I think part of this is because the chainline between top idler and the cranks was incorrect.
One the Trek Session document it has a top idler, which has 2 spacers, 1 on the left, 1 on the right of the idler wheel. They're different thicknesses which means you can offset the idler depending on the crankset
The way they show it is Shimano and SRAM. We used to have SRAM X01 cranks but they snapped and we replaced with Shimano cranks. However at the time i didn't swap the spacers. Recently though i did.... and now i look at the chainline vertically i believe it's incorrect
Screenshot_2025-01-26-10-16-37-718_com.google.android.apps.docs by steveweeks59, on FlickrBoth cranks are 83mm BB external. But i think they're actually the same chainline.
I've 'eyed' it using the fork leg as a straight and vertical edge, when i've got it set in Shimano, it looks skewed. When it's in SRAM, it seems to follow the fork leg.
The other thing i tried was using the chain and dropping it over the idler and down towards the cranks, which again looks straight.
Lastly i used a metal rule, going upwards from the crank on back of chainring up to the idler and that looks correct.
But all 3 of those methods are slightly guesswork. There's another spacer of 0.5m and again, depending on where you put it, moves the chainline.
I understand how to measure the actual chainline, but not sure on the best way to measure the idler.
I could in theory measure from the center of the frame on seat-tube over to the idler, but i'm not sure how precise it'd be.
I supposed part of the question is, how accurate does it actually need to be. On the GT Fury it was 'eyed' and seemed OK and i'm happy that we're not a million miles out but just would like peace of mind.
IMG_20240815_074003 by steveweeks59, on Flickr