2012
DOI: 10.1051/epjconf/20123204017
|View full text |Cite
|
Sign up to set email alerts
|

On the numerical scheme employed in gyrotron interaction simulations

Abstract: Abstract. We report on the influence of the numerical scheme employed in gyrotron interaction simulations. Results obtained with the Crank-Nicolson scheme are compared with those obtained with the Backward Time -Centred Space (BTCS) fully implicit scheme. We present realistic cases where, for discretisation parameters in the range usually used in gyrotron simulations, the results can be very different. Hence, the numerical scheme used can be responsible for obscuring the underlying physics if its convergence i… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
5
0

Year Published

2015
2015
2015
2015

Publication Types

Select...
2

Relationship

2
0

Authors

Journals

citations
Cited by 2 publications
(5 citation statements)
references
References 8 publications
0
5
0
Order By: Relevance
“…(3), boundary condition (4), and trajectory approach are well justified. However, in the simulated dynamic ACI cases, [13][14][15][16][17][18][19][20][21][22] the presence of two frequencies (dominant and parasitic) at the same mode leads to a beating appearing at the complex field envelope A, which broadens the spectrum of A and results in j@A/@tj/(x 0 jAj) $ 10 À1 . This weaker form of (2) may still permit the parabolic equation (3), but the validity of the simplified single-frequency boundary condition (4) and of the trajectory approach become questionable.…”
Section: Modeling Approaches and Numerical Codesmentioning
confidence: 99%
See 2 more Smart Citations
“…(3), boundary condition (4), and trajectory approach are well justified. However, in the simulated dynamic ACI cases, [13][14][15][16][17][18][19][20][21][22] the presence of two frequencies (dominant and parasitic) at the same mode leads to a beating appearing at the complex field envelope A, which broadens the spectrum of A and results in j@A/@tj/(x 0 jAj) $ 10 À1 . This weaker form of (2) may still permit the parabolic equation (3), but the validity of the simplified single-frequency boundary condition (4) and of the trajectory approach become questionable.…”
Section: Modeling Approaches and Numerical Codesmentioning
confidence: 99%
“…In order for this to be true, however, the possibility that the used PIC codes have some kind of inherent numerical dissipation that suppresses ACI in general (e.g., as discussed in Ref. 15), should be excluded. This is done in the next investigated case.…”
Section: -Mode Resonatormentioning
confidence: 99%
See 1 more Smart Citation
“…approximately 0.5-1 % as shown in [1,2] and references therein. The proper description of the wave-particle interaction in the commonly used gyro-averaged slow time scale models [3][4][5] requires to properly define the radiation boundary conditions at the extremities of the considered interaction space. Compared to the seminal paper by Ginzburg et al [6], in which the non-reflecting frequency-independent boundary condition was derived and used in the case of a simplified interaction space with constant radius, in this paper, the interaction space is extended such to include the real gyrotron cavity as well as the uptaper and a simplified launcher following the cavity.…”
mentioning
confidence: 99%
“…Fig 4. The reflection coefficient R(Ω) in the vicinity of Ω 1 = (ω 1 − ω 0 )/ω 0 = −0.06 for different time steps ω 0 Δt and k ⊥ /k 0 = 0.9…”
mentioning
confidence: 99%