The KL beam is generated by the proton beam when it hits the "KL target"; then the beam proceeds in parallel to the KL beam (and is monitored by the "KS tagging" detector), until it hits the "KS target" and generates the KS beam.
The KL beam is generated by the proton beam when it hits the "KL target"; then the beam proceeds in parallel to the KL beam (and is monitored by the "KS tagging" detector), until it hits the "KS target" and generates the KS beam.
Goal: measure Re(ε'/ε)~(1/6)*[1-(Γ(KL->π0π0)/Γ(KS->π0π0))*(Γ(KL->π+π-)/Γ(KS->π+π-))]
The KS channels are easy (they are the dominant decays), but for KL:
Method:
KS and KL are separated according to time difference between the signal in NA48 and signal in the proton tagger. KS: Δt<2ns, KL: the rest.
Lifetime weighting: the KL lifetime (vertex position) distribution is flat, because it's very distant from its production point; for KS it's a falling exponential. This gives a difference in acceptance. So, the KL lifetime is reweighted to an exponential, too.
This makes the acceptances cancel out in the double ratio R, but at the cost of an increase in statistical error.
Vertex resolution: 2.8 m (over 100 m)
K3μ events (i.e. K->πμν) are rejected by using muon chambers.
K3e events (i.e. K->πeν) are rejected by using E/p.
Generical 3 body decays are rejected by looking at Pt of the event (if 1 particle is missing, the others don't have balanced momenta).
In the neutral final state, a χ2 is computed representing the event compatibility with a 2π0 hypothesis.
Control of the calorimeter energy scale:
a shift on energy in the calorimeter translates to an apparent decay vertex displacement; idea: follow the reconstructed position of the source of KS's. (In KTeV it's the position of the regenerator, in NA48 you can follow the KS anticounter.)
Previous experiments: E731 and NA31.
E731 was recording one single final state in simultaneous beams and then changing final state (corresponds to measure KL/KS->π0π0 and then changing to π+π-).
NA31 was recording both final states simultaneously and then changing beam (corresponds to measure KL->π0π0/π+π- and then changing to KS).
KTeV uses simultaneous beams and final states, like NA48, but the way it creates the beams is completely different:
2 parallel KL beams, one passes through a movable "KS regenerator".
Follow up experiments: