Pinning down the LHC schedule

31 May 2010

Daily record of peak luminosity delivered to ATLAS


It has been an eventful couple of months of collisions at 7 TeV, with successive weekends delivering long proton fills and big jumps in the total integrated luminosity recorded by ATLAS. With the summer conferences creeping up fast, the focus is on gathering the maximum amount of data possible, but planning ahead can be difficult.

“I only speak about the short-term schedule. I don't speak about long term any more because it's always wrong!” smiles Deputy Run Coordinator, Martin Aleksa. “The problem is that we experience each morning that the schedule changes for the day with respect to what was planned the day before.”

While the LHC is still in its initial commissioning phase, everything is a new experience. As specific problems or unexpected features crop up, the machine experts slot in an extra shift here or there to investigate, and the schedule is forced to slide.

“For the moment it's the first steps,” says Martin. “They've been using this machine since a few months only, so they don't know it very well yet.” Like a new driver behind the wheel of an extremely powerful car, they're taking it slow.

“They're getting into regions which have an integrated energy which can destroy things, which means that they have to be extremely cautious in going up. It's purely [an issue of] machine protection; it's not something which doesn't work.”

As we go to press, the peak instantaneous luminosity which has been achieved is of the order of 2 x 1029 cm-2s-1. The aim for the end of the year is to hit 1032 cm-2s-1, which will be achieved by increasing the number of protons per bunch to the nominal 1.15 x 1011 (currently they're running with about a fifth of that) and packing more bunches into the beam.

“We would rather like to go first higher in the number of protons per bunch,” explains Run Coordinator Benedetto Gorini, “because the luminosity – or how many collisions you get – goes with the square of the number of protons per bunch, and only linearly with the number of bunches.”

Problems with beam orbit regulation and tripping magnets hindered the commissioning for saturated proton bunches over the last week or so, so attentions turned to adding more bunches into the ring. Over the weekend of May 22nd/23rd, a new record was achieved with 13 bunches circulating in each direction, eight of which were colliding in ATLAS. Nevertheless, in the pursuit of reaching higher instantaneous luminosity as quickly as possible, the working plan remains to complete the proton saturation step before adding serious numbers of extra bunches.

A power cut over the last weekend (May 29th/30th) put things on hold for a couple more days, and May 31st marked the start of a three-day planned technical stop, meaning that the wait for 'fat bunches' will have to be endured for a little longer yet.

Once proton bunch saturation is achieved, the damage risk to the 'interaction regions' of detector will be minimised by reducing the beam squeezing for the time being, so that the beams are less focussed where they cross. The value of ß – the parameter which describes the distance from the interaction point over which the beam size is doubled – will be altered from 2 m to a less aggressive 5 m, reducing the density of protons at the interaction point.

From then on, luminosity will be increased step-wise, by doubling the number of bunches in the ring at weekly intervals until a total stored energy of 1 MJ is reached.

“Already with 16 nominal bunches, you'll get yourself a megajoule,” says Martin “A megajoule will melt you a small block of iron, so for the beam dump this quantity is relevant." It is also significant because it represents the current maximum energy of other accelerators. "That means that once they get to a megajoule, they will want to go even slower.”

The machine is designed to take 2800 bunches eventually, and numbers could reach 700 by the end of this year. The original plan to collect 1 pb-1 of data by the end of June now looks unlikely though. As we go to press, the cumulative integrated luminosity collected by ATLAS stands at 15 nb-1. But as more data flows in, production of useful particles such as Z bosons increases.

“Zs are very nice because they can produce two electrons or two muons, and with those you can calibrate your detector,” explains Martin.

As ATLAS continues to calibrate and the Run Coordinators continue to chase the moving target that is the LHC schedule, the mood is generally positive.

“There were lots of concerns after the accident,” says Benedetto, “but now the machine is working very well. From our side, we'd like them to go a bit quicker, but they're not going slowly because the machine isn't working. They're going carefully.”

 

Ceri Perkins

ATLAS e-News