bug in changeWindowSize ?

[Gabor]

Hi, I used
[new_E_ftfxfy_o,new_E_ftfxfy_e,New_tWindow,New_xWindow,New_yWindow]= ...
    changeWindowSize(obj,E_ftfxfy_o,E_ftfxfy_e,...
    'xWindow',0.002,'yWindow',0.002,'tWindow',2e-12,'Nx',1024,'Ny',64,'Nt',1024);
to reduce x&yWindow size from 0.01m to 0.002m while increase tWindow size from 5e-15s to 2e-12s.
The result for the ordinary pol. is correct.
Before (Itx)

After (Itx)

While in the extraordinary pol.
Before (Itx)

After (Itx)


Could you please check it? Thanks.

[tinolang]

I need more information about the input pulses (wavelength) and where your frequency window starts (shiftFequWindow). Is it possible that the extra ordinary pulse is shorter in wavelength, e.g. the second harmonic of the ordinary?  Maybe this pulse then is not sampled in temporal frequency space any more. Please check what is the shortest sampled wavelength you get with the new Nt and tWindow parameters:

>> obj=chi3D('Nt',1024,'tWindow',2e-12,'shiftFreqWindow',0);
computation time constants SHG example 0.08515 s
>> obj.constVect.lambda(end)

ans =

   5.8610e-07

If this is not solving the problem please send me a simple example script which helps me to reproduce the issue.

[Gabor]

Yes, This was the problem. The eo (400nm) is 2nd harmonic of o (800nm).
The original time sampling was:
'Nt',512, 'tWindow',5e-13,'shiftFreqWindow',0,
and it gives:
obj.constVect.lambda(end)
ans =   2.9334e-07
I have to set: 'Nt',2048, to match the new 'tWindow',2e-12.
Thank you.