August 18, 2012
The Large Hadron Collider is running better than ever this week with record peak luminosities of 7.2/nb/s and a record-shattering 0.3/fb delivered in the last 24 hours.
After they reached the 10/fb total luminosity milestone for 2012 a couple of weeks ago they undertook some messing about that included a polarity switch for ALICE. One outcome of all that was that CMS accidentally lost its solenoid for a few days which gave them an opportunity to do some more tests in the shadow. I am not sure of all the details but there seem to have been some collimator changes and the end result is that everything is now running much better so they should be able to make up the lost time. If anyone has a better description of what went on please do let us know.
From Steve Myers introduction at MAC meeting
A couple of days ago there was also a Machine Advisory Committee Meeting to report on where they go next. There is a new schedule with a few changes. It looks like they want studies for running after the long shutdown as a priority, and are making the case that enough luminosity has been delivered already. I think they still need to get as much as they can to help resolve the Higgs cross-section anomalies. A good compromise would be to attempt 25ns run after the next technical stop with a smaller beta*. Without a webcast of the MAC talks it is hard to tell where they are really heading but they seem to be planning for at least some more 25ns development runs.
Another thing they are thinking about is replacing the injection kicker (MKI8) that has been suffering from overheating since luminosities peaked earlier this year. This has been forcing them to delay some fills while it cools down. The change would be risky and could result in lost time, but if they can test out the replacement before the long shutdown it would put them in a better position when they restart at higher energy in 2015. It looks like a big job though.
August 13, 2012
The SUSY 2012 conference starts in Beijing today. It is the biggest supersymmetry conference of the year and we expect to see the latest results using the 5/fb gathered in 2012 at 8 TeV before the last technical stop. Actually at least some of the results have already appeared with three new conference notes from ATLAS this morning here, here and here. CMS released their results earlier, see their twiki page .
Because of the high masses being searched for the extra TeV of energy over last year’s 7 TeV actually provides 2 to 3 rimes as much sensitivity, so even without combining the new results with the similar amount of data collected last year we get significantly better depth. Sadly there is nothing yet observed in these notes beyond standard model expectations. This is disappointing but there may be other searches released later and there are always places for SUSY to hide from the LHC.
The most promising anomaly at this time is the 1.8 times SM excess in the diphoton channel seen in the Higgs search which currently has 2.5 sigma significance BSM in ATLAS and 1.5 sigma in CMS. If the peaks coincided the combined significance would be about 2.8 sigma but they are at slightly different masses so the combined result is actually no better than ATLAS on its own. You could argue that this might be a callibration error and the 2.8 sigma is good. In any case there will be twice as much data available in a few weeks and we will see if the excess is a statistical fluctuation or not. Looking at the four individual results from the two experiments and last year vs this year they can be plotted on a mass vs signal scale roughly as follows
The green line is the standard model expectation, blue circles are CMS and red are ATLAS. Black is the unofficial combination. The results are comparable to throwing 4 dice and getting four sixes. Was it a fluke or were the dice loaded, and if so, how?
If the effect is not statistical it could easily be a combination of systematic errors. This would most likely be due to errors in the theoretical calculations that would affect both experiments. (TS pointed out this paper which fingers QCD uncertainties) Many people would suggest we wait for the dice to be rolled again and then look at systematics more carefully before taking this too seriously. However, by time that has happened the long shutdown will be on us. If there is a possibility for something to be seen here it makes sense to look at what it could be. Theorists might then make predictions that could be tested this year if triggers can be adjusted in time.
I am assuming that the excess in the diphoton channel is due to extra particles that affect the Higgs decay loop and that the production rate via gluon fusion is close to SM predictions. This may be wrong but it is what the data looks like so far. That being the case, the Higgs diphoton loop can most easily be enhanced if there is a new charged particle that adds to the loop. A boson would probably add to the cross-section while a fermion would subtract from it but some knowledgeable theorists say that “vector-like” fermions are also a possibility and who am I to argue. It must be colourless to avoid spoiling the gluon fusion production rate. It could carry lepton number which would affect its decay possibilities. Mass would be greater than 105 GeV otherwise it would be produced via mediated photons at LEP, but less than about 300 GeV to have a significant affect on the loop. Best candidates are scalar leptons like the stau or charged scalars like a charged Higgs, but vectors such as a W’ are also possible. These things have been searched for and already excluded in the required mass range, but only under model specific assumptions. Hadron colliders ahve big blind spots especially when particles decay via jets. There is still hope that something is being missed.
August 4, 2012
The Large Hadron Collider has now delivered over 10/fb at 8 TeV during 2012 in the middle of a long 11 week summer run between technical stops. The 10/fb is for ATLAS and CMS but LHCb has also passed 1/fb in 2012 to add to their 1/fb from last year.
About 3.5/fb have been added in the first 5 weeks after a slow start with time taken out from pp luminosity production for floating MDs, 90m physics (TOTEM and ALFA) and VDM scans. The collider has now settled into a straight stretch with about 1/fb added each week. Peak luminosities are a little down compared to before the last technical stop due to problems with beam instabilities but if they keep it steady the results will be good. There are six more weeks before the next stop with time scheduled for more floating MD and 500m physics. We can expect them to end on 16th September with about 15/fb recorded this year in addition to the 5/fb from last year.
click on image for summer schedule
While this run is in progress we can expect to see results from before the last technical stops at a series of specialised conferences SUSY 2012, TOP 2012 etc., see the viXra calendar for details. It seems most likely that the next Higgs update will come around early october with 20/fb of data available. This will be in keeping with past updates where the amount of data has doubled each time. With the Higgs discovery behind them the next update may be a little more low-key but I think there is a good prospect for reporting a significant excess beyond standard model in the diphoton channel. It may even pass three sigma in one of the experiments.
This list of LHC Higgs updates looks roughly like this
- Moriond, March 2011 – 0.04/fb
- EPS, July 2011 – 1.2/fb
- Lepton-Photon, August 2011 – 2.3/fb
- CERN council, December 2011 – 4.9/fb
- ICHEP, July 2012 – 10.4/fb
- October 2012 – about 20/fb ?
- Dec/Jan 2013 – ??
Assuming they update at around 20/fb in October, can they double the dataset one last time by the end of the year? The final 10 week proton run schedule looks like this,
click for autumn schedule
If they run with the same parameters they will add another 10/fb to the total luminosity, but with the target for the year already achieved I think they will want to do something different for this run. The scheduled scrubbing run after the technical stop only makes sense if they are considering the option of running at 25ns spacing. Earlier MD tests at 25ns have worked but with reduced beam lifetimes. The scrubbing run will help clean the pipes to make the runs more successful. To run at 25ns they will have to reduce the bunch intensity. The PS will need to split the bunches in half one extra time before injection and also because the present high intensities at 25ns would result in too much heating. This means that luminosity will not increase at 25ns unless they can also improve the squeeze.
In fact the MD tests for tighter squeeze down to 0.2m went very well (as far as I know). Current beta* is 0.6m so they have plenty of scope to at least double the luminosity with the tighter squeeze. The 25ns spacing means less pileup making an increase in luminosity more manageable. I don’t know what the actual plans are but I think a 25ns final run with 0.3m would make complete sense if they can get it to work. As well as giving them a chance at doubling the integrated luminosity yet again it will be a valuable trial for runs after the technical stop which will certainly have to be at 25ns spacing. Running at 25ns this year is a risk but definitely one worth taking.