Suspension Thermal Noise: Incorrect/Naive

The suspension thermal noise calculation overestimates the noise w.r.t. Matlab gwinc: I think this is because the code does not account for the tapered glass fiber.

changed the description

mentioned in commit 070abb18

mentioned in merge request !18 (merged)

This should be fixed by !18 (merged).

The Voyager bounce mode may appear to be wrong after this change. However, that is due to a hack in Voyager matgwinc, which I'll open a separate bug about on that repo.

On second thought, perhaps this is a separate issue from the one fixed by !18 (merged). I assumed this was some cryo-silicon configuration, but the absence of CD noise on the plot would suggest that it's not. @rana-adhikari can you attach the IFOModel that generated this plot?

the first one is just the matlab gwinc for aLIGO from the iscmodeling repo; the second one is what I get by selecting 'aLIGO' with pygwinc (i.e. what's in the repo now)

Oh, of course. What threw me at first is that we have a CI test case that shows reasonable consistency between pygwinc and matgwinc, for SUS noise in aLIGO.

However, pygwinc's aLIGO IFOModel is apparently out of date. It specifies non-tapered fibers, unlike the one from iscmodeling. There may be other discrepancies as well.

The test case presumably doesn't pick this up because it is feeding the obsolete IFOModel to matgwinc - garbage in garbage out.

mentioned in commit 95bfb2a8

mentioned in merge request !19 (merged)

mentioned in issue #26 (closed)

Should be fixed now.

closed