<html>
<head>
</head>
<body text="#000000" bgcolor="#ffffff">
Nathan<br>
<br>
<ul>
<li>First, I was wondering why a hit requires at least two
photoelectrons (in hitCerenkov.c), why not just one?</li>
</ul>
I considered that the randoms from single-photoelectron counting
would be too high to be usable. These are large phototubes and they
will probably have high single-photoelectron counting rates.<br>
<ul>
<li> Next, what is the origin of the 25% efficiency factor which
is applied when storing Cerenkov photons to the stack (in
gustep.F)? I understand why an efficiency factor is needed, but
why this particular value?</li>
</ul>
This is the photon detection efficiency of the phototube, typically
between 20% and 30% in the visible. Cerenkov light peaks in the
blue, where the numbers are closer to 20% for typical bialkali
phototubes, but I chose 25% to be as generous as I thought
realistically possible. The other important number is the material
of the phototube window that determines how much of the Cerenkov
spectrum gets through. I assumed quartz windows (fused silica)
which is a reasonable compromise between efficiency and cost. <br>
<ul>
<li>In your description of the Cerenkov Counter in the CD2
Baseline document (November 16, 2006), you mention that the
mirror optics are not optimized, making the simulation
unrealistic. Do you recall how large of an effect this had on
the number of detected photons? </li>
</ul>
The mirror geometry was worked out by Eugene Chudakov, and
implemented in Geant by me. I think that comment just means that
this is a first-pass attempt to make focusing optics for this
geometry, and that it can probably be improved by studying light
paths generated by the particular particles of interest for Cerenkov
PID. The thing about optimization is that one does not know how
much improvement is possible until the improvement can be
demonstrated.<br>
<br>
-Richard J.<br>
<br>
On 3/20/2011 9:09 PM, Nathan Sparks wrote:
<blockquote type="cite" cite="mid:f750dcc692a6.4d866d10@fsu.edu">
<pre wrap="">Hi Richard,
I have some questions about the Cerenkov simulation. First, I was wondering why a hit requires at least two photoelectrons (in hitCerenkov.c), why not just one? Next, what is the origin of the 25% efficiency factor which is applied when storing Cerenkov photons to the stack (in gustep.F)? I understand why an efficiency factor is needed, but why this particular value?
In your description of the Cerenkov Counter in the CD2 Baseline document (November 16, 2006), you mention that the mirror optics are not optimized, making the simulation unrealistic. Do you recall how large of an effect this had on the number of detected photons? I am now thinking that this is why I am seeing so few hits. Finally, do you have any suggestions for a method to use when optimizing the geometry?
Regards,
Nathan
</pre>
</blockquote>
<br>
</body>
</html>