|
To set the delay, do the following:
-
Ensure that PMT pulses come after the external trigger.
Look at the external trigger (before it is sent through the ground
isolator or the delay) on an oscilliscope along with one (or more)
of the PMT signals that you want to acquire, as demonstrated in
Figure 1. Verify that PMT pulses
are actually coming (that is, that your laser or LED is flashing),
and that those pulses arrive after the external trigger. If
they don't you will need to adjust the way the trigger pulses are
produced or the length of the cable until they do.
-
Ensure that the PMT pulses come after the delayed external trigger.
Do the same as before, but look at the external trigger after it comes out
of the delay and before it is sent to the NIM-ECL converter. This verifies
that pre-trigger electronics (i.e. the transformer isolator and/or gate
generator) are working.
-
Send the external pulse to the trigger.
Connect the delay output to the proper input on the NIM-ECL converter. You
may wish to send one of the other delay outputs to the oscilliscope to
check yourself later on.
-
Read the trigger clock on the scope.
The clock on the trigger is fanned out to all the FBE crates on the same
ribbon cables that carry the trigger signals. There is one spare ribbon
cable resting above crate 4 that is used for diagnostic purposes for ID
triggers (see Figure 2). For OD calibrations, it will be necessary to
use instead the
thinnest of the ribbon cables connected to the back of the OD crate.
In both cases, the two wires on the side of the cable colored brown carry the
clock signal, as shown in Figure 3. To read the clock, you need to connect it
to the scope via
one of our make-shift differential probes. These probes, one of which is
shown in Figure 4, consist of two
wires connected to either side of a resistor on one end, and to a black
connector at the other end. Connect the black connector to the clock wires
on the ribbon cable using a female-to-female ribbon cable connector, or just
using two wires if the connector cannot be found. Insert the ends of the
resistor into two of the oscilliscope inputs, and connect the ground of the
probe to the oscilliscope ground, as shown in Figure 5. The 40 MHz clock signal
should be visible
on both scope inputs (see Figure 6), although one should be the negative of the
other.
(Note: to read the "true" differential signal, one would need to subtract
one of these signals from the other, but that is not necessary for our
purposes).
-
Run the trigger and find the trigger signal.
Login as daq on hoots11 (if you don't know the password, probably you
shouldn't be messing with the trigger on your own. Get some help). Go
to daq/sys/KamDAQ/Trigger/Scripts. Copy KamTrigger.kts to
KamTrigger[your name].kts and open the new file in your favorite editor.
Find the declaration of the variable "enable_calibration_ft" and initialize
it to "1" instead of "0". Then find the declaration of the variable
"calibration_source" and initialze it to the ECL input number for your signal.
Save the file, and at the command line type
"tinykinoko KamTrigger[your name].kts -f". The trigger should now be sending
signals to the ATWDs along the same ribbon cable from which you read the clock
earlier. Move the probe to the next wire over from the clock and look for the
signal on the scope. If you cannot find it, ask a trigger expert.
-
Set the delay.
Look at the trigger pulse from the ribbon cable on the scope and connect a pmt
signal immediately before it is sent into an ATWD input. Adjust the delay on
the gate generator until the earliest pmt signal arrives ~40 ns after trigger
pulse. If the delay is set to its minimum and the PMT pulses are still coming
too early, try by-passing the gate generator entirely. If the PMT pulses are
still too early, you will need to modify the way the external pulse is created
or shorten the cable length to make it arrive earlier.
-
Check with other PMTs
It is a good idea to check the timing with several PMTs in different regions of
the detector. Just keep in mind that if you need to connect an extension for PMT
cables that won't reach the scope, you will have to add 5 ns per meter to get
the timing right.
-
Check the timing on any other input signals.
Some runs, for example US and Japanese laser runs, have required an additional
signal to be sent to the ATWDs. In order for these signals to be captured, they
must be timed to arrive coincident with the PMT signals. This timing should be
adjusted after the delay for the external pulse has been set as described above.
-
Clean up
End tinykinoko on hoots11 by pressing ctrl-c. Copy your KamTrigger[your name].kts
script to your directory if you would like to keep it, but delete it from the
trigger scripts directory on the daq account to keep it clean. Next return all
PMT cables back to their proper inputs. If you make a mistake, we will not take
data from those tubes until your mistake is discovered. So if you are unsure of
yourself, find an expert to help you. It is also important that you do not tangle
the cables up. The cables from one board must not be entangled with those from
another board so that the boards can be completely removed without disconnecting
cables. If you do get the cables tangled, the FBE experts will be upset with you.
Return the make-shift differential probe and it's connector/wires to where you
found them. Clean up any other cabling mess and turn off the scope.
|
Figure 1: a PMT pulse following an external trigger
Pink: external pulse from the OD LED flasher.
Yellow: signal on an OD PMT
Figure 2: the location of the spare ribbon cable that sends clocks
and trigger commands to the FBE crates. This ribbon cable should be
used for setting up ID calibrations. For OD calibrations, use a
similar cable connected to the back of Crate 6.
Figure 3: a close-up of the ribbon cable with the female-to-female
connector attached.
Figure 4: the make-shift probe used to examine the clock and
trigger pulses on the scope.
Figure 5: connecting the make-shift probe to the scope
Figure 6: the clock signal
|
