Procedure for Phase A.7:
 

Step

Activity

Who

Priority

A.7.1

Get Beams into Collision in the X,Y plane

OP

 1

.01

Separator bumps at nominal 0 at all IPs (get settings from best knowledge; beams should be already fairly close).
 

 

.02

Measure beam displacement at the IP using BPMs.

 

 

.03

Adjust beam separation such that the beam 1 and beam 2 difference left/right of the IP is the same. Do this for one IP at the time. [Note 1]

  

 
 .04  Monitor lifetime for all the bunches/empty buckets/abort gap; monitor beam losses.    

.05

"Watch" background.

  

 

.06

  Change mode from ADJUST to STABLE BEAMS (if lifetime and background under control).    

.07

  Start counting delivered luminosity; logging into database (~ Hz).    

A.7.2

Measure and correct longitudinal position

OP/RF

2

.01

  Shift RF phase to monitor the longitudinal position [Note 2].    

A.7.3

Monitor lifetime, beam losses and keep background low and stable (no spikes) [Note 3]

OP

1

A.7.4

Monitor luminosity during the fill provided by the experiments [Note 4]

OP

2

 

 

 

 

Notes on the procedure

 

Note 1. Adjust beam separation such that the beam 1 and beam 2 difference left/right of the IP is the same: measure this with special (beam) directional strip line couplers BPMSW at ~ 21 m L/R from the IP in front of Q1.

Based on BPM information check that:

 

δr (σ)

δx,y (σ)

δx,y (μm) (@ 7 TeV and nominal є)
      b* = 11m b* = 2 m b* = 0.5 m
To see collisions < 2σ < 1.4σ < 133 μm < 44 μm < 23 μm
To optimize lumi and equalize between experiments < 0.5σ < 0.35σ < 33 μm < 11 μm < 6 μm

Table 1: Beam separation values.

At this point, beams should be colliding.

At HERA they also use another method (by B. Holzer): they excite one beam at its tune frequency and observe the spectrum of the other beam. When the beams are close to each other, one can see an increasing amplitude of the excitation frequency in the beam. 

Note 2. In principle not too critical in commissioning. Since first collisions will be without crossing angle and with rather large beta* (11 m), even few ns resolution could be sufficient together with information from the experiments. Once in stable physics this will be monitored by the experiments.

How to detect offsets later? : A new electronic card has been developed. Uses the BPMs around the IP and existing infrastructure and allows to measure the relative beam arrival times with sub ns resolution.

Note 3. List of background sources and how they can be minimized [3,4]:

  1. Beam gas: good vacuum quality in particular around the experiments.

  2. Halo: minimize halo production and maximize cleaning efficiency (well corrected machine, avoid resonances, minimize heating/vibrations) careful transverse feedback, orbit feedback, etc. Optimize lifetime and minimize emittance growth.

  3. Physics Collisions: they have side effects that we can minimize: avoid small offsets, e.g. 0.1 σ is negligible for the luminosity (just 0.25% effect), but it may have effects on lifetime/halo since a small fraction of the collision products can travel to the next IP(s).

HERA procedure on background optimization (by B. Holzer):

  1. First of all beams should collide as well as possible: central collisions, maximum luminosity is crucial.

  2. They optimize the angle of the two beams, again according to the best luminosity, but now also according to the lowest background.

  3. Adjust collimators

  4. Optimize the diffusion rate of the beams (crucial). In the case of HERA the ideal tunes are the ones close to the coupling resonance as they suffer even from 12 order resonances under collisions. And close to the diagonal in the tune diagram there is more free space.

  5. Tune chromaticity (small values)

  6. Optimize the coupling; if there is a measurable coupling the lifetime in HERA is easily reduced by a factor of 5.

  7. The last step is an upstream orbit correction according to the drift chamber currents and background signals of the experiment.

At HERA the background tuning in general is done as a function of the overall loss rate monitoring the BLMs. Clearly they could take the lifetime measurement, but in their case it turns out that the loss rate is faster and much more sensitive.

Note 4. Monitor the delivered luminosity (background subtracted and corrected for any dead-time inefficiency) provided by the experiments.