drop in output voltage indicating output short circuit or overload.

The UCC28C45 is a cheap IC which, having only 8 pins, is far easier to get through prototype test than some of the “all singing all dancing” pwm controller IC’s.

So, since we have a microcontroller which can switch off the pwm controller ic on the board, do we need to use a pwm controller that has in-built short circuit and overload protection.?…it seems like overkill?

L6566B datasheet:
https://www.st.com/st-web-ui/static/active/en/resource/technical/document/datasheet/CD00167474.pdf

UCC28C45 datasheet:
https://www.ti.com/lit/ds/symlink/ucc28c45.pdf

 

[dick_freebird]

I'd spend some time thinking through fault scenarios (FMEA)
with an eye to whether any of the faults you'd want to
sense, have the potential to take out the watchman (or
the circuitry between). For example if shorted flyback output
would sag the controller supply, or lose you some sense circuit
function.

If the controller can be relied upon, then fine. But satisfy
yourself that this is so, in all conditions you're on the hook
for. Just because the controller has the PWM by the neck,
doesn't mean the little guy doesn't have a shank in his
waistband.

 

[treez]

thanks, shorted output would not sag the controller supply, because the controller is supplied by a separate housekeeping (low power) flyback. This has to be so because the output is variable down to 6v and so we cannot use the bias winding to supply the controller.
Question is, we will end up implementing shutdown for the smps in cases of shorted output , or overload, by the micro, is this acceptable?....I mean it surely works, but do the regulatory bodies accept it?