Pepper outputting NaN for powder simulation

General forum for EasySpin: questions, how to's, etc.
Post Reply
JomNarbey
Newbie
Posts: 4
Joined: Fri Mar 29, 2019 3:10 pm

Pepper outputting NaN for powder simulation

Post by JomNarbey »

Hello,

I'm trying to simulate a system containing three collinear exchange coupled S = 7/2 ions, each containing weak ZFS (d only) and a little bit of dipolar. The problem is that pepper runs fine if I only simulate a single crystal orientation (specified by Exp.CrystalOrientation), however when doing a full powder with default value of 'nKnots', I get a bunch of NaN values. Any suggestions for troubleshooting this would be much appreciated!

The code for the spin system is found below.

Code: Select all

S = 7/2;
Sys.S = [S S S];

gx = 2; gy = 2; gz = 2;
g = [gx, gy, gz];
Sys.g = [g; g; g];

Sys.lwpp = 70;

e = 0;
d = convertWave(-.1891)*10^3;
D1 = [d, e];
D3 = D1;
d = convertWave(-.1263)*10^3;
D2 = [d, e]
Sys.D = [D1; D2; D3];

r = 6.761*10^-10;
D13 = 167.9988 * [1 1 -2];
D12 = 1.3446e+03 * [1 1 -2];
D23 = D12;

Sys.eeD = [D12; D13; D23];

J12 = convertWave(.775)*10^3;
J13 = convertWave(.025)*10^3;
J23 = convertWave(.775)*10^3;

Sys.J = [-J12 -J13 -J23];
Stefan Stoll
EasySpin Creator
Posts: 1073
Joined: Mon Jul 21, 2014 10:11 pm
Location: University of Washington

Re: Pepper outputting NaN for powder simulation

Post by Stefan Stoll »

Post a minimal script that reproduces this problem.
JomNarbey
Newbie
Posts: 4
Joined: Fri Mar 29, 2019 3:10 pm

Re: Pepper outputting NaN for powder simulation

Post by JomNarbey »

Thanks for the reply!

Although I'm embarrassed to admit that this appears to be a personal problem... for some reason this works fine on our (much beefier) lab computer but refuses to run properly on my laptop. It is a rather large system to simulate for a powder so it takes quite some time to complete. I'm assuming it's a memory related issue, however matlab doesn't throw any 'out of memory' related errors so I'm not sure what is happening.
Post Reply