User name R. Michaels
Log entry time 14:10:27 on April25,2008
Entry number 234419
This entry is a followup to: 234357
Followups:
keyword=helicity crisis, recommended actionsI looked at the helicity circuit in the counting room. For
background see halogs 234357 and 234120 and many others.
The circuit is far more complicated than it needs to be -- see fig 1.
We could clean it up by making the changes shown in red. I did not
make this circuit. Amazingly though, it does produce a signal that
looks close to the true helicity -- the result is the same polarity
but shifted 360 microseconds. This shift is sloppy but should be
survivable since we blank off 500 usec in between pulses anyway.
Note, Moller signal comes more directly from the fiber (see fig 1).
I do not see any coupling of the HAPPEX DAQ to this circuit, so the
observations that others made about interference with that DAQ make
no sense to me. However, there are lots of things that can go wrong
with gate generators and they are not necesssary.
Recommended action:
1. Simplify the L-HRS helicity by moving to a direct path (see
red dotted line in figure).
2. Otherwise do not change the circuit -- I think the experimenters
may want to do "archaeology" on it this summer. However, this
summer we should really clean up this crate !!
3. When there is no beam, do the following short runs with pulser
trigger T8. Observe the helicity signal at the location indicated
in the fig. Each run is 5 minutes. Page me if you want me to do it.
a) Run with pulser T8, observe good baseline performance.
b) Repeat with helicity to L-HRS unplugged. Thus prove that this
is indeed what we are looking at ("idiot check").
c) Repeat a) with helicity restored.
d) Repeat a) with HAPPEX DAQ running. Observe on scope.
e) If problem is observed, repeat d) by using the recommended
changed path of the helicity signal. And chase the problem ...
Another possibility is to repeat d) with the NIM crate that is
above and to the right turned off, but I don't see any coupling
from this. The possiblity of interference on L-HRS exists but
seems remote to me (no HAPPEX DAQ there) -- e.g. if two signals
are plugged into the input of a logic module. Finally, running
a redundant helicity signal ("second opinion") to the L-HRS or
to the Bigbite crates during an upcoming downtime is highly
recommended.
FIGURE 1
