[Halld-cal] comments to BCAL calibration

[Sean Dobbs]

Hi Elton,

Thanks for your close reading of the document,  I'm not going to be able to
attend tomorrow's Calorimetry meeting due to a conflict, but I'd like to
make a couple comments to help clarify the situation (at least as I
understand it).  I'll restrict myself to discussing Table 2, since as you
said the information in the first table should be stored in the online
databases which are still under development.

You're right that in the end we want a set of constants that represent our
best understanding to be applied during reconstruction (e.g. ADC gains,
pedestals, and time offsets).  My understanding is that it would be useful
to keep values derived from specific data sets in the database as
intermediate steps, to allow for communication between different
calibration programs.  For example, maybe you have one program that
determines ADC constants from cosmic data, another one that determines them
from pi0 and eta decays, and then a final one that combines this
information obtain the final set of constants. In this case, the
ADC_physics_XXXX would be the per-channel values determined from the pi0
and eta decays. Of course, the details of what intermediate information to
save should depend on the details of the procedures...

I'd suggest that since the CCDB allows arbitrarily complex pathnames for
its tables, that any of this intermediate information be kept in
subdirectories (e.g. "/BCAL/cosmics/ADC_gains"), to reduce confusion.


Cheers,
Sean



On Tue, Jan 28, 2014 at 11:39 AM, Elton Smith <elton at jlab.org> wrote:

> Hi Zisis,
>
> Here are some belated comments to your document GlueX-doc-2379. I notice
> I made my comments on the version on Dec 17 and there is an updated
> version. In a  cursory glance I did not the changes, but I could be
> providing redundant suggestions.
>
>
> General: It is good to have the GlueX-doc number at the top. My hard
> copy lacked this and it took me a while to find where I got the document.
>
> Th first table (conditions parameters) contains very similar information
> to what may be included in online data bases. These are still under
> discussion, so the two need to be designed together so that they
> complement each other and minimize duplication. For example, the SiPM
> bias voltages will be recorded and monitored by EPICS and I don't know
> how this information will be needed beyond that.
>
> LED_cntl_clock, LED_cntl_width, LED_bias
> There are a total of 32 independent pulsers (16 upstream, 16
> downstream). Each of 4 pulses are duplicated 12 times and delivered to
> quadrants, up or down.
> I assume the clock is the rate at which the pulser is pulsed.
>
> These next two are clearly DAQ settings and would be part of the
> configuration files
> ADC_threshold. Is this the same as the pedestal subtracted for the
> trigger? Is it a sparcification threshold?
> TDC_threshold. Is this the discriminator threshold?
>
> The next will be recorded and monitored by EPICs
> SPM_bias_hv
> SPM_bias_lv. There are a total of 48 independent channels (24 +5V, 24
> -5V). Each channel serves 4 readout assemblies on one side.
> SPM_crate
> N2_flow. At the moment we do not monitor the N2 flow, they are visual
> flow meters. However, we do monitor temp and humidity inside the readout
> assemblies
>
> This information should be set in a configuration file
> COOLING_temp - I assume this is the set point for the chiller? This
> would be the temperature of the water in the cooling loop
>
> The following information should be monitored and controlled from EPICS
> SPM_temp - There is one RTD temperature probe on each cooling plate,
> i.e. 96.
> Readout assembly temperature and humidity. There are 8 probes (4
> upstream, 4 downstream) in the air (N2) volume of the readout
> assemblies. They measure temperature and humidity, so the dew point can
> be calculated.
> HALLD_humidity - There is only discussion at the moment to add a probe
> to determine the general conditions of the hall.
>
> Table 2 (Calibration constants)
>
> I assume we need an offset and gain for each ADC channel is related to
> the ADC_...._position and width information.   While I agree that there
> may be different methods of determining the gain (cosmics, pulser,
> physics, etc), I think we probably want to store somewhere the "best
> estimate" of the actual gain information for each ADC channel. The
> procedure for determining this value is yet to be determined and may use
> several of the inputs so far mentioned.
>
> ADC_gain 1536 channels
> ADC_offset: 1536 channels
> ADC_time_offset: 1152 channels
>
> For the ADC_physics
> What is meant to the ADC_physics_position or TDC_physics_position, etc
> values? The intent is to use pi0 and eta production to extract the
> information. However, these will not yield a single peak, but rather pi0
> peaks will be determined using combination of two showers (each shower
> with multiple ADC channel). From this a  procedure will have to be
> developed to determine the gains of each channel.
>
> For the TDC_physics_
> What would be used for the source of this information?
>
> Thanks, Elton.
>
>
> --
> Elton Smith
> Jefferson Lab MS 12H5
> 12000 Jefferson Ave STE 16
> Newport News, VA 23606
> (757) 269-7625
> (757) 269-6331 fax
>
>


-- 
Dr. Sean Dobbs
Department of Physics & Astronomy
Northwestern University
phone: 847-467-2826
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/halld-cal/attachments/20140129/0ac6ecf3/attachment.html