Optimisation routine for lightguide coefficients
End of June 2006 I have asked our Edinburgh workshop to build two of the lightguides as shown as dark blue line in the drawing to the right, each will be 50mm thick. Slight changes like a few added holes suggested by the workshop people do not affect the optical focussing.
To obtain the shape I used a macro routine to actually optimise the lightguide shape.
Run the macro from within PHYSICA. The optimisation is one iteration only, check the quality in feeding the out values into a second iteration as input values.
For the PDF writeup and the tarball with all relevant files please see:
- 09 Jul 2006
Some results on the focussing lightguides. Requirements for the two design options
below were a focal plane perpendicular to the disc and a high degrees/mm dispersion.
* compact focussing lightguide:
: compact focussing lightguide
The black dots are focal points on a particular ray. Ideally they would all lie
on top of one another. The position of a focal point actually does imply the
slope of the focal plane in its vicinity. It only can be changed at the cost
of imperfect focussing. The focal plane was twisted to vertical orientation
to allow photon detectors to be oriented parallel to the magnetic field.
The shorter the focal length, the bigger the focussing imperfections both
absolute (in millimetres) and relative (in degrees or beta). The aim was
here to get the dynamic range of 20degrees (45-25) into less than 50mm range,
so the focal plane could be covered by a standard 50mm or 2" side length PMT.
Focussing polynomial is y= -5.4/60^2*x^2 - 0.9/60^3*x^3 - 0.5/60^4*x^4
the focal plane is at x=77mm, these values can be further optimised.
Focussing imperfections [sigma]:
45deg 0.15mm ; 35deg 0.16mm ; 25deg 0.47mm .
Including an additional mirror surface into the lightguide one can keep the
readout completely upstream or downstream of the disc. It may be easier to
accomodate spacial constraints this way than fighting for extra space.
- 05 May 2006
I still think that the Barrel can probably be read out with Time-of-Propagation.
But if photon angle measurement is required, the end of the bars
could possibly look like this with a focussing lightguide, and maybe one can get
away without correcting the fused silica dispersion:
I am not happy with the focussing (sigma=0.3-0.5mm, quick hand optimisation only),
and the vertical focal plane is not the best orientation either, only a compromise
propagated from the Disc calculations above. Getting the focal plane perpendicular
to the magnetic field may be necessary, which is good news as best focussing is
achieved with somewhere between 60-80degrees and not the 90 degrees. No constraint for the focal plane orientation would be even better news...
- 06 May 2006
Photos of first Prototypes
Two focussing lightguides still assembled from polishing sandwiched between
thin protective plates. Some polishing material has squeezed into the gaps.
Disassembly will take place once one has decided one does not want further
Each acrylic glass lightguide is 50mm thick.
- Photo 1:
- Photo 2:
- Photo 2:
- 28 Jul 2006
This page is part of DesignItems.