sPHENIX tracking discussion

[Thomas K Hemmick <hemmick AT skipper.physics.sunysb.edu>]

Hi Dave

Thanks very much!  I found 831 in fortran Friday afternoon and used extern statements to make it callable from C++.  Even though fortran supports things like functions, 831 is written with the return value as one of the arguments.  I'll be delighted to switch to the fable output as it might make the code much more elegant.

Thanks!

Tom

On Sun, Dec 6, 2015 at 3:32 PM, David Morrison <dave AT bnl.gov> wrote:

Hi all,

Tom mentioned that he was looking for a C++ implementation of the modified Bessel functions of (purely) imaginary order to carry out the E_phi calculations from Stefan Rossegger’s thesis for different boundary conditions.

Who would’ve guessed there were any new developments in the calculation of special functions any more?

It looks like ACM TOMS algorithm 831(2004) is what you want.  Yes, it’s in Fortran 77, but that may be OK.  The sources can be downloaded from here http://www.netlib.org/toms/831.  The source compiles fine and there’s an included driver to check the output against values provided by the authors.  I see a few differences in the output – most (but not all) of them seem to be deep in the decimals.  I converted the F77 to C++ using fable (http://cci.lbl.gov/fable/) which worked mostly fine – I needed to address a couple of complaints from the compiler.  That then compiles fine as C++.  I used the C++ to plot Ki(x) and compare to what I get from Wolfram Alpha – those are attached.  Looks OK.  Maybe this is a bit of a run-around for something that “is of second order importance compared to E_r”, but it can be made to work.

Cheers,
Dave

Ki(x) from Wolfram Alpha



Ki(x) from ACM TOMS algorithm 831 converted to C++:




> On Dec 5, 2015, at 10:33 PM, Frawley, Anthony <afrawley AT fsu.edu> wrote:
>
> Tom Hemmick - TPC ion backflow and space charge calculations:
> --------------------------------------------------------------------------------
>
> Tom described the effort that is ongoing to simulate space charge effects in the TPC in sPHENIX.
>
> The bottom line is that first results with a new code are expected in the coming days.
>
> Slide 19: Discussion about the modified gating grid idea suggested by Dick Majka.
> The mesh spacing is a variable that has to be optimized. The estimate of 80-90% efficiency depends on how many mesh layers are distributed along the graded voltage, and how much of the tail is collected.
> Normally, the electric field lines would go through holes in mesh, the gate is operated by turning on the kicker so that the mesh collects ions, and how long it takes to collect the ions depends on how far to the nearest mesh.
> In principle, we could install the mesh and not use it. But there may be complications.
>
> Slide 20: Expect to get results in the next few days.
> It is possible to use Tom's configurable new code to calculate results for the ALICE case, as a check on the code.
>
> Come back in 2 weeks with results!
> Carlos will be the speaker.

David Morrison  Brookhaven National Laboratory  phone: 631-344-5840
                                Physics Department, Bldg 510 C  email: dave AT bnl.gov
                                Upton, NY 11973-5000






_______________________________________________
Sphenix-tracking-l mailing list
Sphenix-tracking-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/sphenix-tracking-l