Re: A logic code error (possibly?)

From: Joseph Comfort <Joseph.Comfort_at_asu.edu>
Date: Fri, 02 Jul 2010 10:21:35 -0700

Hi Vasilis,

I like your figure! It gives me confidence. I am using the latest
versions of the codes. Fluka is V2008.3c.0 (Nov. 2009) and Flair is
V0.8.2 (May 2010).'

Were you perhaps using the latest patched version of Fluka. The missing
strip in my figure is where the error messages had claimed points were
in 2 different regions. The messages disappear when the WHAT(5)=2, and
the code appears to behave sensibly. When will the respin be released?

I'm moving on to production.

With regards,
Joe

Vasilis Vlachoudis wrote:
> Hi Joseph,
>
> are you sure you are using the latest version of FLUKA, I've made the
> same plot and I got the attached figure.
> However you have be careful that the plots of the geometry is not a
> proof that the geometry is ok. For example overlapping regions will not
> show up correctly.
>
> cheers
> vasilis
>
> On 07/02/10 09:53, Joseph Comfort wrote:
>> Hi Vasilis,
>>
>> I made a typing error in my message about the WHAT(5): it is and was on
>> the GEOBEGIN card, not the GEOEND card. It is true that WHAT(5)=2
>> eliminates the error messages.
>>
>> However, as noted earlier, the figure I get from Flair shows a strip of
>> the Wedge region is absent. It includes the steel part of the Wedge and
>> a part of the vacuum hole. The figure is attached again. I checked
>> again, with some tweaks on the file, and the problem still exists. (If
>> the alternate coding in the .inp file is used, the wedge disappears
>> entirely from the figure. I think the two codings should be logically
>> the same.)
>>
>> Contrary to what I wrote before, Fluka runs and particles get all of the
>> way through the geometry. (I had not looked in the right way before.)
>> But I am uncertain as to how reliable the results are. They look
>> sensible, but... I don't know if Fluka or Flair is more at fault in
>> removing the strip from the figure.
>>
>> I think I now have a geometry file (including ARBs, ARBs inside ARBs,
>> and BOXes) that works and does not need a huge number of significant
>> digits (perhaps even fewer than I am using). It has been tested with a
>> few million events and is ready for production with hundreds of millions
>> of particles.
>>
>> How worried should I be about the 'glitch' in the figure? My near-final
>> .inp file is attached.
>>
>> With regards,
>> Joe Comfort
>>
>>
>> Vasilis Vlachoudis wrote:
>>
>>
>>> Hi Joseph,
>>>
>>> The WHAT(5)=2 is on GEOBEGIN card not on GEOEND.
>>>
>>> I've made a geometry debugging test on the region
>>> X[-10:10]x101 Y:[0:0]x1 Z:[550:590]x101
>>> and with WHAT(5)=3 and I get the following errors
>>> **** Lookdb: Geometry error found ****
>>> **** The point: 2.87128713 0. 561.089109 ****
>>> **** is contained in more than 1 region ****
>>> **** (regions: 47 HoleD2a 49 Wedge ) ****
>>>
>>> while setting to WHAT(5)=2 I get no error in that region
>>>
>>> Regards
>>> Vasilis
>>>
>
Received on Mon Jul 05 2010 - 17:50:04 CEST

This archive was generated by hypermail 2.2.0 : Mon Jul 05 2010 - 17:51:04 CEST