#360 closed defect - wrong answer (fixed)

H I rydberg level populations

Reported by: Gary J. Ferland Owned by: nobody
Priority: blocker Milestone: C17_branch
Component: ionization convergence Version: trunk
Keywords: Cc:

Description

This is to open a discussion with results for the problem with the departure coefficients in the highest levels. The attached sim is the baseline model.

Tests reveal the following: Honly - b at the highest n, 90, is 9.9 H He only = 6.5 all elements = 3.3

photoionization is much closer to one than collisional ionization. This should not matter for such highly ionized gas. H onlyl Photo T=5 b=5.7 coronal T=5, b =9.9

Attachments (2)

coronalN1.in (447 bytes) - added by Gary J. Ferland 16 months ago.
baseline model, chosen to have worst result
fiducial_dep_coef.jpg (69.6 KB) - added by Gary J. Ferland 16 months ago.
dep coef for baseline model, C10, C13, and trunk

Download all attachments as: .zip

Change History (5)

Changed 16 months ago by Gary J. Ferland

Attachment: coronalN1.in added

baseline model, chosen to have worst result

comment:1 Changed 16 months ago by Gary J. Ferland

Further investigation into the metals dependence. This is b for n=90

H only 9.79 solar metals 3.236 solar metals, no charge transfer, 3.23 He and metals included, but with very low abundances (solar/1e10) 9.79

this shows that the "goodness" introduced by the heavy elements is not due to CX, but that simply having them exist is not enough - they need ~solar abundances to bring . If the problem were due to some sort of memory corruption the case with very low Z would have had similar effects as solar.

Changed 16 months ago by Gary J. Ferland

Attachment: fiducial_dep_coef.jpg added

dep coef for baseline model, C10, C13, and trunk

comment:2 Changed 12 months ago by fguzman

A branch, RydDep?, was created to deal with this problem and several bugs were fixed. It was reintegrated at r11283 and r11367.

Using the coronalN1.in sim, at r11383 on trunk,

6 resolved and 100 collapsed b(n=90)=1.4 and b(n=100)=1.93

6 resolved and 200 collapsed b(n=90)=0.924 and b(n=100)=0.923 and b(n=200)=0.924

If the number of collapsed levels is increased b->1 which enforces the necessity for treating a big number of levels at low densities

There is no difference between solar abundances and h-only so this problem has been fixed.

comment:3 Changed 11 months ago by fguzman

Resolution: fixed
Status: newclosed
Note: See TracTickets for help on using tickets.