On the bq40z50 datasheet, the Coulomb Counter Integral nonlinearity is stated to be +/-5.2 (typ) and +/-22.3 (max) LSB.
Two questions:
1) Would it be accurate to say that the +/-22.3LSB error occurs right around mid-scale of the range?
2) Is there a linearity error vs raw counts graph, or perhaps a differential nonlinearity spec or something available that will relate the counts to the non-linearity error?
(note: a good answer to #2 would answer #1 as well.)
We are working on a test budget and find that the +/- 22.3LSB causes a a big 'ripple' in the budget. If we knew where in the range the +/-5.2LSB or where the +/-22.3LSB errors were, that could help us a great deal with our budget assumptions.
Thank you!
Cheers,
Mike