Level-2 Calorimeter Trigger Upgrade at CDF Page: 2 of 4
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
FERMILAB-CONF-07-165-E
entire detector as a 24x24 map in the T-0 plane. The trigger
tower energy information is then sent to both L1 and L2
calorimeter trigger systems with 10-bit energy resolution,
using a least significant count of 125 MeV resulting in a full
scale Et of 128 GeV. The Level 1 calorimeter (L1CAL)
subsystem only uses 8 of the 10 available bits for each trigger
tower for Li processing, with the least significant and most
significant bits dropped, giving a least count of 250 MeV and
a full scale of 64 GeV.
As examples, electron and photon triggers are formed at
L1CAL by simply applying energy thresholds to the EM
energy of a single trigger tower while "jet" triggers are formed
using the total EM+HAD of a single trigger tower. For
electrons, tracks from the Level-1 track trigger (XFT) can be
matched to the trigger towers while HAD energy can be used
for rejection. LiCAL also calculates global transverse energy
per event, total transverse energy (SumET) and missing
transverse energy (MET), using the lower resolution 8-bit
EM+HAD energy information. The L2CAL subsystem
receives the full 10-bit trigger tower energy information.
However, the existing hardware-based L2CAL system does
not re-calculate the event SumET and MET using the full
resolution energy information available, rather, it still uses the
SumET and MET information directly from L1CAL. This
design feature limits its trigger selection capability for triggers
requiring either SumET or MET. The main task of the
existing L2CAL is to find clusters using the transverse energy
(Et) of trigger towers. The cluster finding algorithm is based
on a simple algorithm used for Run I, and is implemented in
hardware. The L2CAL hardware forms clusters by combining
contiguous trigger towers that have non-trivial energy. Each
cluster starts with a tower above a "seed"100
80
60
40
2020 40 60 80 100 120 140 160
lumiE30
Figure 2: Rate of the jet trigger selection requiring jets above 40 GeV as a
function of the instant luminosity (1030cm'sec').
threshold (typically a few GeV) and all towers above a second
lower "shoulder" threshold that form a contiguous region with
the seed tower (or each other) are added to the cluster. The
size of each cluster expands until no more shoulder towers
adjacent to the cluster are found. The position of the cluster is
taken as the seed tower position. The existing L2CAL trigger
system has worked well at lower luminosity for Run II. As
the instantaneous luminosity increase the calorimeter tower
occupancy increases due to pile up events. At high
instantaneous luminosity (higher occupancy) clusters start to
merge leading to clusters with artificially high number oftowers which give rise to fake high Et clusters. These miss
identified high Et clusters pass the jet trigger thresholds and
cause rapid and unsustainable growth as a function of
instantaneous luminosity in jet trigger rates and cross sections
at L2 (see Fig 2 & 4).
II. L2 CALORIMETER TRIGGER UPGRADE
The basic idea of the upgrade is to provide the full 10 bit
resolution trigger tower energy information directly to the
Level-2 decision CPU where we can run a new cluster finding
algorithm and recalculate the global quantities such as MET
and SumET. To do that we need to develop a new hardware
path connecting the LiCAL directly to the L2 decision CPU
(see Figure 3). This new hardware path is based on the Pulsar
board [2], a general purpose VME board developed at CDF
and used for upgrading both the Level-2 global decision crate
[3] and the Level-2 silicon vertex tracking (SVT) subsystem
[4]. In the upgraded L2CAL system the full 10-bit trigger
tower information is sent to Pulsars which receive, merge and
serialize the tower data so it can be sent to the L2 decision PC
where the clustering will take place. The details of the
clustering are discussed later.L
Level 2 Leve 2 L2 CPU
algorithms Pulserecrate
L eve/2 ca/orimetric triggers
10 bits8 bits L1 algorithm
T Level calorimetric trigger10 bits
Calorimeter
Figure 3: Hardware configuration for the L2 Trigger upgrade. The red path is
the new hardware path that makes available the full 10 bit resolution trigger
towers to the L2 decision CPU.
JETET>40GeV
102
-Fixed cone clustering
-Old clustering
10
1o20 40 60 so 100
NTowersFigure 4: The number of towers in the online reconstructed jets by the old
clustering (red line) and the proposed algorithm (black line) for jets above 40- - -- - - --- - - -
----~ ---- ---- ---- --
----- ---- ----------------- ------ - ---
. .
l..~- 0
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Flanagan, G. U. Level-2 Calorimeter Trigger Upgrade at CDF, article, April 1, 2007; Batavia, Illinois. (https://digital.library.unt.edu/ark:/67531/metadc889717/m1/2/: accessed April 25, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.