Hi @johntdata,
first of all: Amazing example!
we should use this approach when testing or investigating isochrones…
Just off the top of my head: This might probably be this issue Different isochrones when specifying multiple travel time ranges vs just one range (this links to several threads/issues that explain the problem further)
In that case, there is probably no other fix than editing the algorithm code itself.
But it might also be the switch to a different isochrone algorithm for greater ranges.
Could you provide the docker image/ or git commit hash and the configuration of the ors instance you are running and probably the code/commands to reproduce the used osm file from the scotland.pbf?
Best regards