The fitting in question has an A/B ratio of less than 2, so the analysis should be acceptable to use per the limitations specified in LM 88. You are correct in pointing out that the negative d value results in a value for K1 of greater than what is on the chart, which in turn makes the bending...