Filling Scheme Editor
|
|
Filling Scheme Editor |
|
In general all units are BX. The BX counting starts with 0 corresponding ot the first bucket where a bunch can be injected (this is bucket 1 according to LHC conventions). The last BX of the orbit is No. 3563. The entries "Injection spacing" and "Batch spacing" however, are in units of ns. The numbers must be multiples of 25ns. This unit has been chosen since these numbers are always given in ns, when they are mentioned by LHC experts.
The graphical display is devided into 2 distinct displays. The top one shows the position of the batches in the LHC orbit. At the end of the orbit the AGK is marked in gray and the Abort Gap in a redish color. In Beam 1 there is an overlay of shaded boxes for LHCb (upper row of boxes) and ALICE (lower row of boxes). These boxes indicate the positions where bunches in beam 1 collide in the respective experiment with bunches in beam 2. They are an image of the batches in beam 2, shifted by 1/4th of the orbit (and in the case of LHCb by 3 additional BX). This display is useful when trying to optimise a filling scheme for a maximum number of collisions in LHCb. Injections which are generated with the "intermediate" SPS cycle are drawn in two pieces vertically, i.e. there is a bit of white space visible in the vertical centre of the injection. There is a button to toggle the injection type from nominal to intermediate for selected injections. The lower display shows the bunch classes introduced by John Jowett: Each beam (B1 top, B2 bottom) is divided into 3 horizontal sections. The top one is filled with the green line if the bunch collides in IP8. The middle one is filled with a red line if the bunch collides in IP2 and the bottom one is filled with a blue line if the bunch collides in IP1&5. Black lines indicate bunches which do not collide anywhere. This class display is very useful when designing filling schemes for Heavy Ion runs or Proton Ion runs.
Below the graphical display some tables summarise important information for the scheme. Also the bunch classes are listed. The number before the colon represents the class id and the number after the colon indicates how many bunches of the corresponding beam are in this class. The class id is coded as a three bit number where bit 0 indicates a collision in IP1&5, bit 1 indicates a collision in IP 2 and bit 2 indicates a collision in IP8.
Below this, a table indicates the positions of the injections and the spaces between them in units of BX. Also indicated are the number of occupied BX after the AGK and the number of BX which could be filled with bunches but are not used in the current scheme. (This number does NOT include the empty BX which are required between injections or batches to allow for the kicker magnets to ramp: "Injection Spacing" and "Batch spacing")
Below this you can find strings which can be used to copy the filling scheme into the the filling scheme editor of LHC. For every type of injection there are comma separated lists which can be "coppy-pasted" into the editor. This allows to copy a scheme within a few minutes into the LHC database ready for use in the accelerator complex.
At the bottom of the page information no phase shifts in the RF full detuning scheme is being displayed.
There is a diagram for IP 1&5, one for IP2 and one for IP8. For each
bunch crossing id (x-axis) the plot shows the expected phase (in ps) relative to the LHC clock for
beam 1 and beam 2 (blue and red), the shift of the collision time wrt to
the LHC clock at the IP (in orange) and the displacement of the interaction
point relative to the "nominal" interaction point (in cm: right y-axis).
Since the phase shifts in IP 1&5 are symmetric for beam 1 and 2, no shift of the
interaction point is expected here.
The plots contain gray and white vertical zones. White zones contain colliding
bunches whereas gray zones do not contain colliding bunches at the IP.
Since the phase shift depend on the bunch intensity, the HV in the cavities, and the bunch length, these values can be
adjusted in the form at the top of the plots. Reasonable default values are inserted
per default.
When a new scheme is loaded into the editor, the phase shift information is automatically
calculated. When the scheme is edited the information can be refreshed by clicking on
"estimate phase shifts" or, if "Continuous update" is checked, it will be updated
automatically. Since the automatic updates might slow down the editor during the
process of editing a scheme, the "Continuous update" box is not checked by default.
In order to fill the orbit first the correct AGK has to be chosen or entered.
Then the other parameters (Injection spacing, Bunch spacing, No. Batches, Batch length, Batch spacing) are adjusted with help of the entries on the
top of the page. Clicking on the "Pre-fill" button will fill the LHC. Alternatively press the "Start new scheme" button
and fill the orbit injection by injection (see documentation below).
To prefill the machine different algorithms together with some parameters can be choses from a drop down list.
Single injections can be added to the scheme by adjusting the parameters in the relevant entries and then clicking on "Add injection". If then the mouse is moved over the graphical representation of beam 1 or beam 2, a rectangle appears in the graphcis indicating the position of the injection in the orbit. If the rectangle is green it the injection can be placed at the current position in the orbit by clicking the mouse button. If the rectangle is black there is not enough space to insert the injection. If the check box "LHCb ripple through" is selected, the editor tries to inject the injection in both beams in a way that bunches collide in ATLAS and CMS and then works towards the beginning of the orbit and tries to inject injections in both beams such that they collide in ATLAS and CMS and in LHCb. Be aware that this algorithm only works from the selected injection point towards the beginning of the orbit.
Clicking on Advanced Injections changes some of the parameters to choose from. You should have put your user name into the User entry before clicking on the button, since the editor will load all your previously defined batches and injections from your account. You can click on the "Show editors" button to bring up an editor which allows you to define arbitrary batches and injections. Batches are defined by dragging the mouse over the Batch editor graphics. Alternatively you can load and subsequently save batches by editing or pasting a batch-string into the corresponding entry field. Clicking (and click-dragging) over the graphical field will toggle the contents of the buckets (green fields contain a bunch, empty fields contain no bunch).
Once you have defined and saved the necessary batches you can build up injections in the lower part of the editor. Choose the batches from the drop down list and then click on the graphical display to add the batch to the injection. Batches can be moved around by click-draggin on them. The minimal batch distance is being enforced by the editor. (This number can be changed but must be a multiple of 25ns.) Similar to the batch editor, an injection string can be edited or pasted into the entry field and loaded into the editor. In this case the injection will be saved as a single batch (which probably does not correspond to what happens in the 'real' Injectors but for creating the filling schemes this should not make any difference.) To delete a batch from an injection you can click on the batch while pressing down the Shift key.
To do the fine tuning of the scheme it is useful to shift injections or groups of injections around. For this an injection or a group of injections is selected by clicking on the injections or by click-dragging the mouse over the region of injections to be selected (this can include injections from beam 1 and 2). On top of the graphics there will be an indication on how many bx the selection can be shifted in either direction. To shift the selection press the "Shift" key and at the same time click-drag the mouse WITHIN the graphics. The mouse will move faster than the group of selected injections allowing for a controlled shift of single BXs. The "Shift speed" slider allows to tune this "shift-speed" in order to adapt the interface to your mouse moving skills and mouse quality. You can also shift the selection in steps of 1 BX by using the right and left arrow keys. The number of collisions in the various IPs is updated in real time during the shift.
To save a scheme press Ctrl-S. You must select a folder name where you want to save the scheme in. (You can also create folders in the same dialog). By clicking on "Scheme name" a name for you scheme will be suggested. The dialog also allows you to download a tgz file with all schemes of your user-account for backup purposes. You should do this in case you need your schemes for later reference (the data is not backed up on the server, and, as you know, is worldwide accessible and modifiable by anybody) (Contact the LPC in case you need to play your backup back into the system).
Ctrl-L will pop up the dialog to load or delete a scheme. By double clicking on a scheme it will be loaded into the editor. By right-clicking on it it will be deleted. In this pop up you can also drag schemes (or folders) into new folders to re-arrange your schemes.
|
|
|
|
|
Download Bunch Arrays for both beams in JSON format
Conventions used: positive delays (in ps) denote a delay added to the reference LHC clock (i.e. the bunch comes later to the IP). Positive z-direction points into the direction of Beam 1 (i.e. clockwise).
| : | MV | |
| : | ||
| : | ns | |
| Download detuning data as JSON |