Ok my first "serious" post here. Yesterday I mentioned delaying the start of a "new project" with my absence on vacation: the modification of the ZBB trigger. Let me explain what it is about.
CDF, my experiment at the Tevatron, is a big (5000 ton) detector that studies the debris of proton-antiproton collisions provided by the Tevatron collider at Fermilab (near Chicago). In the core of CDF the highest energy collisions ever achieved happen at the befuddling rate of 3,000,000 per second.
CDF is made up of several hundred thousand electronic channels that record the passage of the debris through every bit of the device. Obviously, we cannot record all of that data on tape: we thus have a trigger system, which is a set of hardware devices built with the aim of deciding on the fly what to store and what to discard, based on the intrinsic "interest" of the collision.
Not all proton-antiproton collisions look the same, in fact quite the opposite: some are just caused by a proton and a antiproton scattering elastically, when they do not break up. The physics of these soft interactions is completely understood and we discard these outright. We in fact are interested mostly in the hardest collisions, when most of the kinetic energy is released as new mass - new particles are formed, some might be still unknown - and this would most likely happen for the hardest, least well understood collisions (the rarest, alas).
The CDF trigger has three levels. The first level works at a rate of 3 MHz, and can do very little more than read the fastest of CDF detector components, and try to make out whether there are high energy particles flowing out of the interaction region. Electrons and muons (gold nuggets in a proton collider) can also be discerned, albeit at the price of large background, at this stage.
The Level 1 trigger has the responsibility of accepting only 20-30 kHz of collisions (one in a hundred), sending the signal to the Level 2 trigger to read in the data and ponder on it for as much as 20-30 microseconds, still very little time for us humans but a respectable time interval for computers. During that time, most of the event is reconstructed and understood, and a much more detailed decision can be made on whether the event is worth retaining. If the Level 2 trigger says so (and it must do so no more than 300 times a second, on average), the Level 3 trigger has a whooping 3 milliseconds to decide whether to store the event, based on a full software reconstruction of the event. The output rate to tape of the several hundred kilobytes of raw data for each event is about 75 Hz. Accepting much more than that would cost a lot of money and would not mean a too significant enhancement of the discovery reach of CDF.
So what is it about the ZBB trigger ? Enter the Z0 boson, a very heavy particle discovered in 1982 at CERN along with the W boson. The Z0 has been a gold mine in the nineties at CERN, where the LEP collider produced millions of them and made exquisite physics measurements with that dataset. At the Tevatron, albeit with much higher energy, we produce fewer of them, because we collide protons, not electrons. We still collect them, because they provide a wonderful calibration for our detector: the mass of the Z0 is known to within a few parts per million, and reconstructing it in our detector, measuring its mass from the energy of the particles produced when the Z0 disintegrates, allows us to calibrate the response of our device very cleanly.
The ZBB trigger is a calibration trigger. Its purpose is to collect the events that contain the signal of the disintegration of the Z0 boson into a pair of b-quark-originated jets of hadrons. If we collect enough of them (we expect to get a few thousand decays per year of running), we will obtain a very clean picture of the response of our detector to the sprays of low-energy hadrons that we call jets.
Using the Z0 to measure the detector response to jets is about the only way to do it, since we know only of three particles that may decay into jets: the Z0, the W, and the top quark. The W is indiscernible from the background when it decays to jets; the top can be seen -and we do- in its decay to jets, but we do not know its mass well enough to use it as a calibration - yet.
Now, I can finally tell you about the project. The ZBB trigger is a set of instructions to the Level 1, level 2, and level 3 of the CDF trigger, that allow the collection of events that have a similarity to those containing the signal of the Z0 decay to two b-quark jets. We recently revised the trigger, to increase its performance and to reject more background, since the Tevatron collider is providing more proton-antiproton collisions these days, and we need to reject more junk if we want to stay within the budget of 75 Hz of events written to tape - the Level 3 output.
Now, in order to make the trigger operative, we have to make a presentation at the Trigger and Datasets Working Group (TDWG), a CDF group that takes the final decision on what instructions are given to the CDF trigger system. I was away last week, and we had to delay our instructions to the TDWG... This is a minuscule loss of productivity for our calibration trigger (the old trigger is taking the data anyway), but still it is a mild disappointment for the few people who have worked with me since October last year to modify our trigger... I will get things straight with them in front of a beer next time I fly to Chicago! Which, by the way, will be next Tuesday, January 11th.
Hi, I'm a collegial student and i'm very interest whit the studies on the Z0 bosons. I heard the Z0 boson was one of the very first particle present 10^-35 secs after the big bang. And one other thing i've learned is that this boson, désintegrated, created some or all of the periodic table's elements. Is it true or am I mistaking.
Thanks a lot
Sébastien Jutras
Posted by: Sébastien Jutras | October 20, 2005 at 09:25 PM