• John
  • Felde
  • University of Maryland
  • USA

Latest Posts

  • USLHC
  • USLHC
  • USA

Latest Posts

  • James
  • Doherty
  • Open University
  • United Kingdom

Latest Posts

  • Andrea
  • Signori
  • Nikhef
  • Netherlands

Latest Posts

  • CERN
  • Geneva
  • Switzerland

Latest Posts

  • Aidan
  • Randle-Conde
  • Université Libre de Bruxelles
  • Belgium

Latest Posts

  • Sally
  • Shaw
  • University College London
  • UK

Latest Posts

  • Richard
  • Ruiz
  • Univ. of Pittsburgh
  • U.S.A.

Latest Posts

  • Laura
  • Gladstone
  • University of Wisconsin, Madison
  • USA

Latest Posts

  • Seth
  • Zenz
  • Imperial College London
  • UK

Latest Posts

  • Michael
  • DuVernois
  • Wisconsin IceCube Particle Astrophysics Center
  • USA

Latest Posts

  • Mandeep
  • Gill

  • Emily
  • Thompson
  • USLHC
  • Switzerland

Latest Posts

  • Ken
  • Bloom
  • USLHC
  • USA

Latest Posts

James Doherty | Open University | United Kingdom

View Blog | Read Bio

Three months in a dark room – only for the LHC

I have spent much of the past three months in a dark room as my research at CERN has involved using lasers to develop a new type of beam position monitor (BPM) for the Large Hadron Collider (LHC). From a technological perspective, this is seriously cool.

What is a beam position monitor?

To thread a beam of particles through a narrow accelerator beam pipe and around various obstacles you need to control and steer the beam using magnets. This is important for two key reasons: firstly, you want your beam to accurately hit its intended target; and, secondly, if you loose beam aperture in high energy accelerators, stray particles will damage your machine – and you don’t want to damage a $10 billion machine such as the LHC.

You can only control what you can measure and so BPMs are of fundamental importance in the operation of particle accelerators.

The current generation of BPMs are mostly electromagnetic in their operation. They use metal strips, known as striplines,  which line the inside of the beam pipe, to measure beam position. A beam of charged particles has an electric field which causes the striplines to become charged – so the striplines essentially act as electrodes. The closer the beam gets to a particular stripline, the larger the charge build-up on that stripline. So by measuring the voltage which builds up across the striplines, one can calculate the position of the beam.

Grand – so we’re sorted for BPMs then? Not quite…

Intra-bunch head-tail instabilities

Yikes, what are those?

Beams are made up of many discrete bunches of particles. Those bunches have beginnings and ends, known as heads and tails. Sometimes the heads and tails vibrate, resonate or even swap position. Such so-called ‘head-tail instabilities’ lead to beam instability and hence accelerator operators need to monitor such phenomena very carefully.

Plots

Plots of intra-bunch head-tail instabilities in the PS in 1974 as recorded on an oscilloscope (courtesy of J. Gareyte, “Head-Tail Type Instabilities in the PS and Booster”, CERN, 1974).

The plots above show such intra-bunch head-tail instabilities observed in CERN’s Proton Synchrotron (PS) in 1974. Note that the time period over which each plot is recorded is 200 nanoseconds (ns), meaning the bunches were approximately 120 ns in length. The problem is that today bunches are hundreds to thousands times shorter than that – so accelerator operators require BPMs of much wider bandwidth to monitor these kind of instabilities.

A new type of BPM…?

I’ve been working on a new type of BPM which uses birefringent crystals to monitor beam position and that, in theory, should have wide enough bandwidth to observe intra-bunch head-tail instabilities in the LHC. To understand how it works we need to understand birefringence…

When light enters a birefringent crystal it is polarised in two different planes – lets say horizontal and vertical. Now each of those planes has a different refractive index due to the structure of the crystal. So the light in the horizontal plane travels at a different speed to the light in vertical plane. This causes light to rotate or twist as it travels through the crystal. The refractive indexes in the crystal change in the presence of an electric field, and this is known as Pockels effect.

An electro-optical (EO) BPM

We can use the birefringent nature of the crystals, combined with the Pockels effect, to monitor the position of a particle beam. As we know, the beam is made up of charged particles and therefore has an electric field. So if you put a birefringent crystal next to the beam, and the beam then changes position, the refractive index of the crystal changes. If we can measure this change of refractive index then we can calculate the position of the beam.

To do this you fire a laser  through the crystal and use a light sensor to detect the intensity of the laser after it exits the crystal. By cleverly placing and orientating a polariser before the crystal (to linearly polarise the light entering the crystal) and then another polariser (known as an analyser) after the crystal, you can block all the light leaving the crystal – so the detector reads zero. This is analogous to rotating a pair of 3D glasses as shown below.

Polar1

Two pairs of 3D glasses with linearly polarised lenses. When the planes of polarisation are parallel, linearly polarised light emerges from the lenses.

Polar2

The same 3D glasses but now with the planes of polarisation perpendicular to each other. The first lens polarises light in the horizontal plane and the second polarises light in the vertical plane.  The result – light is extinguished.

So when the beam is in its optimal position, you ideally wish to be getting a zero reading on your light detector.

Then, when the beam strays from the optimal position, the refractive index of the birefringent crystal changes, causing the light passing through the crystal to rotate to a different degree.  The analyser (which is calibrated to extinguish light emerging from the crystal when the beam is in its optimal position) no longer blocks all of the laser – so you get a reading on your light detector which you can use to calculate the beam position. Easy-peasy!

The advantages of the EO BPM

Birefringent crystals are really sensitive to changes in electric field. So if we can measure changes to the refractive index we can very accurately calculate the position of the beam. Moreover, the laser in the BPM will be coupled into optical fibres and so the signal will not suffer the same levels of attenuation as those carried through the electrical cables of electromagnetic BPMs. We can therefore monitor the position of the beam at much higher bandwidths, as required.

Due to the efficiency of optical fibres in transmitting signals, the crystal may be placed hundreds of meters underground in the accelerator tunnel, while more complicated equipment (including the light detector) can be placed on the surface – where operators may calibrate and control the BPM – without any significant loss of signal bandwidth. The crystals are also tiny so may fit where other BPMs can’t reach.

Laser2

My experimental set-up in the lab – the laser is coupled into the optical fibre using mirrors. It then travels through the optical fibre, is polarised by the system of paddles in the foreground, before emerging from the fibre. The laser beam then passes through the analyser before hitting the light detector.

When will it be in operation?

The EO BPM is still under R&D but we hope to have a prototype ready to test in the Super Proton Synchrotron at the end of Long Shutdown 1 (LS1) in mid-2014. So, the EO BPM is in the pipeline, or at least it will be soon…

The EO BPM is the brainchild of Ralph Steinhagen.

Share

3 Responses to “Three months in a dark room – only for the LHC”

  1. Laurence Cox says:

    James,
    I have a query about the EO BPM operating in the way you have described it. If the detector signal is zero for perfect alignment, then you would get an increasing signal for both too large a polarization rotation and too small a polarization rotation and it is not clear how you tell the difference between the two. If, in fact, your EO BPM is tuned to be slightly off the exact null point, then over-rotation and under-rotation will be distinguishable. Is this what you actually do, but haven’t included this detail in the interests of clarity.
    regards,
    Laurence

    • Hi Laurence,

      Thanks for your comment. I have indeed tried to keep my description of the EO BPM as simple as possible to remain digestible. There’s two points here.

      Like the electro-magnetic BPMs, which have 2 or more striplines around the circumference of the beam pipe, you can also have more than one crystal around the circumference of the beam pipe – so you have separate reference points from which to see what the beam is up to.

      You are also unlikely to have to have an actual value of zero on a detector when the beam is turned on, only when it is off. Ideally you want to minimise the reading on the detector when the beam is in its optimal position, then see how your readings vary when it goes off centre.

      Hope that’s useful – let me know if not.

      James

  2. SRT says:

    That is, indeed, seriously cool. As someone who did thesis work in 17 GHz photocathode guns, the combination of clever nonlinear optics and non-neutral beam physics is very impressive.

Leave a Reply

Commenting Policy