Difference between revisions of "2018-04-13"
Jump to navigation
Jump to search
(Created page with "'''Director - M. Spata'''<br> ''No report.'' Acting Accelerator Division Head ---- '''Accelerator Physics - T. Satogata'''<br> ''No report.'' ---- '''Accelerator R &...") |
|||
| Line 82: | Line 82: | ||
**JSPEC Development | **JSPEC Development | ||
*:-JSPEC GUI starts to obtain outside users. | *:-JSPEC GUI starts to obtain outside users. | ||
| − | It was reported to us that JSPEC cannot read some MAD-X tfs files provided by Dr. Steck from GSI. I revised the lattice file parser to read these files. A check of the element position is also added to the parser. If the position of one element is the same as that of the previous element, this element will not be saved into the lattice. In this way we avoid redundant elements and redundant calculation for IBS rate. For the tfs file from Dr. Steck, the element number is reduced by 50% after this redundant element check is applied. | + | *:-It was reported to us that JSPEC cannot read some MAD-X tfs files provided by Dr. Steck from GSI. I revised the lattice file parser to read these files. A check of the element position is also added to the parser. If the position of one element is the same as that of the previous element, this element will not be saved into the lattice. In this way we avoid redundant elements and redundant calculation for IBS rate. For the tfs file from Dr. Steck, the element number is reduced by 50% after this redundant element check is applied. |
*:-It is also reported to us that JSPEC generates NaNs in some simulations. I checked the input file and output file and found that the problem is due to very large product of the time step and the expansion rate, due to which the coordinates of some particle go out of the range of the double floating point number. JSPEC code is revised to prevent this from happening. If a coordinate is less than 1e-60, some computations will not be carried out and the coordinate will not be updated. However, to make the simulation valid in physics, it is necessary to make sure the product of the time step and the expansion rate is not too large. | *:-It is also reported to us that JSPEC generates NaNs in some simulations. I checked the input file and output file and found that the problem is due to very large product of the time step and the expansion rate, due to which the coordinates of some particle go out of the range of the double floating point number. JSPEC code is revised to prevent this from happening. If a coordinate is less than 1e-60, some computations will not be carried out and the coordinate will not be updated. However, to make the simulation valid in physics, it is necessary to make sure the product of the time step and the expansion rate is not too large. | ||
*:-I am integrating the turn-by-turn model and the RF model to the JSPEC release. This is needed for IMP data experiment and JSPEC benchmarking with the IMP data. This has not been finished. | *:-I am integrating the turn-by-turn model and the RF model to the JSPEC release. This is needed for IMP data experiment and JSPEC benchmarking with the IMP data. This has not been finished. | ||
**JLEIC Cooling Simulation | **JLEIC Cooling Simulation | ||
*:-Simulations were carried out to check what we can achieve without cooling at the cooler for 100 GeV proton beam (44.7 GeV CM energy). I tried to use both round beam and flat beam and assume either the constant proton bunch length with varying momentum spread or both constant proton bunch length and constant vary momentum. The results are summarized in Table 1. If we use a round beam, we see severe reduction of proton beam current without dispersion at the cooler. Using a flat beam helps us, because the charge density increases for a given total charge number and hence the cooling is stronger. For fixed bunch length with varying momentum spread with flat beam cooling, we still see a large reduction of proton current. But if we assume the momentum spread is also constant, there is almost no need to reduce the proton beam current. The flat electron beam provides enough cooling even without dispersion at the cooler. But this means we are able to introduce longitudinal heating to compensate the cooling and maintain the momentum spread. | *:-Simulations were carried out to check what we can achieve without cooling at the cooler for 100 GeV proton beam (44.7 GeV CM energy). I tried to use both round beam and flat beam and assume either the constant proton bunch length with varying momentum spread or both constant proton bunch length and constant vary momentum. The results are summarized in Table 1. If we use a round beam, we see severe reduction of proton beam current without dispersion at the cooler. Using a flat beam helps us, because the charge density increases for a given total charge number and hence the cooling is stronger. For fixed bunch length with varying momentum spread with flat beam cooling, we still see a large reduction of proton current. But if we assume the momentum spread is also constant, there is almost no need to reduce the proton beam current. The flat electron beam provides enough cooling even without dispersion at the cooler. But this means we are able to introduce longitudinal heating to compensate the cooling and maintain the momentum spread. | ||
| − | Table 1. Ion beam current in equilibrium (% of design parameter) | + | Table 1. Ion beam current in equilibrium (% of design parameter)<br> |
| − | + | Round beam / Flat beam<br> | |
| − | + | Fixed bunch length, varying dp/p 40% / 56%<br> | |
| − | + | Fixed bunch length, fixed dp/p 58% / 98%<br> | |
| − | Round beam | + | <br> |
| − | |||
| − | |||
| − | |||
| − | Flat beam | ||
| − | Fixed bunch length, varying dp/p | ||
| − | |||
| − | |||
| − | |||
| − | 40% | ||
| − | |||
| − | |||
| − | |||
| − | 56% | ||
| − | Fixed bunch length, fixed dp/p | ||
| − | |||
| − | |||
| − | |||
| − | 58% | ||
| − | |||
| − | |||
| − | |||
| − | 98% | ||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
---- | ---- | ||
Revision as of 11:06, 16 April 2018
Director - M. Spata
No report. Acting Accelerator Division Head
Accelerator Physics - T. Satogata
No report.
Accelerator R & D - Y. Zhang
No report.
Computational Physics - Y. Roblin
Kirsten Deitrick:
- Mar. 31 - Apr. 13
- Cornell Straight Merger:
- -have made 3D field map out of truncated 1D field map for coil pair
- -run for straight merger experiment
- -run for extended amplitude
- -prepping 3D field map of fundamental + third for comparison
- CSR:
- -combine parameter sweep results, get nominal operating parameters
- -run CSR+SC+shielding for Chris
- -attempt to fix segmentation fault problem
- eCloud:
- -can successfully run PyECLOUD on non-JLab laptop; working to get devl79 install functioning
- -can now run on devl79; beginning to benchmark against Shahid's work (for reference)
- Impedance:
- -scheduled next meeting for May 14th at 10
- Meetings:
- -CASA
- -Cooler
- Misc:
- -put in Linac'18 abstracts (invited talk, Straight Merger, Magnetized injector, eCloud)
- -SL 4/5-4/6, 4/12
- Apr. 14 - Apr. 27
- Magnetized Beam:
- -fix booster in both options
- -fix asymmetries in FEL
- -start scanning for charges/field/spot size
- Cornell/Straight Merger:
- -run for fundamental + third
- -run for fundamental + third using magnetized beam
- CSR:
- -attempt to fix segmentation fault problem
- eCloud:
- -beginning to benchmark against Shahid's work
- -run with current parameters
- Meetings:
- -Cooler
- -JLEIC
Yves Roblin:
- Mar. 30 - Apr. 13
- CEBAF:
- -model team. established a list of tasks to do in near future
- -model team : presented to BTEAM, discussion about prioritization.
- -technote writing/data analysis
- JLEIC:
- -send FOA letter to Manouchehr , setup budget act code .
- -read papers on multivariate polynomial evaluation (in order to use this for fast calculations of higher order maps in BB3D)
- OTHER:
- -interview panel for radiation physicist
- -finalized PERLE optics in ELEGANT (all passes + deceleration)
- -manager training.
- Apr. 14 - Apr. 27
- JLEIC:
- -review /adjust xfer line from CEBAF->Ering
- -prepare for the start of the FOA, plan activities, gather up docs and tools
- CEBAF:
- -machine support
- -BTEAM coordination (04/17 and 04/24)
- -planning/preparing for beam studies (autosteer, bunch length meas, etc..)
He Zhang:
- Mar. 31 - Apr. 13
- JSPEC Development
- -JSPEC GUI starts to obtain outside users.
- -It was reported to us that JSPEC cannot read some MAD-X tfs files provided by Dr. Steck from GSI. I revised the lattice file parser to read these files. A check of the element position is also added to the parser. If the position of one element is the same as that of the previous element, this element will not be saved into the lattice. In this way we avoid redundant elements and redundant calculation for IBS rate. For the tfs file from Dr. Steck, the element number is reduced by 50% after this redundant element check is applied.
- -It is also reported to us that JSPEC generates NaNs in some simulations. I checked the input file and output file and found that the problem is due to very large product of the time step and the expansion rate, due to which the coordinates of some particle go out of the range of the double floating point number. JSPEC code is revised to prevent this from happening. If a coordinate is less than 1e-60, some computations will not be carried out and the coordinate will not be updated. However, to make the simulation valid in physics, it is necessary to make sure the product of the time step and the expansion rate is not too large.
- -I am integrating the turn-by-turn model and the RF model to the JSPEC release. This is needed for IMP data experiment and JSPEC benchmarking with the IMP data. This has not been finished.
- JLEIC Cooling Simulation
- -Simulations were carried out to check what we can achieve without cooling at the cooler for 100 GeV proton beam (44.7 GeV CM energy). I tried to use both round beam and flat beam and assume either the constant proton bunch length with varying momentum spread or both constant proton bunch length and constant vary momentum. The results are summarized in Table 1. If we use a round beam, we see severe reduction of proton beam current without dispersion at the cooler. Using a flat beam helps us, because the charge density increases for a given total charge number and hence the cooling is stronger. For fixed bunch length with varying momentum spread with flat beam cooling, we still see a large reduction of proton current. But if we assume the momentum spread is also constant, there is almost no need to reduce the proton beam current. The flat electron beam provides enough cooling even without dispersion at the cooler. But this means we are able to introduce longitudinal heating to compensate the cooling and maintain the momentum spread.
Table 1. Ion beam current in equilibrium (% of design parameter)
Round beam / Flat beam
Fixed bunch length, varying dp/p 40% / 56%
Fixed bunch length, fixed dp/p 58% / 98%
Diagnostic Development - K. Jordan
No report.
LERF - S. Benson
No report.