The purpose of this procedure is to:

1. Identify the correction elements breakpoints, obtain orbit data at that breakpoint, and determine whether a smooth is required.
2. Set the DP/P at a specific breakpoint to keep any global offsets to the correction elements to a minimum.
3. Understand which set of desired positions to use for smoothing a specific breakpoint.
4. Understand how why correction elements should track their reference during smoothing.

    In order to determine whether a smooth is required, BPM data must be gathered for the correct energy breakpoint. These energy breakpoints can be found on I50  under C453 and then GI(I), (figure 1). The breakpoints will be listed in momentum and are common to all ramps. Once the correction element breakpoints are understood, the next step is to acquire BPM data for the specific time in a cycle when the breakpoint is active. This can be be accomplished by using the TPI feature on I2 for the ramp being smoothed and setting this time into the BPM display frame timer. Another method is to set the BPM profile timers to span the ramp's beam time. One can then pick the appropriate data out of the profiles acquired. Using the profile frames in this manner also allows one to look at the orbit in the machine between breakpoints. Finally, the simplest method is to use I50 to set the profile frames at the breakpoints. The program will set the BPM profile timer to acquire one frame at each of the correction element breakpoints, (figure 2). The BPM flash frame generally should not be used for smoothing since it only acquires a single turn of data. The display and profile frames use an averaging technique and return data better suited for this procedure.

    Once the BPM data for a breakpoint has be acquired, look at the RMS values on the BPM display. If these values are much greater than 6.5 horizontally or 2.5 vertically, the closed orbit should be considered "wretched" and one should smooth. Another way to determine whether a smooth should be done is to look at the magnitude of the oscillations around the ring on the BPM display frame. If the oscillations are greater than +/- 2mm, one should smooth.

 

Breakpoints window                                      Figure 1

BPM setup window                                              Figure 2

 Click  any image to enlarge

    Before smoothing at the injection energy the proper momentum must be set as specified in the  injection tuning procedure . Once the correct momentum has been set, smoothing at the injection energy is usually a simple matter of running the program. Upon entering I50, the program will select the low energy desired positions. It is a good idea however to check this file before performing a smooth to make sure the file is in the state it should be ( i.e. detector locations are not masked or the file is corrupted for some reason). There are several situations which can occur in the accelerator that will confuse the smoothing program. First,bad BPM data, usually NO BEAM, is fairly obvious. Second, stale BPM data from a processor not receiving a prepare for beam. Third,broken or non tracking correction elements can be a problem because the smoothing program thinks it has a corrector to work with when in reality it does not. If one smoothes when any of these situations occur and the smoothing program has not been informed, the result will be a non local orbit distortion which will probably make the orbit worse. Another problem which can occur as a result of orbit smoothing is to introduce a DC offset into the correctors. Figure 3 shows an example of this problem. The momentum correction feature in I50 is supposed to prevent this problem from happening. However, after many smoothes the offset tends to add up to a significant value. Before smoothing look for this DC offset problem, (figure 3). If this problem is present, correct it using the  Removing DC offset procedure .
 

Example of correction element settings with DC offset                                            figure 3

 

    Smoothing the orbit up the ramp is similar to injection smoothing but the mechanics is a bit different. It is important to check the DP/P of the orbit and center the beam before smoothing. Centering the beam before smoothing will help keep the DC offsets that develop in the correctors to a minimum. Unless a specified non zero DP/P excises, one should center the beam before smoothing. At energies above injection this is accomplished with a three bump at the 530 location. For example, I:H530[x]:3 where x= the breakpoint number starting with 0 being injection. The desired position files are different at various energies throughout the Main Injector ramp. The injection file (file 1) is used up to 25 Gev. File 2 is used for 25Gev through 75 Gev. File 3 is used for 120 Gev and file 4 is used for 150 Gev.  Before sending out the new settings for the correctors, check for the DC offset. If this problem is present, correct it using the  Removing DC off set procedure . At the higher energies it is also important to look at the absolute value of the settings. The correctors are only capable of 18A and tracking errors or RMS trips will occur if the settings are not reasonable. After sending out the new settings check and fix any tracking errors before proceeding further. Tracking errors can usually be fixed by plotting the corrector with the problem against its reference. A three bump at the breakpoint of interest can usually be used to fix the problem . Tracking problems are usually associated with an improper desired position file and three bumping the corrector back into tracking will change the position at that location.

     When smoothing at flattop one needs to be careful of conditions that occur as a result of LLRF states. Specifically, at the flattop energies Fset commands in the LLRF may cause changes to the DP/P. Cogging and coalescing can also cause confusion. For these reasons it is usually best to smooth at these energies with study cycles that keep the LLRF feedback loops active.

    DC offsets in the correction elements cause problems for the machine as they aid or oppose the bend field. This creates a situation where it becomes difficult to specify the energy the accelerator. A result of this problem is an uncalibrated tune table. If the correction elements DC offset changes with time, the tune table will follow this offset. This situation makes it difficult to maintain stability, reproducibility, and performance over time.

     Figure 3  is an example of this DC offset. Removing this offset can be a simple matter of smoothing with momentum correction off or it can require readjusting settings of other devices depending on the magnitude and breakpoint energy of the offset. To remove this offset, set the DP/P of the machine to a non zero value and smooth with the momentum correction off. For example, if the offset is in the positive direction, as in figure 3, put the beam in the negative DP/P direction, the inside of the machine, and smooth. Before sending out the corrections look at the settings to make sure they are correct. For small offset this operation can be done in a single pass. For larger offsets it will be necessary to iterate and possibly retune between steps.

    At injection the DP/P is controlled by I:MBOFF as specified in the  injection tuning procedure . At energies above injection, the DP/P is controlled with the three bump,  I:H530[x]:3 where x= the breakpoint number starting with 0 being injection.