[Halld-cal] Notes on Cabling and labeling of Fcal and Bcal
Elton Smith
elton at jlab.org
Wed Jul 20 09:06:26 EDT 2011
Hi Matt,
Thanks for the note. I did not know this area was not yet available
externally. I have copied the message from May 2008 below.
Cheers, Elton.
-------------------------------------------------------------------------------------
Suggestion for routing the Bcal and Fcal cables
To: halld-cal at jlab.org, halld-me at jlab.org, halld-electronics at jlab.org
Subject: Suggestion for routing the Bcal and Fcal cables
From: Elton Smith <elton at jlab.org>
Date: Sat, 10 May 2008 11:58:27 -0400 (EDT)
Reply-To: Elton Smith <elton at jlab.org>
Sender: owner-halld-cal at jlab.org
Dear collaborators,
During preparations for the system review we had some discussions about
cabling since how the cables are connected to the electronics can affect
the options for creating energy sums for the trigger. The engineering
staff is also starting to route cables and establish cable trays, so it is
not to early to consider their layout. Below are two suggestions for
discussion about how to map the Bcal and Fcal cables to the crates. These
cabling schemes are motivated by considerations for the trigger. Generally
a phi-symmetric trigger is desirable for our detector.
Comments/suggestions are welcome.
Cabling for the Bcal.
--------------------
General: For the trigger we want to cable opposite (left/right) ends into
the same fADC module so that sums will approximately compensate for
attenuation. Adjacent modules should be cabled together to create local
sums of energy deposition.
The crate sums are delivered to the system processors and therefore are
created naturally the trigger now, without the use of additional data
lines. Therefore, it makes sense to cable the crates in a way that might
be useful for the trigger.
One cal wedge: Outer 4 x 2 (left/rigth) = 8 inputs
Inner: 24 x 2 (left/right) = 48 inputs
-> One wedge fills 3.5 fADC (16 channel) modules.
-> Two wedges fill 7 fADCs
-> Four wedges fill 1 crate (14 fADCs). This leaves 2 empty slots, which
might be sensible from the heat dissipation stand-point, but also
associates a single crate with 1/12 of the Bcal (30 degress in phi).
With this scheme we need a total of 12 crates for the Bcal (6 north, 6
south). Note that 10.5 crates would be used if all slots were cabled
completely (16 fADCs per crate). However, the even number of 12 splits
evenly between north and south which would be required by the present
concept for the location of racks.
I do not know if there is already a convention for numbering wedges or
crates, but I assume it makes sense to tie them to the coordinate system,
were y is up and x is pointing north, and phi clockwise relative to x.
north crates phi (deg)
1 0-30
2 30-60
3 60-90
10 270-300
11 300-330
12 330-360
south crates phi (deg)
4 90-120
5 120-150
6 150-180
7 180-210
8 210-240
9 240-270
In order to ensure coverage across the boundary between crates, the
trigger would need to create overlapping crate sums. These would include
1+2, 2+3, 3+4, 4+5, 5+6, 6+7, 7+8, 8+9, 9+10, 10+11, 11+12, 12+1. This
granularity corresponds to an azimuthal angle range of 60 degrees. For
example [I'm guessing here] a cosmic-ray trigger could be formed in the
SSP using opposing regions such as 2+3 * 8+9 or 3+4 * 9+10 or 4+5 * 10+11.
This concept should be checked by the trigger experts.
Cabling for the Fcal.
--------------------
General: For the purpose of making coincidences between the Fcal and the
FTOF, one might consider cabling the Fcal in either vertical or horizontal
columns (rows). Background rates, however, are a very strong function of
the distance from the beamline, and so it makes more sense to combine Fcal
blocks within concentric rings into the same crate. This also keeps the
azimuthal symmetry of the detector which I believe is a very desirable
feature of the trigger.
A rough division of rings (within two radii) is given below, where each
block is 4x4=16cm2. Each crate is assumed to house 16 fADCs (256 blocks)
For 11 crates this corresponds to 2816 blocks (nominal number is 2800).
The last crate corresponds approximately to a single ring of blocks at the
outer layer of the Fcal.
crate area radius(cm)
(cm2)
1 4096 36.1081486
2 8192 51.06463346
3 12288 62.54114794
4 16384 72.21629719
5 20480 80.7402748
6 24576 88.44653962
7 28672 95.53318149
8 32768 102.1292669
9 36864 108.3244458
10 40960 114.1839917
11 45056 119.7571808
At the trigger level, different trigger thresholds or selection criteria
could be easily applied at the crate level. These thresholds should be
applied to everlapping sums to ensure that showers that straddle the
boundaries have the same efficiency in the trigger as showers in the
middle of a sum. This scheme with rings, requires the least number of
elements in the sum which is two. [Note that cabling in for example
quardrants will require four elements into the sum, since a shower could
in principle hit at the corner of for regions].
Energy sums which can be used for selection with a threshold would be the
sum of two crates, i.e. 1, 1+2, 2+3, 3+4, 4+5, 5+6, 6+7, 7+8, 8+9, 9+10,
10+11, 11+12, 12. This granularity provides a reasonable granularity even
for the simplest trigger scheme which is presently proposed.
This scheme requires cables around an entire Fcal ring routed to a single
crate. In particular this means that cabling would not be partitioned
north and south which might be the most elegant cabling solution.
-----------------
Elton Smith
Jefferson Lab MS 12H5
12000 Jefferson Ave
Suite #16
Newport News, VA 23606
(757) 269-7625
(757) 269-6331 fax
On 7/20/11 8:51 AM, Matthew Shepherd wrote:
> The linked email is not accessible.
>
> -Matt
>
> On Jul 19, 2011, at 8:29 PM, Elton Smith wrote:
>
>> Dear collaborators,
>>
>> I had promised to send out a summary of my understanding of the labeling for Bcal and Fcal detectors. Fernando's 1 page summary is attached and gives the previously agreed labeling scheme for both detectors.
>>
>> The motivation and discussion of cabling related to the trigger can be found on an archived e-mail
>> http://www.jlab.org/ccc/mail_archives/HALLD/halld-cal/msg00795.html . Note that the numbers for the Bcal must be reduced by the granularity of the Bcal sums (see below), but the strategy remains the same.
>>
>> The numerology for the Bcal is given in the lists below for the current nominal configuration and how they would change if an additional inner layer is added. For the FADCs, one might be able to get away without adding any more crates, but the crates would be completely filled if the additional layer is added. For the TDCs, I think would need to add two crates (one on each side of the beamline), although the crates would only be a little over half full.
>>
>> Given that the cabling into the electronics depends on the trigger, which has not yet been completely defined, and the uncertainly on whether an additional layer would be added, I do not think we can specify completely the electronic assignment of each detector element at this time. Therefore, I suggest that for the moment we label the cables with their detector labeling only.
>>
>> Cheers, Elton.
>>
>>
>> -----------------------------------------------
>> Nominal configuration
>>
>> FADCs (16 inputs)
>> Outputs per Bcal 2x12
>> Number of slots per crate used 12
>> Number of Bcal mods per crate 8
>> Number of crates north 3
>> Number of crate south 3
>>
>> TDCs (32 inputs)
>> Outputs per Bcal 2x8
>> Number of slots per crate used 12
>> Number of Bcal mods per crate 24
>> Number of crates north 1
>> Number of crates south 1
>>
>> Adding one radial layer
>>
>> FADCs (16 inputs)
>> Outputs per Bcal 2x16
>> Number of slots per crate used 16
>> Number of Bcal mods per crate 8
>> Number of crates north 3
>> Number of crate south 3
>>
>> TDCs (32 inputs)
>> Outputs per Bcal 2x12
>> Number of slots per crate used 9 (over half full)
>> Number of Bcal mods per crate 12
>> Number of crates north 2
>> Number of crates south 2
>>
>> Cheers, Elton.
>>
>>
>> Elton Smith
>> Jefferson Lab MS 12H5
>> 12000 Jefferson Ave
>> Suite # 16
>> Newport News, VA 23606
>> elton at jlab.org
>> (757) 269-7625
>> (757) 269-6331 fax
>>
>> <HallDNamingConventions.pdf>_______________________________________________
>> Halld-cal mailing list
>> Halld-cal at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/halld-cal
More information about the Halld-cal
mailing list