I have a workbench that transforms feature classes within the same sde to topojson files for each feature class. Currently I have the same feature in two feature classes that I am writing to topojson. However, the output topojson files are different for this feature. The topojson files are used in Power BI custom maps. One file does exactly what it should while the other doesn't. The issue is with that particular feature that is located in both readers and writers. I checked the number of vertices for this feature in both feature classes and they are the same (3045 - it is a large coastal feature). The output topojson contains different vertices count than the feature classes that is likely expected. But both output topojson files have different vertices count for the same feature: 3431 (topojson file that is not performing as it should within Power BI) and 3421 (topojson file that performs as it should in Power BI).

I processed the feature class that produces the wrong output topojson file in https://mapshaper.org/ and the output topojson file works as expected in Power BI. I do not have a lot of experience working with topojson files but I find it strange that the same feature in 2 feature classes would be written differently as topojson files in FME. This issue occurred when we changed the reader to use the feature classes in our sde. Prior to that the feature classes were read from a file gdb and we had no issues writing them to topojson in FME. Is there a known fix that can be applied here?