[dsg-hallb_rich] [EXTERNAL] Re: More d0 tests of mirror 5C

Wed Oct 20 08:24:06 EDT 2021

Hi Tyler,
I would like to suggest you some more measurements to do on mirror 5C.

1) If you had the chance to polish the fiber, redo the measurement in 
the exact same conditions of the last one with the light intensity 
adjusted, including the exposition time. It would be interesting to know 
if you have more light than before. Only in the case the camera 
saturate, you can readjust the light intensity following the procedure 
and redo the measurement.

2) Check the background subtraction and the shape of the spot far from 
the minimum. If the results are satisfactory, we solved the issue and 
you can move to mirror 5.

3) If not, try to change the CCD. Again, check the background 
subtraction and the shape of the spot far from the minimum.

4) If still there are problems, try to redo the measurement covering the 
edges of the mirror with some black paper.

I hope we can conclude this program by the end of this week.
Next week I would like to discuss the results with CMA to understand if 
there is any real issue with the mirrors or we can move on with the 
production.

Best regards,
Marco

Il 2021-10-12 16:50 Tyler Lemon ha scritto:
> Hi Marco,
> 
>  For all of the analysis I did, I used the file "bkg1.txt" for the
> background subtractions.
> 
>  For photo 5, I read in all files for the background images that we
> took yesterday and then plotted the CCD counts in a histogram where
> the X axis is the CCD count and the Y axis is the number of pixels
> that had that many CCD counts. My main point for that image is that
> when you overlap these histograms on the same plot, they all fall in
> the same area and are the same shape, indicating that the background
> is more or less constant over time.
> 
>  I've moved all of the files to my home directory on JLab Linux
> servers so it should be accessible from jlabl1 or ifarm. The directory
> is "/u/home/tlemon/RICHII/d0". If you can't get to the data there,
> I'll figure out a new place on ifarm or adjust the privileges in that
> directory.
> 
>  In that directory, there's a subdirectory called "data" where I've
> put in the raw data for all tests we've done (I'll also put any future
> data in that area).
> 
>  There's a second subdirectory called "code" where I've put some of
> the code we have used so far. I only have the source code for
> "PlotD0.C" that does the d0-vs-Z plot and "d0-valG.C" that does the
> analysis and outputs the results of the d0 calculation.
> 
>  We do not have the source code for "ximea-shot" that is used to take
> the CCD image and "txt-image-sub" that is used to subtract the
> background from the measurement image. I've asked for the source code
> for ximea-shot a few times but never got a reply or the source code.
> Both executables have outputs that name Luca Barion in case that helps
> you track down the source code.
> 
>  -Tyler
> 
> -------------------------
> 
> From: Marco Mirazita <Marco.Mirazita at lnf.infn.it>
> Sent: Tuesday, October 12, 2021 7:53 AM
> To: Tyler Lemon <tlemon at jlab.org>
> Cc: dsg-hallb_rich at jlab.org <dsg-hallb_rich at jlab.org>
> Subject: Re: [dsg-hallb_rich] [EXTERNAL] Re: More d0 tests of mirror
> 5C
> 
> Hi Tyler,
> I attached plots for the two sets of measurements 01 and 02.
> Both show a problem in the bkg subtraction, the 02 less than the 01.
> This could hopefully be because 02 has a shorter exposition time.
> I suggest to try a measurement with exposition time 0.5 ms, increasing
> 
> the light source intensity if it is necessary and possible.
> Then, few questions
> - which background did you subtract in these two sets of measurements?
> - what are you showing exactly in your photo5?
> - can you post in your web page the code to do the analysis of the
> data
> (subtract the bkg, make the root files, etc)?
> - do you know if it is possible to access the location of your web
> page
> from ifarm or jlabl1? Today it took hours for me to download all the
> data from the web page (and then I had to copy back everything to the
> farm!).
> Marco
> 
> Il 2021-10-12 12:52 Marco Mirazita ha scritto:
>> Hi Tyler,
>> one possible reason why you don't the d0 vs z curve when you covered
>> the edges of the mirror could be because the white foam you used is
>> reflecting the light everywhere. You should use something black.
>> Marco
>> 
>> 
>> 
>> Il 2021-10-11 21:00 Tyler Lemon ha scritto:
>>> Hello,
>>> 
>>>  This morning, Mindy and I repeated d0 measurements with mirror 5C
>>> taking some of the considerations discussed in last week's
>>> collaboration meeting. Below is a (pretty long) summary of what we
>>> did.
>>> 
>>>  The raw data (and also the raw data from previous tests) have been
>>> posted at the link below under "October 11, 2021". The files are
>>> password protected with the username DSG and password RICH2mirrors
>>> 
>>> 
>>> 
> https://www.jlab.org/physics/dsg/technical_documentation/hall_b/rich/mirror_data
> [1]
>>> 
>>> 
>>>  To maybe give an indicator of how some areas of the mirror surface
>>> look, attached is a photo of an area of the surface that looks to
> not
>>> be of very high quality (photo 1).
>>> 
>>>  The first thing we did was to ensure that the test set up is
> properly
>>> aligned. Results for this are all in the subdirectory "01_realign"
>>> This was done by taking the alignment laser pointer and collimators
> to
>>> ensure the beam from the CCD stand is parallel to the z-axis of the
>>> CCD stand (CCD stand set up in attached photo 2, resulting
> alignment
>>> on mirror stand in photo 3). This didn't seem to have any effect on
>>> the measurements and gave the same result as before.
>>> 
>>>  The second thing we did was increasing the intensity of the source
>>> light using the sources trim potentiometer. These results are in
> the
>>> subdirectory "02_source-int-adjust". This also did not seem to have
> a
>>> significant effect on the data.
>>> 
>>>  At this point we tried to cover the edges of the mirror as
> suggested
>>> by Marco C. We did this by cutting a hole in the foam packaging
>>> material the mirror is stored in and placing the foam on to the
> mirror
>>> (photo 4). The results for this situation are in the directory
>>> "03_center-only". The results for this situation were not good at
> all
>>> and we did not get a d0-vs-z fit plot that is in line with previous
>>> measurements. We'll try this again in the future, but I included
> the
>>> data from this test just as an indicator that we tried something.
>>> 
>>>  For all three situations we tried above, we took several
> background
>>> measurements before and after the d0 procedure. In total,
> throughout
>>> the measurement session, we took ~30 backgrounds. I took all of
> these
>>> background files and plotted their data as histograms, which
> resulted
>>> in a plot of ~30 histograms that all overlap in the same general
> area
>>> of the plot (photo 5).
>>> 
>>>  Is there any thing else we should try with the d0 measurements? If
>>> there is a procedure on how we should polish the source's fiber,
>>> please let us know so we can try that.
>>> 
>>>  In the Friday meeting, we talked about the initial prototype of
> the
>>> spherical mirrors and how those d0 measurements looked. Do we still
>>> have that raw data? If so, we can try measuring that mirror again
> and
>>> comparing results to the original measurement to further verify
>>> whether the test station is working as expected.
>>> 
>>>  Thanks,
>>>  -Tyler
>>> _______________________________________________
>>> dsg-hallb_rich mailing list
>>> dsg-hallb_rich at jlab.org
>>> https://mailman.jlab.org/mailman/listinfo/dsg-hallb_rich
>> _______________________________________________
>> dsg-hallb_rich mailing list
>> dsg-hallb_rich at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/dsg-hallb_rich
> 
> 
> Links:
> ------
> [1] 
> https://www.jlab.org/physics/dsg/technical_documentation/hall_b/rich/mirror_data
> _______________________________________________
> dsg-hallb_rich mailing list
> dsg-hallb_rich at jlab.org
> https://mailman.jlab.org/mailman/listinfo/dsg-hallb_rich