The purpose of this page is to develop as general an argument as I can for comparing the relative sensitivity to new physics of CDF versus CMS. This is driven by a desire to understand how much data CMS needs before it starts exceeding the discovery reach of CDF in some, any signatures.


For simplicity, let's just look at yields, and ask ourselves what the yield for producing some high mass object depends on:

yield = integral ( parton luminosity * matrix element * phase space ) * trigger efficiency * acceptance

Here the integral is meant to first go over matrix element times phase space at fixed sqrt(s). This gives you a differential cross section as a fucntion of parton-parton center of mass energy. You then integrate that times the parton luminosity over all center of mass energies available.

significance = signal yield/sqrt(signal + background yield)

Let's go through these effects one by one in a very crude fashion to see which ones matter.

Parton Parton Luminosity

The figure below (courtesy Claudio Campagnari 2006) depicts the ratio of parton luminosities for Tevatron (i.e. ppbar @ 1.96TeV) versus LHC (i.e. pp @ 14TeV). The idea here is that protons are composite objects, leading to hadron colliders being broadband machines that have parton-parton collissions across a wide range of center of mass energies (i.e. sqrt(s)). Claudio has taken some standard set of parton density functions, and calculated parton-parton luminosities for ppbar@1.96TeV and pp@14TeV, and then divided the two. He has done this for both quark-antiquark as well as gluon-gluon collissions.

The plot is interpreted as follows. A high mass particle produced via quark-antiquark annihilation at a mass of ~200GeV will get produced roughly 10 times as copiously per pb-1 at the LHC than at the Tevatron. However, if you make that particle heavier, say 1TeV, and allow it to be produced via gluon-gluon then almost 1 Million times as many of those particles are produced per pb-1 at the LHC versus the Tevatron. For example, gg -> H for mH ~150GeV: 100pb-1@LHC gives the same yield as 10fb-1@Tevatron. Though, this is a bad example.


Trigger Efficiency and Acceptance for CDF versus CMS

The main difference between CDF and CMS in terms of triggering is that CDF has a track trigger at level 1, and a secondary vertex trigger at level 2. CMS triggers primarily on electrons, muons, and photons with thresholds of 10-30GeV and 80GeV respectively. Di-object triggers are slightly lower in all cases. Single MET and jet trigger thresholds are 90GeV and 400GeV respectively, with a variety of multi-object triggers. It's worth noting that the single e,mu triggers are quite comparable to the 20GeV e,mu trigger in CDF, while MET and especially jet triggers have significantly higher thresholds than CDF.

Acceptance is likely to be slightly better at CMS than CDF as the detector is much more uniform, and covers a larger eta region. However, except for the SVT trigger in CDF, there's really no difference large enough to significantly impact the discovery potential of the two experiments when compared to each other. We thus might as well ignore the trigger and acceptance details.

-- FkW - 31 Dec 2006

  • HLT-menu-part1-PTDRvol2.gif:

  • HLT-menu-part2-PTDRvol2.gif:

  • L1triggerMenu? -PTDRvol2.gif:

  • partonpartonlumi.gif:

Topic revision: r2 - 2007/01/02 - 07:17:10 - FkW
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback