Malund still the point is missing if a switcher is not upto its mark and you try to filter the crap , it has a potential to harm switcher electronics in the first place .This would result in what is know as a failed powersupply , see for yourself how the Damn TINY/TOP crap switches perform and fail .
are you going formal?, I'll still refer to you as Richard. Anyhow, you do not get the point, do you?
1) We design a filter according to the switchers datasheet and then verify the combo.
2) we DO verify the result of the combo
3) with the filter designed according to the switchers datasheet, there is no risk of "harming switcher electronics" (see 4))
4) we do not use "crap switchers"
5) if the combo had not performed "better than good enough", we would have gone into detail.


Erik Malund said:
I freely admit that I 'blindly' trust that my uC will execute the instructions as stated in the bible/datasheet. I also freely admit that I 'blindly' trust that my uC will be capable of delivering Voh min at Ioh max.
Richard replied:
I freely admit I leave the datasheet as well as the component in which the term 'TBD' is mentioned . I dont trust datasheets blindly.
this begs the question: "how do yo design?" if you do not trust the datsheet in the design phase how do you ever get to a point where you can do your "measure' 'analyze' or whatever.

Of course, anything has to be verified, the only point where we differ is that I say "if a signal is solid, it does not matter if it is solid because of hard drive or low load" where you say "I need to know WHY it is solid". Of course, the drive has been designed to match the load.

You, evidently, am not in any hurry to get stuff out the door, I am satisfied if things are "better than good enough" and have no interest in knowing why it is so.

As an example I will not accept "5V tolerant" if I interface between 3V3 and 5V I always use a level converter because it is "better than good enough" even if accepting the 5V tolerant might be "good enough".

Of course, the fact that my stuff represent a very small part of the "system cost" gives me some latitude to use a 33 cent component that might not be needed in order to make my design "better than good enough".
Were I to consider using "the Damn TINY/TOP crap switches" in order to save a dollar or two, many thing would change.

one final note:
we are international and do not have a lot of serveice reps (our stuff is "better than good enough") and thus most warranty repair involves travel etc. Thus our average cost for a warranty repair exceed $500 and to save on warranty repairs I can spend quite a few $0.33 components to make things "better than good enough" and still come out ahead.

Erik