Dear FLUKA users,
I am trying to relaunch some simulations that last completed several years ago. After updating a couple of points in the .inp file to comply with the current version of FLUKA, I tried to run it (in flair on WSL/Ubuntu 20), but it crashed with an error that I unfortunately can’t really make sense of.
The magnetic field maps are loaded and the simulation seems to run for a very short time, but then it would crash with a floating point exception noted in the .log file.
The .log file is attached.
lhcb_UPGRADE_MK1_14TeV_19.4.2021_RICH_PEshield001.log (1.6 KB)
The core dump refers to pho_abort.f at pho_event.f:155, which seems to be the following:
pho_event.f:
SQS = SQRT((P1(4)+P2(4))**2-(P1(1)+P2(1))**2-
& (P1(2)+P2(2))**2-(P1(3)+P2(3))**2)
IF ((SQS > DEPS).AND.(SQS >= SQSGLOBMAX)) THEN
WRITE(LO,*)
& 'PHO_EVENT: Error: Higher energy requested then sqsglobmax.',
& SQS,'>', SQSGLOBMAX
CALL PHO_ABORT
END IF
!**anfe Safety margin for initialization energy is 20%
SQSGLOBMAX = 1.2D0*SQRT((P1(4)+P2(4))**2-(P1(1)+P2(1))**2-
& (P1(2)+P2(2))**2-(P1(3)+P2(3))**2)
Because of the code seemingly referring to the primary beam parameters, I was suspecting that maybe it also could be connected to the SPECSOUR card. This has changed a bit since I last used it, as I noticed an addition of a second continuation card since my last runs.
For the slim chance that this might have to do with my problem, regarding the new options in flair I’d also like to ask the following:
When using PPSOURCE for actual pp collisions (7+7 TeV, head on pencil beams), with a BEAM and BEAMpos card, do I still have to select the atomic (&mass) number of the second particle (“A:” and “Z:” in flair) or would the default still be a proton?
Also, since flair shows all 3 options of NonElastic, Elastic and EM dissociation as OFF instead of empty/default, do I have to actively select them or does this represent the default described in the manual?
The main cards for my simulation (without geometry) are as follows, please let me know if you need more.
(The full simulation needs several user routines, and magnetic field maps which are loaded via OPEN cards, but I can provide this if needed.)
GLOBAL 5000. DEPRBODY
OPEN 33. READONLY
/FLUKAfiles/lbfluka-masterRun319.4.2021/MagFieldMaps/c1_onemap_down_c.cdf
OPEN 34. READONLY
/FLUKAfiles/lbfluka-masterRun319.4.2021/MagFieldMaps/c2_onemap_down_c.cdf
OPEN 35. READONLY
/FLUKAfiles/lbfluka-masterRun319.4.2021/MagFieldMaps/c3_onemap_down_c.cdf
OPEN 36. READONLY
/FLUKAfiles/lbfluka-masterRun319.4.2021/MagFieldMaps/c4_onemap_down_c.cdf
*
* Next options are for activation studies
PHYSICS 3. EVAPORAT
*USERWEIG only for PMI scoring in combination with modified fluscw.f ic1.f
USERWEIG 3. 0.0
PHYSICS 1. COALESCE
IONTRANS HEAVYION
RANDOMIZ 1.0 46852433.
USRICALL
DEFAULTS PRECISIO
* Calls subroutine mgfld.f
MGNFIELD
BEAM 10000.0 0.0 PROTON
BEAMPOS 0.04 0.001 1. 0.0 0.0
* Magnet UP&DOWN 14TeV pp
SPECSOUR 0.0 0.0 7000.0 0.0 0.0 -7000.0PPSOURCE
SPECSOUR 0.005 0.004 4.1 10. &
...
EMFCUT -0.001 0.0001 0.00001 BLCKHOLE @LASTMAT 1.0PROD-CUT
EMFCUT -0.001 0.0001 R1 @LASTREG 1.0
Thanks in advance for any help!