I have a question regarding the bunch frequency setting. I have read some previous discussions on the forum, but I am still not fully clear about it.
My current understanding is that this parameter is related to the space-charge calculation. Therefore, I would like to ask:
1.Are there any recommended guidelines or rules of thumb for choosing this parameter?
2.Can it be determined based on certain beam properties (such as beam size, energy, etc.) to ensure a reasonable range?
Looking forward to your reply!
Question on bunch frequency [SOLVED]
Re: Question on bunch frequency
This question really only arises in the case of DC beams; otherwise, you need to specify the frequency of your RF structures.
For DC beams, for meshing reasons, it is preferable to choose the frequency so that the longitudinal size in metres is of the same order of magnitude as the transverse dimensions.
Regards,
Didier
For DC beams, for meshing reasons, it is preferable to choose the frequency so that the longitudinal size in metres is of the same order of magnitude as the transverse dimensions.
Regards,
Didier
Re: Question on bunch frequency
Dear Didier,
After carefully reviewing the TraceWin user manual and related materials, I would like to confirm my understanding of the space charge settings in Partran and ask a few questions.
My current understanding is as follows:
1.Calculation step / Space-charge step**
These parameters define the discretization along the beamline, i.e., how many steps per meter (or per βλ) are used for particle tracking, and how frequently the space-charge force (SC kick) is applied.
2.Partran space charge options**
These define the 3D mesh used for space-charge calculations, including the number of cells and the transverse coverage (e.g., mesh/rms).
3.For DC beams**
The bunch frequency is not a physical bunching frequency but rather a numerical parameter used to define the longitudinal size of the computation domain. In the bunch frame, this corresponds to: Lz∗=γβc/f
Based on this understanding, I have the following questions:
1. Recommended values for steps
Are there typical or recommended ranges for the calculation step and space-charge step (e.g., steps per meter) that ensure sufficient accuracy?
2. Choice of energy for bunch frequency
Since the beam energy (and thus β) varies along the beamline and space-charge effects are strongest at low energy, should the bunch frequency be determined based on the low-energy section rather than the high-energy section?
3. Choice of transverse scale for matching (Lz∗)
According to the user manual, setting mesh/rms = 5 implies that the transverse mesh covers approximately ±5×rms, so the transverse computation domain is:L⊥∗≈2∗5∗(rms)x
In this context, my understanding is:The longitudinal size (L_z^*) should be chosen to be of the same order as the full transverse computation domain (defined by the mesh), rather than the rms beam size itself, in order to maintain a roughly isotropic 3D mesh (aspect ratio close to 1).
Additionally, since the beam size evolves along the beamline (e.g., initial rms_x ≈ 2 mm but potentially increasing significantly in the low-energy region), I would like to ask:
3.1 Should the transverse scale be based directly on the mesh definition (e.g., ±5σ)?
3.2 Or should it be based on the maximum beam envelope along the beamline?
Thank you very much for your time and guidance.
Best regards,
ZY Song
After carefully reviewing the TraceWin user manual and related materials, I would like to confirm my understanding of the space charge settings in Partran and ask a few questions.
My current understanding is as follows:
1.Calculation step / Space-charge step**
These parameters define the discretization along the beamline, i.e., how many steps per meter (or per βλ) are used for particle tracking, and how frequently the space-charge force (SC kick) is applied.
2.Partran space charge options**
These define the 3D mesh used for space-charge calculations, including the number of cells and the transverse coverage (e.g., mesh/rms).
3.For DC beams**
The bunch frequency is not a physical bunching frequency but rather a numerical parameter used to define the longitudinal size of the computation domain. In the bunch frame, this corresponds to: Lz∗=γβc/f
Based on this understanding, I have the following questions:
1. Recommended values for steps
Are there typical or recommended ranges for the calculation step and space-charge step (e.g., steps per meter) that ensure sufficient accuracy?
2. Choice of energy for bunch frequency
Since the beam energy (and thus β) varies along the beamline and space-charge effects are strongest at low energy, should the bunch frequency be determined based on the low-energy section rather than the high-energy section?
3. Choice of transverse scale for matching (Lz∗)
According to the user manual, setting mesh/rms = 5 implies that the transverse mesh covers approximately ±5×rms, so the transverse computation domain is:L⊥∗≈2∗5∗(rms)x
In this context, my understanding is:The longitudinal size (L_z^*) should be chosen to be of the same order as the full transverse computation domain (defined by the mesh), rather than the rms beam size itself, in order to maintain a roughly isotropic 3D mesh (aspect ratio close to 1).
Additionally, since the beam size evolves along the beamline (e.g., initial rms_x ≈ 2 mm but potentially increasing significantly in the low-energy region), I would like to ask:
3.1 Should the transverse scale be based directly on the mesh definition (e.g., ±5σ)?
3.2 Or should it be based on the maximum beam envelope along the beamline?
Thank you very much for your time and guidance.
Best regards,
ZY Song
Re: Question on bunch frequency [SOLVED]
1) in maunal chapter "Space Charge in Partran simulation", you can find some links to 2 articles about this specific point. Generally speaking, I tend to start with a coarse step and reduce it until I no longer see any significant changes in the results.
2) If you have a high-energy section, it means your beam is bunched at a frequency definie by the first RF structure, so you need to use this frequency.
3) The point is not to keep the short-term to long-term ratio close to 1, but to ensure it is not 100, 1000 or more; the space-charge routine will work but is not optimised for such huge discrepancies.
3.1&2) At each calculation step, the mesh is recalculated based on ±5σ
Regards,
Didier
2) If you have a high-energy section, it means your beam is bunched at a frequency definie by the first RF structure, so you need to use this frequency.
3) The point is not to keep the short-term to long-term ratio close to 1, but to ensure it is not 100, 1000 or more; the space-charge routine will work but is not optimised for such huge discrepancies.
3.1&2) At each calculation step, the mesh is recalculated based on ±5σ
Regards,
Didier