Today I updated by as-built drawing for the RFM networks down the X and Y arms (DCC D1200562). I found that the order of nodes in the X and Y arm loops are not the same (see figure below). This may explain why the LSC receiving error rate of data from h1iscey is greater than that from h1iscex. In the Y arm loop, there is the maximum number of nodes between the iscey and the lsc, while on the X arm loop the two other nodes at the end station and not "in the way".
We will work next tuesday to make the two loops the same.