|
|
Dear Axel,
thank you for your replay and clarification!
Indeed I'm using release-0.1.2 version. So just to double check let me repeat your answer.
So PULSE_LENGHT in the laserConfig.param file is the intensity sigma. Which means that for 25fs of intensity FWHM, PULSE_LENGHT = 25fs/k, where k = 2*sqrt(2*log(2))=2.3548. So PULSE_LENGHT = 10.616fs. This matches your answer and comments in the code. However, there is a line
60 const double PULSE_LENGTH_SI = 5.0e-15; //25.0e-15 / 1.17741;
which is a bit confusing, cause I don't understand the number 1.1774 (and 25.0e-15 / 1.17741 will not give 5.0e-15).
I'm sorry for being picky however I just want to have a clear picture of the initialization procedure and reading the comments is a good way to get familiar with unknown code.
Once again thanks for your support!
Cheers,
Danila.
________________________________________
From: picongpu-users@hzdr.de [picongpu-users@hzdr.de] on behalf of Huebl, Axel [a.huebl@hzdr.de]
Sent: Wednesday, May 04, 2016 11:05 AM
To: picongpu-users@hzdr.de
Subject: Re: [PIConGPU-Users] laserGaussianBeam initialization
Dear Danila,
we cross-checked our implementation and it looks correct.
We will reference the full derivation in our wiki in the next days.
To give you a non-latex explanation for now:
*laserConfig.param*
defined in laserConfig.param you found correctly the variable
"PULSE_LENGHT" which is as described the sigma of a standard gaussian of
the intensity.
(For the experimentalist conversion to FWHM of intensity, see conversion
factor in the comment.)
*laserGaussianBeam.hpp*
In the implementation (great you found it on your own! :) ) we set the
electric field.
the relation of electric field and intensity sigma's are:
sig_Ex^2 = 2 sig_I^2 | * 2
2 sig_Ex^2 = 4 sig_I^2
So instead of converting into sig_Ex by sqrt(2) we just implemented
( 2 sig_I) ^ 2 == 4 sig_I^2
for setting the E-field component.
Cheers,
Axel
On 03.05.2016 19:42, Huebl, Axel wrote:
> Hi Danila,
>
> thanks for the question!
>
> Which version of PIConGPU you are using?
> The latest release, 0.1.2, changed the documentation in the
> laserConfig.param a bit because it was not precise before it.
>
>
> I got your question regarding the temporal envelope (pulse length) of
> the Gaussian Beam. I actually just fixed it for 2D last year but we can
> double check tomorrow if the documentation is up-to-date.
>
> https://github.com/ComputationalRadiationPhysics/picongpu/pull/1090
>
> So far, great you found the implementation yourself! :) I am missing the
> sqrt on the first view, too but let us check tomorrow.
>
>
> Cheers,
> Axel
>
> On 02.05.2016 13:06, Khikhlukha Danila wrote:
>> Hi
>> all,
>>
>>
>>
>> I have a couple of short questions regarding the Gaussian beam
>> initialization.
>>
>>
>> 1. As far as I understood from the code comments
>> (simulation_defines/param/laserConfig.param)
>> the initialization for laser pulse is done based on the sigma of the
>> intensity envelope. However in the code
>> (src/picongpu/include/fields/laserProfiles/laserGaussianBeam.hpp) it
>> looks like PULSE_LENGTH is the sigma of electrical field envelope.
>> I also wasn't able to find any conversion (namely
>> sqrt(2))
>>
>>
>> 2. So at the moment I'm assuming that the initialization of the Gaussian
>> beam is done based on the sigma of an electrical field envelope.
>> So if we consider k = 2*sqrt(2*log(2))=2.3548 as a coefficient between
>> FWHM and sigma of the Gaussian curve
>> then the coefficient between sigma of an electrical field and FWHM of
>> the intensity becomes sqrt(2)/k = 0.6006.
>> However, in comments for laserGaussianBeam it is stated the number
>> 1./1.17741 = 0.8493 and I'm not sure what is it.
>>
>>
>> Can you please double check if comments are up-to-date? Otherwise I got
>> something wrong in the description.
>>
>>
>> Thank
>> you,
>>
>> Danila.
>
--
Axel Huebl
Phone +49 351 260 3582
https://www.hzdr.de/crp
Computational Radiation Physics
Laser Particle Acceleration Division
Helmholtz-Zentrum Dresden - Rossendorf e.V.
Bautzner Landstrasse 400, 01328 Dresden
POB 510119, D-01314 Dresden
Vorstand: Prof. Dr.Dr.h.c. R. Sauerbrey
Prof. Dr.Dr.h.c. P. Joehnk
VR 1693 beim Amtsgericht Dresden
|
|