LHC vs. Tevatron Update

Over the weekend the LHC had a first successful physics run with nominal intensity beams, in 3 bunches. A peak luminosity of about 5 x 1029cm-2s-1 was achieved, and the total integrated luminosity per experiment is now around 30 nb-1. While this is quite a bit behind optimistic schedules of earlier this year, it may now be possible to much more quickly increase the LHC luminosity as the number of bunches is increased. The current plan foresees an integrated luminosity of about 1 pb-1 in July, and another 3 pb-1 in August.

The report about this from BBC News has the LHC’s Mike Lamont trash-talking about the Tevatron:

“It’s clear that the LHC is the new boy in town, but in two years running we’re going to put Fermilab out of business,” operation group leader Mike Lamont told BBC News.

John Ellis is enthusiastic about the possibility of producing black holes:

Professor Ellis added that as the luminosity increases, one of the things physicists at Cern will be looking for is a mini- black hole.

“It would be absolutely, fantastically exciting if we produced black holes at the LHC,” he said.

“Then we would test our ideas about gravity, quantum physics, string theory. This would be much more exciting than finding a… Higgs boson or even dark matter.”

Meanwhile, over in Batavia, the Tevatron has been regularly operating at peak luminosities of 3-4 x 1032cm-2s-1, nearly a 1000 times that of the LHC, accumulating integrated luminosity of around 50 pb-1 a week. They’re getting the total number of collisions produced at the LHC this year about every couple of minutes. So far this year they are doing even better than planned, with over 2000 pb-1 of integrated luminosity in FY 2010. Last week, the Physics Advisory Committee met to consider plans to get in Mike Lamont’s face, and keep operating the Tevatron past its planned closing date of end FY 2011, possibly for another three years. This would take their total data set from about 10 fb-1 to possibly as much as 20 fb-1. With this amount of data they expect to be able to provide 3-sigma evidence for a Higgs over the entire expected mass range, as well as stay ahead of the LHC in several different measurements, including the sort of possible non-SM CP-violating effects that recently have been in the news.

Update: More about CERN’s competition with the Tevatron here:

The LHC now has to produce as many collisions as possible in the next two years in order for the various experiments at CERN to essentially prove their worth among other established particle physics laboratories.

The past failures of the LHC weighed heavily on operations group leader Mike Lamont who talked about some of the criticism from the media.

“The Americans in particular can be quite aggressive,” he told Deutsche Welle.

“It’s quite clear that we’re competing with the States, and we’ve had setbacks, and you can see journalists occasionally being aggressive about that,” he said. “I mean ‘You’re spending taxpayers’ money, and you’re still messing up,’ which can be a fair comment.”

Only if the experiments meet their goals for collected data by 2012, Lamont said, would CERN pull ahead of the research performed at the US-based Fermilab, a particle accelerator located near Chicago, Illinois that measures 6.3 kilometers in circumference.

“We’ve got reach in energy, but they’re still sort of chasing at our heels,” Lamont said. “So if we can collect enough data in 2010 and 2011, we essentially put them out of business, then we can relax in 2012 and fix the [LHC] properly.”

Update: The latest news (01:51) on the LHC Vistar doesn’t sound good: “soon access in US15 for fire brigade”. The beam was lost around 01:00, soon after beams had been ramped to 3.5 TeV. US15 is an underground service cavern next to the ATLAS detector.

Update: Not clear what that was about, but as of 4:30 things are back to normal and they’re getting ready to inject another beam.

This entry was posted in Experimental HEP News. Bookmark the permalink.

10 Responses to LHC vs. Tevatron Update

  1. VVAA says:

    Wow, you seem disappointed that things went back to normal!

  2. Peter Woit says:

    Not at all, I’m very glad the problem wasn’t serious, and they’re now back in business, as of a few minutes ago colliding beams at even higher luminosity. That’s great.

    I hope both the LHC and the Tevatron do great physics over the next few years, in a friendly and vigorous competition…

  3. Pingback: EE.UU. contra Europa, o el Tevatrón del Fermilab contra el LHC del CERN « Francis (th)E mule Science's News

  4. nbutsomebody says:

    It is unlikely that Elis takes the proposal of black hole production very seriously. Last time I heard him ( in february) he was very critical about the idea.

  5. Dear Peter,

    the problem of the Tevatron experiments with the Higgs search is that they will have to face degrading performances of their silicon detectors in the second part of a 20/fb dataset. Silicon sensors are sensitive to the radiation dose they get, and those of CDF and D0 were not designed to withstand such high integrated fluxes. They are currently working fine, but signs of increased leakage currents are clear, and bias voltages need to be knocked up. Signal to noise has already decreased. For a while things will not directly affect the b-tagging efficiency, but we are on untested ground to a non negligible extent.

    Another issue is that the predictions that are being sold for the Higgs reach at the Tevatron are based on 2xCDF, which assumes that D0 will eventually catch up with the better sensitivity that CDF is showing in the low-mass analyses. I know how hard CDF has worked on the low-mass region, and have serious doubts that D0 will ever match CDF there, especially given the serious undermanning of D0.

    In summary, the reach for the Higgs at low mass is, IMHO, not as good as it is made to appear.

    Best regards,


  6. Bill K says:

    “A peak luminosity of about 5 x 10^29/cm^2/s was achieved, and the total integrated luminosity per experiment is now around 30/nb. While this is quite a bit behind optimistic schedules of earlier this year, it may now be possible to much more quickly increase the LHC luminosity as the number of bunches is increased.”

    Does the luminosity go up quadratically as the number of bunches is increased? I.e. does each bunch going one way get to collide with each bunch going the other way?

  7. lumo says:

    Luminosity and bunches ~ typically yes.
    The luminosity goes as the product of the number of particles in each bunch, divided by the cross-sectional area. For convenience assume the beams have equal cross-sectional areas, say an ellipse with rms semi-axes (sigma_x, sigma_y). Assume also the bunches collide head-on and the longitudinal axes are coincident (not merely parallel). This is all fairly typical operation. Then if the circulation frequency is f_rev, the luminosity is

    L ~ N1 N2 f_rev / (sigma_x sigma_y ) * (factors of pi, etc)

    Hence the units of luminosity are cm^-2 s^-1.

    Suppose there are B1 bunches in one ring and B2 in the other ring, then

    L ~ B1 B2 N1 N2 / (sigma_x sigma_y ) * (factors of pi, etc)

    Typically B1 = B2. Otherwise the operation is less than optimal. Also typically N1 = N2. In practice this is only approximate because some particles are lost during the acceleration process, one cannot guarantee that the transmission efficiency is equal for both rings. But one can say N1 ~ N2.

    Because the beams are proton-proton, they are in two separate rings, and collide only at the interaction points. This is all the same for LHC and RHIC.

    But in a p-pbar collider, or an e+e- collider, the counterrotating bunches pass through the SAME ring, in opposite directions. Then if there are B1 = B2 = B bunches, they collide at 2B points.
    This leads to unwanted collisions in the ring arcs, which disrupt the beams. So the beams have to be artificially separated where collisions are not desired.

    But to answer your question ~ yes.

  8. PhilG says:

    The bunches in one direction do not intersect with every bunch in the opposite direction because they are in separate rings and are only brought together at the points where the experiments are housed.

    The exact number of interactions per turn depends on the filling scheme, i.e how they arrange the pattern of bunches in the buckets. For large number of bunches you can assume one collision per turn per bunch. This means the luminosity increases linearly with the number of bunches, not quadratically.

  9. lumo says:

    Mea Culpa! I stand corrected.
    You are absolutely correct that a bunch in one ring does not collide with every bunch in the other ring. FYI in RHIC for the polarized protons, the polarization pattern is ++– in one ring and +-+- in the other ring. That way one gets all 4 spin combinations (++) (+-) (-+) (- -). I don’t know if there are any unpolarized bunches.

  10. Chris Oakley says:


    I assume you cannot be Lubos Motl (“Mea Culpa” is not something he would be capable of saying), but this nickname is often used by him. Can I therefore suggest that you use another to avoid confusion? Or, better still, just use your real name!

Comments are closed.