Mr. Danneger,

Perhaps your point is obscured by your grammar.

You wrote: "The described behavior causes never, that the "circuit isn't working" nor that the "code "crashes"" !

I'm not sure what you are exactly saying but perhaps you missed the point. I will re-explain.

The 2051 data sheet indicates a max current that can be expected to be dissipated when in power down. The errata clearly indicates that this is not met. If your circuit was designed based upon this expectation then it "is not working". The very general case where this is troublesome is that of a battery powered device.

Your disinterest in this characteristic because you "use the 2051 mostly not battery powered" does not invalidate the implication of this problem for those of us who do.

You correctly infer that this problem has only been noted by ATMEL and myself to be characteristic of the "E" revision.

I am please your home project can survive 8 hours of continuous execution, however mine must perform across the industrial environment at peak specification for, at least, 3 years.

The relocation of the invocation of the power down bit (the work around) does not consistently work in implementation and that can be demonstrated so i am not sure what you mean when you say "And also nothing about, that the work around (use revision F) not fix the problem."

Your last point "Please read the errata sheets more carefully in the next time !" could also be applied to message posts.


regards,
p