jump to navigation

Go Tevatron! May 24, 2007

Posted by dorigo in games, news, personal, physics, science.

The owl shift at CDF is a boring business, but during night time the accelerator usually is able to perform at its best. The guys at the Main Control Room tend to favor a scheme whereby they build a large stack of antiprotons during the day, and inject protons and antiprotons (a new store) in the late evening, to run at the highest luminosity during the night hours. This way things appear to run more smoothly. Moreover, things are quieter even in our control room, with fewer people messing around… This results in a better chance of high efficiency in the data taking.

The eight hours of owl shift are thus well placed to produce a record integrated luminosity to CDF. In our control room, a big white board is full of instructions written with erasers, and on the top right there is a part which is seldom updated: it contains the records of data collection, a sort of home-made “hall of fame” for the shifters. Right now, the board contains the following information:

  • Best efficiency per shift including shot setup and integrated luminosity greater than 1/pb: 07/31/06, eff=93.7% (a shot setup is when the beams are injected in the accelerator, which results in some initial obligatory losses of efficiency to ensure there is no harm in turning on the silicon detectors);
  • Best efficiency per shift with integrated luminosity greater than 1/pb: evening shift, 8/16/06, eff=97.2%;
  • Integrated luminosity record: evening shift 02/24/07,  delivered = 4429.8/nb, acquired = 3876.6/nb;
  • Record of integrated luminosity for a single run: 2/24/07,  run #235056: Delivered 8.48/pb, acquired 7.51/pb.

Today is my last day on shift, and we have a chance to get close to the integrated luminosity record… I doubt we will break it, but so far, after 4.47 hours of data taking out of the 8 hours of the shift, we are at 2796/nb delivered, 2589/nb acquired. Unfortunately, the luminosity drops exponentially during the store, due to particle losses in the ring for beam-gas interactions and for the beam-beam interactions going on in CDF and D0. So I expect that if the run continues to go as smoothly as it has so far, we will get close to 3.6/pb collected… It would still be a success, and maybe the second or third best shift. Go Tevatron!

UPDATE: 6.11 in the morning, 1.49 hours to go, and the acquired luminosity is 3167/nb! if we go on like this, we will get very close to the record!

UPDATE: 6.33 and we are at 3324/nb! I am not starting to consider cheats like waiting a few minutes past 8AM to close the shift, which would buy us a bit of extra luminosity… According to my estimates, if we do not get more downtime we can easily get over 3.8/pb tonight. I am proud of myself, despite having done virtually nothing tonight (the smoother things go, the less a shift crew has to do).

UPDATE: 7.05 and we are at 3433/nb… Unfortunately we got a hit in efficiency due to downtime, since we had some DAQ problems with our Level-2 trigger. This means we will not make a record tonight, boo hoo. Still a great night though…

FINAL UPDATE: the final number is 3.74/pb at 91% efficiency. Quite good, but I now think we would not quite have made a record even with 94% efficiency (3.74*0.94/0.91=3.85/pb).



1. Kea - May 25, 2007

You sound like the stereotypical Ferrari-driving Italian! Can’t wait til the data is analysed … will it take long? Lol.

2. dorigo - May 25, 2007

Unfortunately I do not own the necessary heaps of money to buy half-million-dollar cars… As for the data, it usually takes about a couple of months before they are available for offline analysis. The real bottleneck is analysis itself…


3. peter - May 26, 2007

he sounds more like a sterotypical moron.
I am told it takes about 6 months to process the data.
The bottleneck for him is catering to his blog.
Be compassionate.

4. dorigo - May 26, 2007

Dear Kea,

Peter is just Paolo Giromini who is trying to keep alive a twice dead joke. Anyway, he is wrong, since he is told something which does not correspond to the truth. The data has recently been made available in two months, and I am told by Donatella Lucchesi, my colleague in Padova and co-head of the CDF offline group, that there is effort to reduce this time lapse.


PS for comments about the fake identities used by Paolo here, see the thread in “Idiot’s guide to running a multi-million dollar experiment”.

Sorry comments are closed for this entry

%d bloggers like this: