[Clas12_rich] Mirror 5C-Autocad3D measurements
Brian Eng
beng at jlab.org
Fri Jul 1 17:43:53 EDT 2016
Doing a least squares fit I get pretty much the same numbers for mirror 5C as the survey group did. All the numbers are in units of mm.
Back surface sphere radius: 2735.65632782 with RMS of 0.0891485148111
Sphere center: 1757.05061408, -1419.56605411, -0.169281068618
Mirror surface sphere radius: 2696.26510045 with RMS of 0.0696200623859
Sphere center: 1746.89907226, -1410.89622076, -0.0705877320044
It should be noted that their calculated center points are off about ~10 in X,Y which might explain getting a difference of ~35 instead of the expected 25.
I've put the code and data file (the same as the one Jim sent) here if anyone wants to mess with it.
https://userweb.jlab.org/~beng/RICH/
It isn't very elegant, but it should be fairly obvious what it is doing.
> On Jul 1, 2016, at 3:48 PM, Tomassini Sandro <sandro.tomassini at lnf.infn.it> wrote:
>
> Hi Pablo
>
> Thanks for your analisys. As you can see the outer radius is smaller than the inner one...
> It should happen the contrary.
> Next time we should distribute better the measured points on the mirror surface.
> Have a nice weekend
> Sandro
>
> Dr Eng. S. Tomassini
>
> Il 01/lug/2016 21:13 Pablo Campero <campero at jlab.org> ha scritto:
>>
>> Hello Sandro,
>> I took your suggestion and I made the calculations of the radius for the inner and outer surfaces, I took more points (21). It looks better now. The results that I got were:
>> -Mirror Surface (Inner): <2694>±81[mm]
>> -Back surface(Outer): <2678>±73[mm]
>> -For both surfaces the St.DEV. is ~3% of the average.
>>
>> Both values cover the theoretical specs. I propose in the future, if is necessary; a program .can be written to take all combinations (32*32*21*21=451584) of the measurements.
>> Have a nice weekend.
>>
>>
>> Best Regards
>> Pablo Campero
>> Detector Support Group.
>
> _______________________________________________
> Clas12_rich mailing list
> Clas12_rich at jlab.org
> https://mailman.jlab.org/mailman/listinfo/clas12_rich
More information about the Clas12_rich
mailing list