Dear friends

From the earliest days of CMS, when the design was made on a scrap of paper in a local restaurant, the basic concept of the detector was that it should be assembled in large pieces on the surface and then lowered underground. This concept became reality with the first “heavy lowering” operation, which took place little more than a year ago. With the lowering of the final heavy piece – the “YE-1” endcap disk – CMS has all but completed the assembly of the detector (only the Pixels and the Endcap Electromagnetic Calorimeter – including the Preshower – remain, and will follow shortly). This milestone was deservedly celebrated with the opening of several bottles of champagne!

Some of you may ask yourself how these huge pieces of the detector could be lowered down without any major problems. For this special edition, Elizabeth Gibney interviewed many of the team involved with the lowering, in order to explain why this method was chosen for CMS and how the lowering was organized from the very beginning until the last disk successfully landed on the cavern floor.

Congratulations go to everyone involved in this immense, but ultimately successful, undertaking.

With kind regards,

Dave Barney
Marzena Lapka
Elizabeth Gibney

From Nature, 30 January 2008:

Snapshot: Search for Higgs primed to start

The final element of the Compact Muon Solenoid (CMS) detector is gently lowered into place at CERN, Europe's particle-physics laboratory near Geneva in Switzerland. The detector will look for physics beyond the 'standard model' in the high-energy collisions of protons at the Large Hadron Collider, CERN's next-generation particle accelerator...

Read more: http://www.nature.com/news/2008/080130/
full/451507a.html

From Ethnos (Greece), 25 January 2008:

The picture that appeared in the article.

Read the full article (in Greek): http://www.ethnos.gr/article.asp?catid=11386
&subid=2&tag=8784&pubid=326534#

Click here to download printable PDF version.

Long before the sun rose on 22 January the CMS surface hall buzzed with press, coffee and croissants, as the crowd watched YE-1 begin its descent. But by 6pm the spectacle had moved 100 metres underground as the fifteenth and final heavy piece of CMS came to a halt on the cavern floor.

Endcap disk YE-1 sinking below the surface hall horizon.

The morning started with a few minor problems as the poor conditions outside took their toll on the equipment: “There were some huge gusts of wind that made the gantry sway and one of the cables got jammed in the winches,” explained Technical Coordinator Austin Ball.

Even with heavy duty equipment, the solution is often simple: “We cut out the broken section of cable and secured the strand with tape before feeding it through manually,” explained Hubert Gerwig, Lead Engineer for the operation. “Once it was passed the critical part we could carry on as normal. There was no safety issue; we just needed to take it slowly.”

Maf Alidra next to YE-1. The endcap is lowered by steel cables onto the air pads of the experiment hall floor.

It took about an hour to recover but from then on it all went smoothly. The descent itself was slow (as normal!), but clearly visible to bystanders watching the 1430 tonne disk submerge below the surface hall floor and emerge from the sky into the cavern. Press from Canada, the Ukraine, Russia and France observed the occasion eagerly and watched the final landing from three levels.

The journey took a full 12 hours, and champagne corks popped as the element came to a rest on the airpads and, by the time the cables were finally slackened, it was already in full flow.

After a long and hard day the crowd of mechanics, engineers and observers shared in celebrating the remarkable achievement. “It’s a combination of relief and slight sadness,” said Austin. “There is a touch of regret though that it’s all over, but we’re entering a new exciting phase now: we’re going to be doing the physics. So just like any good day; it’s a mixture of relief and satisfaction.” 

The decision to employ a method that involves hoisting thousands of tonnes of expensive machinery, the result of endless hours of work, down a 100 metre gaping hole was never going to be taken lightly. But CMS’s unique method of slicing up the detector and lowering each piece into the cavern ready-made, has been nothing short of a success.

“It’s a huge amount of responsibility; you have all the physicists waiting for an element to arrive, and if you make a serious mistake, it could be the end of CMS,” says Hubert Gerwig, whose task it was to ensure each sub-detector’s safe arrival.

Lifting gantry for a possible implantation at Point 2.

But lowering CMS by means of heavy lifting was a decision taken at the very beginning, some 16 years ago, inspired by experiences with LEP. “The concept of building large objects on the surface and transferring them completed to the underground area was the clear way to go,” says Alain Hervé, CMS’s original technical coordinator. Though not in LEP’s initial plans, success in lowering two pieces into position in L3, weighing 300 and 350 tonnes, suggested the method could work.

Being able to work in parallel on the civil engineering and the detectors was clearly a huge advantage, and the ‘slicing’ also meant that, through complicated loops of cable chains, water, gas and cooling leads, each piece remained accessible within the cavern, a contrast with experiments built in a ‘Russian dolls’ style, from the inside, out.

Perspective view of CMS as shown in the LoI of 1992.

But the implications of using this method needed careful consideration too: “As soon as I saw the sketch I had a clear vision of how such an experiment had to be organized,” says Alain. “We were already two years inside LEP running, and lessons from the construction and the first shut-downs were clear to me. The way the experiment had to be sectioned and installed was the first priority, not developing the design,” says Alain.

And so the experimental area, pits and surface hall all had to be designed around the lowering method. Foundations for the gantry and a massive strong plug to hold the pieces over the shaft before the cranes took the strain were incorporated. Anchor points for lifting were integrated in the design of the yoke as well as HCAL (Hadron Calorimeter) and HF (Forward Calorimeter) platforms and cradles.

The first lowering took place in November 2006, and the last on 22 January 2008. “Everything has been calculated and calculated again, but that’s still not the real thing,” says Hubert. “In the end we were successful and nothing was damaged. So there is a sense of relief; it’s a bit like an exam, you feel better once it is over and you can celebrate!”

Ten companies internationally were capable of taking on such a project as CMS, but VSL Heavy Lifting, a Swiss company, were the eventual winners of the 2 million CHF contract.

Though VSL were used to lifting such heavy loads as the 7500-tonne roof of the Airbus Assembly Hall, and the walkway between the Kuala Lumpur Twin Towers, CMS presented unique challenges. For a start, the equipment needed to lower, not lift, and a long way. Adding to this was the fact that the scientific cargo was far more delicate than its industrial counterparts.

The gantry support system was rented and made from existing towers, but the 3.4m high horizontal beam, that took the strain of the load, had to be custom made. Even such an enormous beam still bent 5cm under the weight of the heaviest element, the 2000 tonne central barrel ring known as YB0.

The gantry support system seen from outside the surface hall.

During each lowering, the element is supported at four points, but by many more cables as each bunch is made up of 55 individual 15.7mm diameter strands of cold-drawn steel, each with a capacity to hold 28.4 tonnes. Within each bundle, half of the strands are twisted one way and half the other to make overall turning force zero and preventing any turning of the load during the descent.

The lowering in fact happens in 200 small steps as the cables are fed through the hydraulic system half a metre at a time. This works on a grip and release basis, similar to your own hands lowering a bucket down a well, and at all times the load is safely supported by one or other of the clamping mechanisms. Once the cargo reaches the ground, it comes to a soft landing on hovercraft-like airpads that take the load off the cables and later on allow easy horizontal movement within the experiment hall.

The enormous central barrel, YB0, emerging safely from the shaft's bottleneck and into the cavern.

Each lowering was a big event, requiring a week’s preparation to manoeuvre the plug, element and attachments in place, and each was a success. Though there were also, of course, a few challenges that arose along the way: YB0, for instance, the largest segment to enter the cavern, at one point cleared the sides of the shaft by just 10cm either side, and the team needed surveillance video cameras to carefully monitor its movements and ensure that the element passed the bottleneck safely.

The slight slope in the CMS cavern presented another issue. Whilst unnoticeable underfoot, the 1.23% gradient meant that when one side of any element touched down, the other still had 3cm to go. “Initially we tried to compensate for the slope, so that there was an even landing,” explains Hubert Gerwig, the engineer in charge of heavy lifting, “but soon we realised it wasn’t necessary because under the load, the metal strands actually elongate by 10cm, making them somewhat elastic.” The elasticity meant they could simply carry on lowering the piece until both ends were on the ground, also giving rise to the slight “bouncing” effect visible as each piece finally came to rest on the experiment hall floor.  

That some of the elements couldn’t be supported on just four attachment points presented another issue and the endcap disks , for example, were instead carried down on a special crate. But being supported nine metres below their centre of gravity made them naturally unstable. The team solved this dilemma by using stabilisers at the top of the endcap such that if it tilted, it would simply lean against the lowering cables.

In the end each slice of CMS found its way to the experiment hall safely, thanks to the hard work and dedication of all involved, and the group and their equipment can now be disbanded. The VSL team will now take their gantries and hydraulic equipment to Durban in South Africa to build a stadium for the 2010 World Cup. Some parts, such as the cables that have carried down every piece of CMS, are now too fatigued to be used again. Finishing the lowering is a sad event for some, but the VSL mechanics like to travel the world and are eager to move on. “I think some of the workers get the opposite of home sickness,” says Hubert, “they cannot wait to get away again!” And we wish them good luck.