To get to my flasher photonics tables, click here.

New Photonics Features Documentation

What follows is the documentation for the new photonics features that I've implemented and that I'll talk about (or have already talked about) at the Berkeley Collaboration Meeting. These features include a new flasher LED source, a fully configurable flasher board source, and an option to simulate a hole ice cylinder of increased scattering when using the flasher board source.



Flasher LED:

Photonics already supports a regular LED source as source option #4. The photon direction is chosen from a cosq distribution in zenith angle and from a uniform distribution in j (both with respect to the source axis, i.e. the LED axis). The flasher LED's that are actually in the ice are considerably more beamed than this and are closer to Gaussian in q with a s of around 14 degrees.

In order to more realistically simulate flashers, I've added a flasher LED source option, available as source option #7 in the photonics steering file. The source can be rotated in the steering file just as any other photonics point source. Currently, each photon propagation involves one binary search of a vector in choosing the q direction for each photon. It's the same process as the one implemented for the EM and hadronic shower sources.

The difference between the regular LED and the flasher LED sources is clearly visible in arrival time histograms. It also makes a difference when simulating timing differences between different DOM's on a single string. It's not clear yet whether it makes any difference in inter-string calibrations.

Full Flasher Board Source:

I've also implemented a full flasher board source consisting of 6 horizontal flashers and 6 flashers tilted at 48 degrees. This is available as source option #8 in the steering file. Individual LED's can be turned on or off from an additional line in the photonics steering file. In each photon propagation step, one of the active LED's is chosen at random to fire. A photon direction is chosen locally just as for the single flasher LED source and then the direction is transformed into the global coordinate system.

The full flasher board source is never meant to be tilted in the photonics steering file. All the tilting has already been taken care of. Also, since we have 6 pairs of LED's distributed around the flasherboard at 6 different locations, photons will not be generated uniformly in j as they are for all other photonics point sources.

In the following images, a point is plotted on the unit sphere for the direction of each outgoing photon. The red lines show the positions of the LED source axes.

Above, only the 6 tilted LED's are turned on. We can clearly see 6 "lumps" that are rotationally symmetric about each LED axis and that are reasonably beamed.

This is the same as the image above but viewed from above. We can see that we get far fewer photons directly out of the top of the board.

Here, only the 6 horizontal LED's are turned on.

Here, the full board is turned on. As expected, we get many more photons in the upper hemisphere than in the bottom. Very few photons travel downwards.

Hole Ice Cylinder:

We know quite a bit about the ice properties of South Polar ice (i.e., see Kurt's page). In the process of inserting a string into the ice though, it's necessary to melt this pristine ice and let it refreeze. This introduces bubbles, and in general, there is a column of ice around the string that is different in its properties from the rest of the ice. Currently, this is taken into account in photonics only through the efficiency file that is applied to DOM sensitivity.

In order to see if we could learn anything about the nature of scattering in the hole ice, I've implemented a photonics option to simulate a hole ice cylinder of different ice properties around the string. It currently only works with the full flasher board source, option #8. From the steering file, one can also configure the radius of the cylinder as well as it's effective scattering length. The cylinder is infinite and uniform.

As of the Berkeley Collaboration Meeting, this has not been committed to the photonics CVS archive. It's implemented and fairly well tested, so it should be committed soon.

It's not clear yet how much of a difference including hole ice will make in measurements of things like timing differences on a single string.

The following images illustrate the hole ice scattering option. Each point is the location of a scattering point for a propagated photon.

Above we have a flasher board with all 12 LED's turned on. I've propagated 100,000 photons with a hole ice cylinder of 30 cm radius and 50 cm effective scattering length. The red points are scatters inside the hole ice.

Top view of a different hole ice simulation with fewer photons.

Same as the first image but with hole ice scattering turned off. There are 17% fewer scatters in the hole. I'm investigating why the scatters inside the hole extend further up than with the hole ice turned on. It may have to do with the propagation legth and how the absorption length gets set...

New Steering File Options

To configure the flasher board source and to utilize the hole ice cylinder, I've added four new lines to the photonics steering file. Existing steering files are unaffected by the changes I've made. The photonics reader module only looks for the existing lines if you've selected the full flasher board source (option #8) in the steering file. To see an example of a full steering file, click here.

The additional lines following the source type are:

000000 111111 #Flasher board configuration (6 horiztonal and then 6 tilted, separated by a space, 1=ON, 0=OFF, e.g. 111111 000000)
1     # Hole ice cylinder-1 to turn on, 0 to turn off
.3    # Radius of hole ice cylinder (m)
.50   # Hole ice effective scattering length (m)



The first line specifies the configuration of the flasher board.  The reader expects six 0's or 1's followed by a space and then
another six 0's or 1's.  A 0 turns a particular LED off while a 1 leaves it on.  The first six LED's are the horizontal ones and 
the second six the tilted ones.  In each grouping, the LED's start at 0 degress azimuthally and move around the board in 60 degree
increments.



The next lines configure the hole ice cylinder around the flasher board.  The first turns it on or off and the next two set its
radius and effective scattering length, respectively.  The cylinder is infinite and uniform in the vertical direction.



Good luck and happy simulating.  If you have any questions, don't hesitate to contact me.

Photonics: take the plunge...