PFC controller should not be near PFC inductor?

[cupoftea]

Hi,
The UCC28070A datasheet (page 38) says that the UCC28070A should be no nearer than 1 inch from the PFC inductor....but isn't this mainly referring to gapped ferrite inductors?...because torroid, eg sendust, has no gap and so will be much less noisy?...so presumably the controller could get much nearer than one inch for torroid?

UCC28070A datasheet

https://www.ti.com/lit/ds/symlink/ucc28070a.pdf?HQS=dis-mous-null-mousermode-dsf-pf-null-wwe&ts=1650057335817&ref_url=https%253A%252F%252Fwww.mouser.co.uk%252F

...All seems a bit inconsistent though, because the Innoswitch controller is always recomended to be placed directly beneath the gapped flyback transformer as the following (below) shows.....so howcome the innoswitch doesnt suffer from such noise but the UCC28070A does?

Innoswitch page 52

https://www.power.com/sites/default/files/documents/der535.pdf

 

[Easy peasy]

Are you forgetting the noisy node at one end of the choke ?

 

[cupoftea]

Are you forgetting the noisy node at one end of the choke ?

Thanks i see your point...but in the innoswitch PCB, there is the noisy drain node in the flyback transformer too.
Admittedly the switching node of the flyback transformer is buried at the bottom of the winding stack-up, but the exit wiring still comes out of it........or are you implying that even with that, its nowhere near as exposed and noisy as the boost inductor switching node bit?

 

[KlausST]

Hi,

the UCC28070A should be no nearer than 1 inch from the PFC inductor.

I´ve never seen such a requirement before. And never had bad experience with interference like this.

Maybe because I try to use some kind of "shielded" or "closed" inductors, maybe because of the optimized layout and bypassing..
But maybe this controller has some sensitive circuit inside.
--> So better follow the manufacturer´s requirement. They will have some experience.

Klaus

 

[FvM]

The datasheet is specifically talking about magnetic coupling. I agree that low voltage circuit branches e.g. current sense input may be affected by near by magnetic fields, of course depending on the magnetic component design.