The
simple
implementation of FPD L1 Trigger is based on detector geometry
only, with no confirmation of individual tracks and no sorting by
|t| or
xi. It only requires that both
position detectors that form a spectrometer (
e.g. P1U and P2U) have at least one
valid hit.
A valid hit is ideally defined by simultaneous signals in fibers of the
three different views,
U,
X and
V, which spatially superpose to each
other, as we expect a particle traversing the detector to leave a
signal in at least one fiber per view. Nevertheless, we can think of
relaxing this requirement in order to allow coincident signals in only
2 out of the 3 views (
e.g. U and
X only, with no
V fiber having a signal above
threshold) do be accepted as a valid hit.
New equations have been generated to implement these two schemes, which
are referred to as
3 out of 3
(3/3) and
2 out of 3 (2/3)
schemes. Following the strategy used before, we write the trigger
equations in terms of segments, instead of fibers themselves. This
makes them simpler (of course, first we use the fibers to define the
segments). The equations can be implemented for fine segments (1/3 of a
fiber width) as can be for wide segments (4/3 of a fiber width,
presently).
Links to the VHDL files implementing the two schemes
for wide segments are given next.
vw2x3.vhd
- File with the VHDL code implementing the
2/3 scheme
for wide segments
vw3x3.vhd
- File with the VHDL code implementing the
3/3 scheme
for wide segments
More details can be found at:Test vectors will be used for test
purposes in the FPD L1 Trigger implementation.
In their simplest form, for a single detector, they are combinations of
U , U' , X , X' , V ,
V', representing the fibers turned ON/OFF. For instance, '
19 0
2 2 10 0' stands for fiber U(
19)=ON, X(
2)=ON, X'(
2)=ON, V(
10)=ON and no fiber turned ON
for planes U' and V'. This kind of vector has one hit at most per
plane, but we can build more complex vectors by mixing two or more
test vectors.
In this section we provide test vectors properly formated to be used in
the test stand (CTS). It consists in strings of hexadecimal numbers,
named
frames, like the
following one,
00000040 00000000
01000002 00000000 00000000
00000000 00000000 00000000 00000000 00000000
representing data coming from AFE to DFE. Each group of 8 hex numbers
in a frame represents data coming from a single detector, through one
link. For the present
tests, only
link 0 (1st
link) and
link 2 (3rd
link) carry data, indeed. Each of these links brings in data from one
of the two detectors comprising a spectrometer. Seven consecutive
frames make up a whole
event, as shown below.
Detector 1
Detector 2
00001002 00000000
01100100 00000000 00000000
00000000 00000000 00000000 00000000 00000000
01000000 00000000
00000000 00000000 00000000
00000000 00000000 00000000 00000000 00000000
00000000 00000000
00000040 00000000 00000000
00000000 00000000 00000000 00000000 00000000
00080000 00000000
00000000 00000000 00000000
00000000 00000000 00000000 00000000 00000000
00000000 00000000
00000010 00000000 00000000
00000000 00000000 00000000 00000000 00000000
00000004 00000000
00000000 00000000 00000000
00000000 00000000 00000000 00000000 00000000
00000001 00000000
00000001 00000000 00000000
00000000 00000000 00000000 00000000 00000000
22-Jun-2004
Here, we make two sets of files
available with test vectors: one, named good, has 11
test vectors corresponding to valid combinations of U, U', X, X', V, V'
fibers in both detectors, that will cause trigger to fire; the other
one, named bad, contains a
single invalid combination, that will not trigger both detectors of a
spectrometer.
In each
case we provide:
- the formated file (.tv)
to be used in the CTS;
- its 2 X 6 fibers decimal
representation (.dec);
- its 2 X 6 fibers binary
representation (.bin).
Links for new sets of test vectors
are given in the table below. Vector names will begin with "tv" string and follow the
convention: