search

RadioAstronomySoftwareGroup / pyuvsim

A ultra-high precision package for simulating radio interferometers in python on compute clusters.
https://pyuvsim.readthedocs.io/en/latest/
BSD 3-Clause "New" or "Revised" License
42 stars 7 forks source link

Point sources are not points #21

Closed jpober closed 6 years ago

jpober commented 6 years ago

Simulating an array (in this case, MWA) with three point sources gives blobs of flux density. Things don't quite seem to line up coherently.

screenshot 2018-05-24 16 28 38
dannyjacobs commented 6 years ago

lmns seem to be the same on all baselines uvws seems to be ok, but the triangle test was only short baselines. @aelanman will run a triangle baseline test with ~1km baselines the only other thing in the visibility equation is the beam. @bhazelton is adding a "tophat" beam object to remove the complexity of the hera beam from the equation

bhazelton commented 6 years ago

We tried the tophat beam and saw no improvement. Now we're considering whether it's a problem with pyuvdata's phasing. We're going to try running on the uvdata_time_array_fix branch which has better phasing.

jpober commented 6 years ago

Fixed. It was a minus sign in the fringe equation!