Check for invalid sources at r=0 in cylindrical coordinates

[oskooi]

Adds a check within fields::add_volume_source src_vol_chunkloop for two three different types of invalid sources at $r=0$ in cylindrical coordinates.

[oskooi]

There is a single failing test in test_adjoint_cyl.py because of an $E_r$ line source extending in $r$ which includes $r=0$ for an $m=1.2$ simulation:

meep/python/tests/test_adjoint_cyl.py

Lines 39 to 42 in b86b476

	source_center = [design_r / 2, 0, -(sz / 2 - dpml + design_z / 2) / 2]
	source_size = mp.Vector3(design_r, 0, 0)
	src = mp.GaussianSource(frequency=fcen, fwidth=fwidth)
	source = [mp.Source(src, component=mp.Er, center=source_center, size=source_size)]

,

The current results therefore do not seem to be accurate because of the $\Delta r$ error that is being introduced by this source.

Perhaps we need to modify this test to exclude $r=0$ in the line source?

cc @mochen4

[stevengj]

This check is not reliable, because the user-specified amplitude function A(x) could be zero at r=0. But I don't know a good way to check that without sampling A at every point along the z axis?

(Also, note that for m=0 the phi and r components should be zero at r=0.)

[mochen4]

As Steven pointed out, the source can be valid if the amplitude is 0 at $r=0$. The test looks good to me, though.

[mochen4]

This check is not reliable, because the user-specified amplitude function A(x) could be zero at r=0. But I don't know a good way to check that without sampling A at every point along the z axis?

Don't we already have A at every point when we create src_vol to add the sources? Can we do the check here?

meep/src/sources.cpp

Line 273 in 601609f

amps_array[idx_vol] = IVEC_LOOP_WEIGHT(s0, s1, e0, e1, 1) * amp * data->A(rel_loc);

[oskooi]

(Also, note that for m=0 the phi and r components should be zero at r=0.)

Added an additional check for this case.

Don't we already have A at every point when we create src_vol to add the sources? Can we do the check here?

Good suggestion. I have moved the check from add_volume_source to src_vol_chunkloop. With this modification, the case of a user-specified amplitude function is also included.

I tested these changes internally with a couple of tests. It seems to work correctly.

[codecov-commenter]

Codecov Report

Merging #2392 (7ae6a22) into master (b86b476) will not change coverage.
The diff coverage is n/a.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

@@           Coverage Diff           @@
##           master    #2392   +/-   ##
=======================================
  Coverage   72.57%   72.57%           
=======================================
  Files          17       17           
  Lines        5166     5166           
=======================================
  Hits         3749     3749           
  Misses       1417     1417           

Impacted Files	Coverage Δ
python/adjoint/filters.py	60.58% <0.00%> (-0.29%)	⬇️
python/geom.py	94.15% <0.00%> (+0.01%)	⬆️