Everything is illuminated

On Monday, 26 January, CMS installed one of the final pieces in its complex puzzle: the new Pixel Luminosity Telescope. This latest addition will augment the experiment's luminosity measurements, recording the bunch-by-bunch luminosity at the CMS collision point and delivering high-precision measurements of the integrated luminosity.


Installing the PLT in the heart of the CMS experiment.

No matter the analysis, there's one factor that every experimentalist needs to know perfectly: the luminosity. Its error bars can make or break a result, so its high precision measurement is vital for success. With this in mind, the CMS collaboration tasked the BRIL (Beam Radiation Instrumentation and Luminosity) project with developing a new detector to record luminosity for Run 2. Working with experimentalists from across the CMS collaboration and CERN, BRIL designed, created and installed the small - but mighty - Pixel Luminosity Telescope (PLT).

"During Run 1, our primary online luminosity measurements came from the forward hadron calorimeter, which we compared to the offline luminosity measurement using the pixel detector," says Anne Dabrowski, BRIL deputy project leader and technical coordinator (CERN). "But as we move to higher and higher luminosities and pile-ups in Run 2, extracting the luminosity gets harder to do." That's where the PLT comes in. Designed with the new LHC Run 2 in mind, the PLT uses radiation-hard CMS pixel sensors to provide near-instantaneous readings of the per-bunch luminosity - thus helping LHC operators provide the maximum useful luminosity to CMS. The PLT is unconnected to the CMS trigger and reads out at 40 MHz (every 25 ns) with no dead-time.

The BRIL team includes collaborators from CERN, Germany, New Zealand, the USA, Italy and Russia.

Research and development on the PLT began ten years ago, with diamonds first considered for the pixel telescope planes. A PLT prototype was even installed along the LHC beam line during Run 1. "Diamond sensors would have been an excellent choice, as they do not need to be run at low temperatures to have an acceptable radiation damage signal loss," says David Stickland, BRIL project leader (Princeton University). However - while the potential for a diamond PLT remains - the prototype results led the team to use a more tested and reliable material for Run 2: silicon.

However, this practical decision would create new issues for the BRIL team to resolve: "Suddenly, heat was a real concern," explains Anne. "If we wanted to get a good signal out of silicon sensors, we had to bring the telescopes down in temperature." With only 18 months to go until installation, the BRIL team had to go back to the drawing board to try and fit a cooling structure into an already-constrained space.

The PLT is comprised of two arrays of eight small-angle telescopes situated on either side of the CMS interaction point. Each telescope hovers only 1 cm away from the CMS beam pipe, where it uses three planes of pixel sensors to take separate, unique measurements of luminosity. (Image: A. Rao)

"We were successful thanks to the ingenuity of the CMS engineering integration office and PH-DT engineers, in particular Robert Loos,” says David. “Rob designed an extraordinary 3D-printed cooling structure using a titanium alloy, using the 'selective laser melting (SLM)’ technique in order to 'grow' the cooling structure we needed." Despite the internal diameter of the cooling channels being less than 3 mm, the cooling structure can make right-angle turns at the drop of a dime and withstand pressure up to 15 bar. "It's tremendously strong, light and compact. I don't know how it could have been made without this technique," David adds.

This is only the first example of the innovative design used by the BRIL group. So while the telescope's installation may be complete, our coverage of their work is not yet over. Look out for an article in the next edition of the Bulletin to find out more...

by Katarina Anthony