Linear Collider Forum



Home » Simulation » Full Simulations » Protoype Simulation - Cell Numbering and Coordinate Frames
Protoype Simulation - Cell Numbering and Coordinate Frames [message #105] Wed, 09 June 2004 08:04 Go to previous message
musat
Messages: 57
Registered: February 2004
Hi,

It was suggested previously to move this discussion to the full simulation forum.

Here at the LLR, we wrote some suggestions concerning the coordinate system and the cell numbering. The document is

http://polywww.in2p3.fr/~musat/Proto/Ecal.html

I'm currently writing the Ecal prototype implementation and would like to make a new Mokka release this month, so perhaps we could try to find an (at least first) agreement on these points in a delay of, say, ten days.

What do you think about that?

Cheers,

Gabriel

============================================================ ===
Roman wrote:

Dear Colleagues,

I would like to pick up the discussion on coordinate frames again. First of all thanks to all of you for contributing to the discussion
and sorry that I didn't react earlier to all the E-Mails.

I guess we have agreed ...

- to align the detectors along the z-axis. To other items
see my answers to Davids E-Mail below

- to try and find an as far as possible common index scheme to be packed into one LCIO word

A good time to settle the discussion is the Calice-Meeting by the end of June. Should we strive for a written proposal to
be presented to the other Calice members ? If yes, who's going to write it (Gabriel was proposed, I would volounteer, too) ?
I am quite sure that any proposal will trigger a lively discussion at CERN .
Once the conventions are fixed it might be good to determine a time when
all the outcome has to implemented into Mokka.

Looking forward to your comments.

Cheers,

Roman

David Ward wrote:

> Hi Roman,
>
> Sorry, I should have replied sooner. A few comments below...
>
> On Thu, 6 May 2004, Roman Poeschl wrote:
>
>
>
>> Dear Collegue,
>>
>> over the last few weeks people at NIU and DESY have been working on the
>> implementation of the hadronic tile calorimeter into the MOKKA simulation
>> framework. A number of points have come up which should be - we believe -
>> coordinated with the other groups doing simulations for the test
>> calorimeters.
>>
>> The main issues are
>> - a common coordinate system
>> - (possibly) a central cell numbering scheme for the prototype.
>>
>> We are aware of at least two different coordinate systems currently in
>> use. Obviously we all would profit very much from a common system. I like
>> to propose the following convention:
>>
>
>
> I agree we should harmonise things, and try to settle it once and for all.
>
>
>
>> * the z-axis is defined as the direction of the beam.
>>
>
>
> This seems the most natural choice, though the LLR-defined prototype in
> Mokka uses y.
>
>
>> * We use a standard right handed coordinate system
>> * the origin of the system is the symmetry point of the calorimeter in x,y
>> * the z=0 point is at the face of the ECAL.
>>
>>
>
>
> For z=0, what exactly do you mean by the face? You mean the first layer
> of any material in the ECAL (epoxy, probably)?
>
>

Yes, I mean the first layer of any material. The z-Position of the first layer in the ECAL would then be
z_1stlay = 0 - (Layer-Dimension)/2. or similar.

In the HCAL we currently have put the z-position of the layer into the middle of the scintillating tiles.


> For (x,y) I assume you mean the symmetry point of the HCAL? The ECAL is a
> bit more complicated because of the staggering of layers of Si, but if we
> specify (x,y)=(0,0) in terms of the HCAL, the ECAL just has to be aligned
> w.r.t. the HCAL.
>
>
Yes, I mean the symmetry point of the HCAL.

>
>
>
>> To make a fast access to the cell possible a cell numbering
>> scheme is needed.
>> We propose a simple numbering scheme, based on cells of 1x1cm2 for the
>> HCAL. The numbering starts at the lower left corner of each layer.
>> * i: row of cell
>> * j: column of cell
>> * k: layer number of cell
>> We propose to start the numbering of layers new for each subdector type.
>> (ECAL, HCAL, Tailcatcher). The three indices should be packed into
>> one word for output.
>>
>
>
> Starting at 0 in each case, or 1?
>
>
I would always start with 0.

> The existing ECAL scheme has two levels, the wafer indices and the pad indices within each wafer. I assume you are proposing a single index in each direction instead. That's probably OK, I just wanted to be clear.
>
>
Yes, I would prefer this.

Maybe the ECAL experts can work out a numbering scheme which is oriented on what was discussed so
far and which is best for their needs.

> I'm not quite clear about the 1x1 cm cells. These are meant to be virtual
> cells, I assume, at least in the HCAL. In the ECAL how do these map onto
> the physical pads (because of gaps between wafers, staggers between layers
> etc). Just need to clarify this. Maybe in the ECAL they are just the
> physical pads. How about the HCAL - do 1x1 cm cells map exactly onto your
> complicated pattern of different-sized cells? Are there issues relating
> to gaps between tiles, of are the gaps always exactly on a 1 cm grid?
>
>
We can map the 1x1 virtual cells onto the geometry of the 'real'-HCAL. The cell dimensions of the
real HCAL are integer multiples of the 1x1cm2 geometry. In first approach there will be no gaps between
the tiles in lateral direction.

>
>
>> To make exchanging and using data simpler, it would be great if
>> all groups could agree to a common index scheme and also to a
>> common packing scheme of these indices into one LCIO word.
>>
>>
>
>
> I agree.
>
> So, I have no problem with this in principle, I just think we have to make
> any proposal absolutely clear, and be sure it does all we need.
>
> Best regards,
>
> David.
>
>
>
> +----------------------------------------------------------- --------+
> | Dr. David R. Ward University of Cambridge, |
> | E-mail: drw1@hep.phy.cam.ac.uk Cavendish Laboratory, |
> | Phone: 44 1223 337242 (Room 939 Rutherford building) |
> | or 44 1223 335630 (College) Madingley Road, |
> | Fax: 44 1223 353920 Cambridge, CB3 0HE, U.K. |
> +----------------------------------------------------------- --------+
>
>
>
 
Read Message
Read Message
Previous Topic:Simulation Requirements Document
Next Topic:Protoype Simulation - Cell Numbering and Coordinate Frames
Goto Forum:
  


Current Time: Wed Apr 25 03:51:37 Pacific Daylight Time 2018
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.1.
Copyright ©2001-2010 FUDforum Bulletin Board Software