Fix: Correction to laser pulse formular for few-cycle wavepackets #875
Conversation
|
@stetie Great job. 👍 @ax3l Please add @stetie to the developer list. |
|
added to developer team, welcome and thx for the first pull request! 🚀 just |
|
P.S.: pls reformat the style of your commit message a little :) |
Correction to laser pulse formular for few-cycle wavepackets (same as for ComputationalRadiationPhysics#851)
b07c1fd to
ebb77a2
Compare
| @@ -56,6 +59,9 @@ HINLINE float3_X laserLongitudinal(uint32_t currentStep, float_X& phase) | |||
| const double endUpramp = -0.5 * LASER_NOFOCUS_CONSTANT; | |||
| const double startDownramp = 0.5 * LASER_NOFOCUS_CONSTANT; | |||
|
|
|||
| const double tau = PULSE_LENGTH / sqrt(2.0); | |||
There was a problem hiding this comment.
gentle reminder to your spotted correction. same as in plane wave: should be
const double tau = PULSE_LENGTH * sqrt(2.0);|
Sinus pulse: sum/amax = 1e-2 is OK |
|
Cosinus pulse: sum/amax = 1.5e-2 is OK |
Fix: Correction to laser pulse formular for few-cycle wavepackets
|
weird offset in the envelope... hm. what are the cells per lambda and how many steps did you propagate? -> 90/lambda_0 and 2k steps is this just a small offset due to dispersion and the envelope is from the analytical formula (-> yes)? |
Similiar to #851, this corrects the laser pulse formular to account for extra terms arrising from deriving the formular from the vector potential. It ensures the integration along the propagation direction is 0. This is important for few-cycle laser pulses.