Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://curl.sai.msu.ru/mass/download/doc/descript_eso.pdf Äàòà èçìåíåíèÿ: Sun Jan 18 20:44:44 2004 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 20:15:44 2012 Êîäèðîâêà:

Optimized MASS device for synchronous measurements with Paranal DIMM. Optical and mechanical design. Alignment.
Kornilov V., Potanin S., Shatsky N., Shugarov A., Voziakova O. January 15, 2004


Contents
1 Optics design 1.1 Basic principles . . . . . . . . . . . . . . 1.2 Principal geometry of MASS device . . . 1.2.1 Entrance and exit pupils. System 1.2.2 Fabry lens . . . . . . . . . . . . . 1.3 MASS optics . . . . . . . . . . . . . . . 1.3.1 Pupil segmentation unit . . . . . 1.3.2 MASS channels A, B, C, and D 1.3.3 MASS sp ectral resp onse . . . . . 1.4 Field ap erture, centering mechanism and 7 7 8 8 10 11 11 11 14 15 17 17 17 17 20 21 22 22 23 24 24 24 25 25 26 26 26 28 28 29

........ ........ magnification ........ ........ ........ ........ ........ viewer . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

2 Mechanical design 2.1 General description . . . . . . . . . . . . . . . . . . . . . 2.1.1 General characteristics . . . . . . . . . . . . . . . 2.1.2 Device skeleton . . . . . . . . . . . . . . . . . . . 2.1.3 Optical b ench . . . . . . . . . . . . . . . . . . . . 2.1.4 Other assembly units . . . . . . . . . . . . . . . . 2.1.5 Electronics mo dule design . . . . . . . . . . . . . 2.2 Alignment p ossibilities . . . . . . . . . . . . . . . . . . . 2.2.1 Fabry lens and viewer optics . . . . . . . . . . . 2.2.2 MASS channels optics . . . . . . . . . . . . . . . 2.3 Disassembling and assembling . . . . . . . . . . . . . . . 2.3.1 Disassembly sequence for alignment, maintenance 2.3.2 Disassembly of the electronics mo dule . . . . . . 2.3.3 Assembly . . . . . . . . . . . . . . . . . . . . . . 3 Alignments 3.1 Preliminary alignments . . . . . . 3.1.1 MASS PSU alignment . . 3.2 Device alignments at the telescop 3.2.1 Fabry lens p osition . . . . 3.2.2 Viewer alignment . . . . . ... ... e .. ... ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.. .. .. .. .. .. .. .. .. .. or .. .. . . . . . . . . . .

... ... ... ... ... ... ... ... ... ... repair ... ... . . . . . . . . . . . . . . .

. . . . . . . . . . . .. .. . . . . . . . . . .

1


4 Critical parameters determination 4.1 System magnification . . . . . . . . . . . . . . . . . . . . . . . . 4.2 MASS detectors parameters . . . . . . . . . . . . . . . . . . . . 4.2.1 PMT optimal voltage and discrimination determination 4.2.2 Non-linearity and Non-p oissonity determination . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

30 30 31 31 32 34 34 35 36

A Optical parts sp ecifications A.1 The sp ecifications for MASS purchased optical elements . . . . . . . . . . . . . . A.2 The sp ecifications for MASS sp ecial optical elements manufactured by the contractor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B List of mechanical parts

2


List of Figures
1.1 1.2 1.3 1.4 1.5 1.6 2.1 2.2 2.3 2.4 2.5 3.1 3.2 4.1 4.2 Principal optical layout used for calculation . . Optical layout of MASS device in ZY plane . . Optical layout of the MASS device in ZX plane MASS segmentator . . . . . . . . . . . . . . . . Sp ectral resp onse of the MASS device . . . . . Image scans pro duced by centering mechanism View of the device . . . . . . . . . . Main dimensions of the MASS device View of the device without cover . . Optical b ench unit . . . . . . . . . . View of the electronic mo dule . . . . ... with ... ... ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 12 13 13 15 16 18 19 20 21 23 27 28 32 33

.... optical .... .... ....

...... interface. ...... ...... ......

.. . .. .. ..

Photo catho des p osition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . System magnification adjust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Counting functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dep endence of non-Poisson parameter p on flux F . . . . . . . . . . . . . . . . .

3


List of Tables
1.1 1.2 1.3 1.4 The telescop e parameters adopted for MASS optical design. . The MASS main optical parameters . . . . . . . . . . . . . . PSU and entrance segment dimensions . . . . . . . . . . . . . MASS sp ectral resp onse in relative photon units. Wavelengths .. .. .. in ....... ....... ....... nanometers . . . . . . . . 8 10 14 15

4


Intro duction
This do cument describ es the optical and mechanical design of a low-resolution turbulence profiler (MASS) optimized for synchronous observation with Paranal DIMM, according with a Prop osal to ESO [1]. The pro ject was implemented in frame of the ESO contract No. 69255/ODG/02/9124/GWI. The electronics of the device and details related to it are presented in a separate do cument [7]. Also, separate do cuments contain Turbina Software reference guide, Turbina user guide [11], and Sup ervisor user guide [8], which complete the full description of the MASS instrument and its control software. The MASS optical scheme was sp ecially calculated for the use with a short refractive feeding telescop e C102 from Celestron company or similar. The principles of the work of MASS are describ ed in [4]and [2]. Meanwhile, according to the exp erience obtained in a year-long exploitation of the original MASS device [5], some changes have b een intro duced in the geometry of the main optical comp onent of MASS ­ pupil segmentation unit. The Chapter 1 of the do cument presents the final optical parameters of elements together with the tolerances for the critical measures. In addition, the tables give the full sp ecifications for the optical elements, b oth for the standard ones for purchasing in commercial companies and the sp ecial elements manufactured by the contractor. The Chapter 2 describ es the general mechanical design of the instrument. The sequence of assembly/disassembly is presented as well. The next chapter is a guide for alignment of the optical scheme elements ­ the op eration which is mandatory after the device assembly or while installing the device on the telescop e. Exit pupil optics tuning (MASS segmentator), fo cusing and lateral p ositioning of the Fabry lens, checking the entrance pupil p osition are the sub jects of particular attention. Lastly, the Chapter 4 helps to compute the principal parameters of the device resulted from the finished alignment pro cedure. This is critical for correct interpretation of the scintillation data which is p erformed by the MASS software. App endices which follow give technical parameters of the optical and mechanical device comp onents.

5


Bibliography
[1] Kornilov V., The optimization of MASS device for synchronous measurement with Paranal DIMM. A Prop osal to Europ ean Southern Observatory (ESO). Decemb er 11, 2002 [2] Kornilov V., Potanin S., Shatsky N., Voziakova O., Zaitsev A. Multi-Aperture Scintil lation Sensor (MASS). Final design report. February 2002. [3] Kornilov V., Potanin S., Shatsky N., Voziakova O., Shugarov A. Multi-Aperture Scintil lation Sensor (MASS) Upgrade. Final report. January 2003. [4] Kornilov V., Tokovinin A., Voziakova O., Zaitsev A., Shatsky N., Potanin S., Sarazin M. MASS: a monitor of the vertical turbulence distribution. Pro c. SPIE, V. 4839, p. 837-845, 2003 [5] A.Tokovinin, V.Kornilov, N.Shatsky, O.Voziakova, Restoration of turbulence profile from scintil lation indices, MNRAS 2003, V. 343, P. 891 [6] Sarazin M., Ro ddier F., The E.S.O Differential Image Motion Monitor Astron. Astrophys. 227, 294-300 (1990). [7] Kornilov V., Shatsky N., Shugarov A., Voziakova O. Optimized MASS device for synchronous measurements with Paranal DIMM. Electronics and Device control. Novemb er 2003. [8] Kornilov V., Shatsky N., Voziakova O. Supervisor program User Guide. SV version 0.22, January 2004. [9] Tokovinin A. Polychromatic scintil lation. JOSA(A), 2003, V. 20 P. 686-689 [10] Kornilov V., Potanin S., Shatsky N., Voziakova O. Optimized MASS device for synchronous measurements with Paranal DIMM. Optical parameters and general design. July 2003. [11] Kornilov V., Potanin S., Shatsky N., Voziakova O. MASS Software User Guide Version 2.04. Decemb er, 2003.

6


Chapter 1

Optics design
1.1 Basic principles

We shortly remind here the principles of a Multiap erture scintillation sensor (MASS) instrument. MASS measures four scintillation indices in small central circular ap erture and 3 concentric annular ap erture as well as 6 differential scintillation indices for all p ossible pairwise ap erture combinations. Scintillation indices pro duced by a turbulent layer at some altitude h dep end the turbulence intensity, on the ap erture geometry and on the sp ectral range (so called weighting function W (h) [4, 5, 9]). Using these 10 measured values, a calculation of some integral characteristics of the atmospheric turbulence and a restoration of turbulence vertical profile with low-resolution (5 ­ 6 fixed layers) are p ossible. All the weighting functions drop to zero at zero altitude. DIMM (Differential image motion monitor) measures fluctuations of the angular distance b etween two images pro duced by two circular ap ertures ab out 8 ­ 10 cm diameter separated by ab out 20 cm. Theory states that the rms image motion is prop ortional to the turbulence integral over full atmosphere [6]. So, the weighting function for DIMM is un-dep ending on altitude. The idea to use two different turbulence measuring devices synchronously p ointed at the same star using one mount, grew from the following: ­ Multiap erture scintillation sensor (MASS) do esn't sense turbulence lo cated in b oundary layer (first 1 km ab ove ground). Possible solution -- a generalized mo de of MASS measurement, was tested with original MASS device. It was found that this metho d will b e able give a reasonable results with large ap erture feeding telescop e only. On the other hand, DIMM measures b oth low and high turbulence. ­ Numerical simulations show [5] that reducing the largest MASS annular entrance ap erture (the segment D) down to 8.5 cm optimizes the metho d sensitivity for middle altitudes. This means that a non-exp ensive refractive telescop e with diameter 10 cm can b e used to feed the MASS device. Such a refractor can b e installed in parallel of the Paranal DIMM as a piggy-back device. The following general solution was chosen and implemented in optimized MASS instrument for turbulence measurements: ­ in order to enlarge short fo cal length of the refractor, the sp ecial optical interface is included in optical scheme. 7


­ to re-image the plane of entrance pupil of feeding telescop e to the exit pupil plane; ­ to split the light with the help of a segmentator unit (see [2]) onto four MASS channels, and to re-image the exit pupil at photo catho des of MASS detectors. These general ideas follow the principles of the original MASS design. The following Sections describ e this scheme in details.

1.2

Principal geometry of MASS device

The MASS device was designed and calculated for usage with a compact telescop e ­ refractor. For the calculation of the design parameters, the data listed in the Table 1.1 were used. The schematic drawing of the feeding refractor with principal part of the mo dified MASS device is shown in Fig. 1.1 where the measures used in further calculations are shown as well. The high-p ower negative lens is placed b efore the fo cal plane to reduce the b eam convergence and obtain thus the equivalent fo cal length ab out 2 m. This combination of the ob jective PR and the negative lens (transfo cator) TF is called the "telescop e" throughout this do cument. Position x of TF lens from the fo cal plane F P R defines the magnification m at the lens, the new effective fo cal length of the telescop e, and lo cation of the new fo cal plane F (similarly to the role of a secondary mirror in Cassegrain telescop e). In our case, the fo cusing of the instrument is done by change of the distance b etween the ob jective and TF lens. Comparing to Cassegrain telescop e, we have more freedom to parameters optimization, b ecause the fo cal length of the TF can b e varied, to o. Table 1.1: The telescop e parameters adopted for MASS optical design. All dimensions are in millimeters Parameter Nominal fo cal length FP R Diameter of ob jective DP R C102 telescop e 500 102

1.2.1

Entrance and exit pupils. System magnification

To make an image of the entrance pupil, a Fabry lens with some fo cal length F abry must b e placed on optical axis of the instrument. Exit pupil will b e lo cated at a distance af (according to Fig. 1.1) from the lens. Really, the Fabry lens re-builds not original entrance pupil but its image pro duced sequentially by ob jective of a telescop e and transfo cator TF lens. The dimension and lo cation of this image dep end on the exact geometry of the telescop e (recall that telescop e includes TF lens) and change slowly when telescop e is refo cused. Physically, the p osition of the exit pupil is fixed, and the lo cation of the entrance pupil plane is defined by the telescop e system as well as by the Fabry lens fo cal length and p osition. The entrance pupil lo cation is not imp ortant for the results of measurements. Since the elements splitting light b etween channels are placed in the exit plane and work as ap erture stop, the 8


9 Figure 1.1: General optical layout. S -- pupil segmentation unit, F P R -- feeding refractor fo cal plane, F -- refractor+transfo cator fo cal plane, F' -- refractor+transfo cator+Fabry lens fo cal plane, PR -- refractor ob jective, TF -- transfo cator lens, LF -- Fabry lens, S' -- image of S pro duced by Fabry lens, S -- image of S' built by TF lens, E -- entrance pupil of the system, which is the image of S pro duced by refractor ob jective. All values designed by right arrow are p ositive, otherwise -- negative. Other values are explained in text.


working entrance pupils for the MASS devices can b e defined in reverse light path as the images of these pupil stops pro duced by the Fabry lens + telescop e optical system. The ratio of the diameter of an ap erture in the entrance pupil plane to the diameter of resp ective physical element placed in the exit plane, is the magnification K of the instrument. It dep ends on the telescop e and device geometry as well. More exactly: the lengths a, c, Fabry lens fo cal length Fabry and its lo cation da define b oth the magnification K of the instrument and the lo cation of the entrance pupil for a given telescop e with parameters F P R and FT F . For the design, the lo cation of the entrance pupil plane was chosen to b e 50 mm b efore a p ole of a feeding refractive ob jective near the front edge of the telescop e tub e. Such a p osition is very convenient for instrument check and alignment. The nominal value of K = 15.00 was chosen (K 20 in the original MASS). The exact value of system magnification influences final results of the MASS, b ecause the geometry of entrance pupil is used in computation turbulence intensity from directly measured values. Table 1.2: The MASS main optical parameters, transfo cator including. All dimensions are in millimeters Optical parameter Transfo cator lens fo cal length Fabry lens fo cal length Fabry Fabry lens diameter Dabry Distance b etween FP and EP Distance b etween FP and TF LF p osition with resp ect fo cal Instrument magnification K Entrance pupil plane p osition Scale in the fo cal plane scaleF F
TF

Nominal value -18 60 25 123 116.7 -38 16.6 -50 101

planes a lens c plane da b
E

1.2.2

Fabry lens

The optical layout of the MASS instrument (without feeding telescop e) is presented in Fig. 1.2. The co ordinate system which is used here and further, is as follows: ­ Z-axis along optical axis of the instrument ­ Y-axis lies in plane of symmetry of the instrument towards viewer. ­ X-axis is p erp endicular to b oth Z- and Y-axes. In order to diminish overall dimensions of the device, the Fabry lens LF is placed b efore the fo cal plane of the telescop e. In this case the Fabry lens shifts the fo cal plane to a new p osition FP where we have a real image of a target star. In this plane the field ap erture FA is placed. Additionally, such LF placement protects the interior optics of the device from dust. Such design requires re-fo cusing the telescop e each time the Fabry lens is shifted along the axis. The p ossibility to move Fabry lens along optical axis is necessary in order to adjust the 10


system magnification, which can differ from its nominal value due to lens fo cal length tolerance and uncertainties in telescop e geometry. In the Table 1.2 the main optical parameters of the MASS device are listed. Catalog of Edmund Optics company was used to select the Fabry and transfo cator lens. The exact formulae for computations are presented in [10]. The consideration showed that transfo cator and Fabry lenses fo cal lengths are mutually constrained. Note, that short TF fo cal length providing more compact optical interface b etween refractor and MASS, requires shorter LF fo cal length. Optimal pair was chosen as is shown in Table 1.2.

1.3

MASS optics

The optical scheme of the MASS device itself is shown in Fig. 1.2 as ZY plane view, and in Fig. 1.3 as ZX plane view (from viewer side). The main optical element of the MASS device is the pupil segmentation unit (PSU). PSU forms four reflected b eams and reflects them in different directions. The chosen magnification as well as telescop e ob jective diameter, force to adopt the maximal outer diameter of the MASS pupil segmentation unit as 5.5 mm.

1.3.1

Pupil segmentation unit

The PSU is lo cated on the instrument optical axis (see Fig. 1.3). The exact p osition of the exit pupil at PSU is provided by shifting the Fabry lens in Y and X directions. This is done during Fabry lens alignment. To provide the light reflection from the PSU segments in needed directions to the re-imaging mirrors RA, RB, RC, and RD, the segment mirrors are pro duced tilted by 8.0 to the PSU rotation axis. Then, the segments of the PSU are rotated around PSU axis so that the rotation angle b etween the adjacent segments equals 30 . In order to comp ensate partially for the large segments tilt and to place the re-imaging mirrors closer to the instrument optical axis, the PSU as a whole is inclined by 1.65 around X axis. Angles b etween the segment normals and incident/reflected b eam are 6.9 for outer channels A,D and 6.4 for inner channels B,C. Segments have concave surface with a curvature radius of 250 mm that ensures non-divergent b eams after the reflection from segmentator. This p ermits to use small re-imaging mirrors. Despite the segments tilt, their pro jections are circular with high accuracy. Using the optimal set of MASS ap ertures [5] the dimensions of the PSU segments were chosen as listed in Table 1.3. The PSU is fabricated from hard bronze, its mirrors are p olished as a whole with optical quality. The reflecting coating is made by evap oration under vacuum and consists of 3 layers: a Chromium layer dep osited on the bronze, an Aluminum layer, and a protective S iO overcoating. The microphotograph of the PSU is shown in Fig. 1.4. Final diameters of PSU segments, measured with the help of such microphotographies for all pro duced segmentator, agree well with the nominal diameters of Table 1.3.

1.3.2

MASS channels A, B, C, and D

MASS Pupil segmentation unit pro duces four reflected b eams. Each b eam falls on the corresp onding re-imaging spherical mirror RA, RB, RC, and RD. 11


FP LF MV FA RD(RA) CC PMTs SF

ExP

PSU

Z Y
V1

RC(RB)

V2

EP

Figure 1.2: Optical layout of MASS device in ZY plane. LF -- Fabry lens, FP -- instrument fo cal plane, FA -- field ap erture, ExP -- plane of exit pupil, PSU -- pupil segmentation unit, RA, RB, RC, RD -- re-imaging mirrors, SF -- sp ectral filter, PMTs -- MASS detectors, MV -- viewer removable mirror, CC ­ glass plate with central hole, V1, V2 -- transmoving ob jective, EP -- eye-piece.

12


RA LF FP RB FA TK RC SF

PMT A ExP PMT B PMT C PMT D PSU

X
RD

Z
Figure 1.3: Optical layout of the MASS device in ZX plane (corresp onds in Fig 1.2). Viewer is not shown. LF -- Fabry lens, FP -- instrument field ap erture, TK -- triangle knife of centering unit, ExP -- plane of exit segmentation unit, RA, RB, RC, RD -- re-imaging mirrors of the A-, B-, SF -- sp ectral filter, PMT -- four detectors. to the b ottom view fo cal plane, FA -- pupil, PSU -- pupil C-, and D-channels,

Figure 1.4: On the left: Top view of one of the MASS segmentator, illuminated by scattered light. On the right: Segmentator mounted in its holder.

13


Table 1.3: PSU segment dimensions and entrance segment dimensions. All values are in millimeters Segment/ Channel Segment Segment Segment Segment Segment Segment Segment D D C C B B A outer inner outer inner outer inner outer Physical diameter 5.50 3.90 3.85 2.20 2.15 1.30 1.27 Entrance diameter 91.3 64.7 63.9 36.5 35.7 21.6 21.1

MASS re-imaging mirrors are chosen to b e the same as in original MASS, i.e. 12.5 mm diameter and 51 mm fo cal length. The distances from the PSU to mirrors are equal to 120 mm for all mirrors, the distances from mirrors to corresp onding PMTs are 88.7 mm. The angles b etween the mirror normals and incident/reflected b eams are 12.3 for outer channels A,D and 9.7 for inner channels B,C. Usage of simple spherical mirrors under such angles pro duces significant astigmatism. Re-imaging mirrors are tilted slightly to direct light to PMT photo catho des. The outer mirrors are symmetrically tilted by 4.6 around Y-axis and by -2.3 around X-axis. The inner mirrors -- by 2.0 and 2.7 , resp ectively. The resulting tilt of the outer mirrors (normal to Z-axis angle) is 5.12 . The normal pro jection onto the XY plane has a p osition angle ±117 with the axis Y. For the inner mirrors, this angle is 3.31 while their p osition angle is ±36 . Reimaging mirrors are coated by multi-layer dielectric film, reflecting up to 99% in the blue-green region of the sp ectrum. Light falls onto photo catho des under relatively large angles (13 to 15 ), contributing to the distortion of PSU segment images. The effect of star motion into field ap erture was estimated, and no significant energy re-distribution at photo-catho des was found. The size of the segment images on the photo catho des is reduced by 0.73, so the largest image (D segment ) has diameter ab out 4.0 mm. The glass sp ectral filter FS are placed b etween the PSU and field ap erture into central blind. The filter defines the short-wave cutoff of the MASS sp ectral resp onse.

1.3.3

MASS sp ectral resp onse

Compact photomultipliers R7400P from Hamamatsu are used as light detectors. These PMTs have bi-alkali photo catho de of 7 mm diameter. The sp ectral sensitivity is typical of bi-alkali photo catho des. The sp ectral resp onse of MASS is shap ed by the PMT sp ectral sensitivity, the transmittance of the sp ectral filter SF and the reflectance of the re-imaging mirrors. The final sp ectral resp onse is shown in Fig. 1.5 and numerical data are presented in the Table 1.4. Such sp ectral resp onse pro duces a dep endence of MASS magnitude on star color. In Fig. 1.5 (right) the dep endence is

14


S()
1,0

MASS-V 0,6

0,8

0,4

0,6

0,2

0,4

0,0
0,2

-0,2
400 450 500 550 600

, nm

0,0

0,5

1,0

(B-V)

0

Figure 1.5: On left: Sp ectral resp onse of the MASS device. On right: Color equation b etween MASS magnitude and star color index B-V. plotted. Transformation from standard V magnitude is describ ed as follows: M AS S = V + 0.347(B - V )

Table 1.4: MASS sp ectral resp onse in relative photon units. Wavelengths in nanometers 420 430 440 450 460 470 480 490 S () 0.000 0.004 0.206 0.720 0.963 1.000 0.956 0.891 500 510 520 530 540 550 560 570 S () 0.823 0.729 0.636 0.552 0.467 0.391 0.317 0.255 580 590 600 610 620 630 640 650 S () 0.203 0.160 0.061 0.025 0.029 0.005 0.004 0.000

The integral parameters of the MASS sp ectral resp onse are: effective wavelength for A0 star 496 nm, effective sp ectral bandwidth ab out 85 nm.

1.4

Field ap erture, centering mechanism and viewer

A field ap erture is lo cated in the fo cal plane of the instrument. It serves to limit the contribution of sky background to the light measured by MASS detectors. The ap erture also limits the field of view of the centering mechanism: for this reason, a compromise size of the field ap erture as large as 2.2 mm (4 ) was chosen. The ap erture is made as a hole in a flat thin steel plate. 15


Figure 1.6: Image scans pro duced by centering mechanism. On left -- scan of a p oint-like image, on right -- scan of uniformly illuminated ap erture. The size of a wide field of view for star finding is defined by the viewer design and is ab out 9 mm or 16 . To view this field, a moving mirror MV is shifted onto the optical axis of the device. The selected star must b e placed into the central hole (it is seen as red circle when FOV illumination is on) of a glass plate CC, which is co-aligned with the field ap erture to b etter than 0.2 mm (ab out 20 ). In this case, after removing of the mirror, light passes to MASS detectors. Further star centering can b e done by the centering mechanism. Star centering unit is based on the triangle non-transparent knife moving across field ap erture. Edges of the knife are tilted at 45 to direction of the motion along X axis. When the knife is in the center, it fully closes the ap erture. Metho d calibration is based on a following fact: if the field ap erture is uniformly illuminated (by bright sky, for example), the level 0.5 of a maximal signal is achieved when one of the knife edges crosses exactly the ap erture center (see Fig. 1.6). There are two such p oints, obviously. Comparing a star scan slop es p osition with these reference p oints, the star image shift can b e obtained. The viewer is not used in a normal work since the parallel DIMM provides the telescop e p ointing at the star, star searching if necessary and guiding during measurements. In practice, the viewer serves as an auxiliary to ol for coalignment of DIMM and MASS feeding optics and in extraordinary cases. The optical layout of the viewer is presented in Fig. 1.2 where the main viewer parts are shown. Removable mirror MV has dimensions 12 â 18 mm. When inserted in the b eam, it is placed at 40 angle with resp ect to the optical axis of the device. This provides the viewer axis tilt equal to 80 with resp ect to the instrument optical axis. The re-imaging system of the viewer consists of two achromatic lenses V1 and V2 ACH18x50 (Edmund optics; fo cal length 50 mm, diameter 18 mm) and repro duces the instrument fo cal plane with magnification -1. The lenses are separated by 75 mm distance. Standard 1 1 inches eye-piece with fo cal length 12 ­ 15 mm is used with the viewer. The 4 eye-piece is lo cated at 220 mm from the axis of the device (or telescop e) to provide easy access for the observer. The viewer is placed in plane ZY.

16


Chapter 2

Mechanical design
2.1
2.1.1

General description
General characteristics

As it was shown ab ove, the optimized MASS device is installed at feeding telescop e­refractor using mechanical and optical interface. The interface can b e considered as the part of the device, to o. The general view of the MASS device for synchronous atmospheric turbulence measurement is presented on Fig. 2.1. In the Fig. 2.2 overall device dimensions are shown. Note, that the main b ox of the device is practically the same to one designed for CTIO/TMT pro jects. Practically all parts of the device are fabricated from hard aluminum alloy, black-ano dized. Only few critical parts are made from steel. Total weight of the MASS instrument is ab out 1.5 Kg. The device is mounted at refracting telescop e Celestron C102 with the help of a inner metric thread M 56 â 1 6G which is pro duced in the interface tub e. The interface tub e with device is tighten up the corresp onding thread on the fo cuser tub e of the refractor and fixed by lo cking nut. The device electronics is enclosed in a separate case which can b e removed and attached again easily. The connection to interior electronic elements, such as illumination LED, is provided via a sp ecial plug connector. The mechanical parts are designated on the drawings as "MDnnS", where MD ­ prefix for MASS device, 'nn' ­ assembly unit numb er, 'S' -- suffix for the sp ecific part. The prefix is omitted when a designation is mentioned b elow.

2.1.2

Device skeleton

The force structure of the device consists of 3 elements: device base 01A, U-profile main b eam 01C and b ottom tie 01D. These parts are screwed together and form rigid through-like frame. One can see it in the Fig. 2.3, where photo of the device without cover 01B is presented. This structure b ears all other units and assemblies. Do not disassemble the device skeleton if there are other solutions! Inside the device base, the mechanism for lateral shift of Fabry lens unit is mounted. On the outside, a mount ring (flange) 02C is screwed. The latter holds the Fabry lens unit and tub e 09A of the optical interface. The transversal b eam 01E (called b elow ­ optical b ench) is attached to the main b eam.

17


Figure 2.1: MASS device attached to refractor, which is installed atop of the La Silla DIMM.

18


250

62

95

80

Figure 2.2: Main dimensions of the MASS device with optical interface.

19

335


Figure 2.3: View of the device without cover. Used co ordinate system is shown, to o. The optical b ench b ears most parts of the device optics. On the outside of the main b eam, the switching knob of the viewer mirror, 05G, is placed. The optical plate 04A with PSU is mounted on the b ottom tie and covered by 01F. Also, the electronic b ox is set on the tie and fixed to the tie. The cover 01B (the second half of the device b ox) is fastened to the base and to the tie.

2.1.3

Optical b ench

The optical b ench 01E is a central assembly unit of the device. On the top plane of the b ench four functional units are mounted: ­ ­ ­ ­ unit of the re-imaging mirrors RA, RB, RC, and RD; viewer removable mirror unit; fo cal plane unit; star centering mechanism.

On the b ottom planes the central blind are fastened. The top and b ottom views of the optical b ench are shown in the Fig. 2.4. On this photo the central blind is removed. The re-imaging unit consists of the mirror supp ort 07A with so ckets for mirrors, where the mirrors lie free, and the cover plate 07B, which fixes the mirrors. The supp ort is fastened to the optical b ench with help of four M2.5 screws. The elastic separator b etween the part and the b ench p ermits to adjust a little the total tilt of the mirrors holder.

20


Figure 2.4: The top (left) and b ottom (right) views of the optical b ench. Star centering unit is removed. The viewer removable mirror is a more complex unit. It contains the supp ort 05A, the clamping cramp 05C which limits the mirror rotation and b ears the Hall sensor plate, the mirror holder 05B with two half-axes 05E and 05U, and the switching -like spring. The mirror MV is cemented to its holder. Also, the cover plate 05D which holds the glass plate CC with central hole is screwed to the supp ort. Illuminating FOV LEDs are mounted at the cover plate. The plate CC is glued to the holder. The fo cal plane unit includes the field ap erture 06A pressed into the so cket of ap erture supp ort 06B. A blind 06C is utilized to prevent direct light passing from the field ap erture to the PMT photo catho des. It also reduces the scattered light from the exit pupil elements. All electronic parts placed in the main case of the device are lo cated on the optical b ench. The b ench b ears a connector for this electronics that matches the connector in the electronic b ox. More information ab out electronic elements which are inside the main b ox can b e found in the Do cument [7]. In addition to the ab ove-mentioned electronics, a Control light LED PCB is mounted to the b ench directly. Also, a controller of a stepp er motor of the centering unit E3 is fastened to the b ench (see Fig. 2.4).

2.1.4

Other assembly units

Fabry lens unit The Fabry lens unit includes the shifted square nut 02B with a thread for the Fabry lens holder 02A. The thread serves to fo cus the lens. The lens itself is installed in the holder using a thin lo cking nut 02C. The square nut is clamp ed b etween the device base and the mount ring. The clamp pressure is regulated by a wide brass washer which can b e either corrugated or planished. The lateral shift of the square nut with LF in one direction is provided by a cam 02F with finger 02G which transmits the motion of an adjusting screw to nut motion. For another 21


direction, the second cam is installed. This mechanism is mounted under the device base, the fingers pass through the slots outside. Viewer The viewer consists of three parts: the eye-piece so cket 03A, the viewer tub e 03b. and the viewer flange 03C. The latter is p ermanently screwed to the b ox cover. The re-imaging lenses V1 and V2 are installed in the so ckets of the eye-piece part and of the viewer flange with the help of lo cking nuts 03D. Optical plate The optical plate 04A b ears the PSU holder 04B, which is fastened by 3 screws -- 2 pulled and 1 pushed. From one end the holder is pressed against a low central pad (height 0.6 mm), which is placed in the center of the optical plate. The pupil segmentation unit is set in its holder with the help of the cover plate 04C. The optical plate is mounted in the so cket of the b ottom tie of the device b ox. It may b e removed and fixed back for checking or cleaning. The sp ecial cover protects the PSU and its fixing and alignment screws.

2.1.5

Electronics mo dule design

The electronics mo dule (see Fig. 2.5) consists of two parts: the PMTs housing and the electronics case. The parts are screwed together and are not detachable from each other. The PMTs housing 08A contains 4 PMTs, 3 PCBs of the photon counting electronics, the teflon spacer 08C, which prevents PMT photo catho des from contact with housing, and the shutter mechanism. The shutter mechanism consists of two blades 08D with holes, the cramp 08E, the lever 08F, and the axis 08G. The axis passes through the hole in the housing, its rotation closes (clo ckwise) or op ens (counterclo ckwise) the shutter. The shutter do es not provide full darkness when the electronics is detached from the device, but protects the PMTs from direct daylight. The housing must b e always closed with the cover 08B when p owered. The PCB which b ears auxiliary electronics, two external connectors, and the connector to the main-case electronics, is mounted on the frame 08H of the electronics case. The cover 08I protects the electronics from the outside. In the cover, a window for the LED indicators is made. When electronics is p owered, the green LED shines. A presence of HV is indicated by the red LED and data exchange ­ by the yellow LED.

2.2

Alignment p ossibilities

Some alignment features are provided here. Most of them are intended for assembly pro cess only. Other alignments are done when the device is attached to the feeding telescop e. The alignments are: ­ fo cusing of the Fabry lens; ­ lateral shifts of the Fabry lens; ­ tilt of the viewer mirror; 22


Figure 2.5: View of the detached electronics mo dule. A photomultiplier PMT R7400 is shown separately. A p en on the photo is placed for comparison. ­ ­ ­ ­ centering of the CC plate; rotation a PSU segments around their axis; tilts of the PSU in XZ and YZ planes; tilts of the MASS re-imaging mirror assembly in two directions;

2.2.1

Fabry lens and viewer optics

The fo cusing of the Fabry lens is done by rotating the LF holder in the thread of the square nut 02B. The error of 0.5 mm in the Fabry lens p osition pro duces the magnification error less than 0.5% and the shift of the entrance pupil plane along optical axis ab out ±100 mm. So, an accuracy of the Fabry lens fo cusing ab out 0.5 mm is more than sufficient. The full range of fo cusing of ±4 mm around the nominal p osition is provided. The nominal Fabry lens p osition dep ends on the particular feeding telescop e. The lateral shifts of the LF with an accuracy of ab out 0.1 mm (which corresp onds to 1.5 mm in the entrance pupil) and with the full range of ±2 mm (±30 mm in the entrance pupil) in b oth directions provide practically identical shifts of the exit pupil in the PSU plane. This alignment is aimed to comp ensate for the imp erfections in the Fabry lens centering and in the device centering and tilt. It is pro duced by two screws accessible via 3 mm holes in b ox cover 01B with help of Hex1.5 key. The inclination of the viewer mirror is fixed during the device assembly and should provide b eam axis parallel to the viewer mechanical axis. The tilts are regulated with help of set screw 23


in the viewer supp ort cramp 05C. The residual offset of the on-axis star image from the viewer center may b e eliminated by shifting manually the glass plate CC, up to ±1 mm in b oth directions.

2.2.2

MASS channels optics

After the MASS segmentator is fixed in its place, the p osition angles of its segments should b e tuned to their correct values of ±15 and ±45 ; the segmentator is inclined as a whole as well. The aim of these alignments is to direct the reflected b eams precisely into the centers of the resp ective MASS re-imaging mirrors. The p osition angles must b e set to within ±1 and the segmentator inclination is tuned with a precision of ±0.1 in b oth directions to place the b eam sp ots in the re-imaging mirrors with sufficient accuracy. These alignments are provided by the push-and-fix screw pairs having the full range of ab out ±1.5 (this is enough, given the roughly correct initial segmentator setting under these angles). Finally, to center the PSU images on the PMT photo catho des, the re-imaging mirror assembly is aligned with an accuracy not worse than ±0.2 (corresp onds to the centering errors of ab out 1 mm on the PMTs). Given that the mirror supp orts are already made with correct angles, the alignment range of ±1 is sufficient.

2.3
2.3.1

Disassembling and assembling
Disassembly sequence for alignment, maintenance or repair

Do not forget to close the PMTs shutter b efore disassembly of the device! Disassemble the parts only to the state needed for the device maintenance or optics alignment or cleaning. Some parts of the device can b e removed without device op ening, in arbitrary order: ­ The Fabry lens can b e removed with its holder only. Before, mark the p osition of the holder inside the square nut to re-establish the fo cusing at assembly. ­ The viewer can b e detached if the instrument is aligned and further work is planned in automatic mo de. For this, unscrew the viewer tub e with eye-piece together from viewer flange. Protect the first viewer lens V1 by any plastic cup. ­ To check or clean a V2 lens, unscrew the viewer tub e from an eye-piece so cket. ­ The electronics mo dule can b e detached to do some checks or alignments. Turn off the device, b e sure that PMTs shutter is closed. Remove 4 M3 screws (2 near the viewer and 2 from the b ottom tie) completely, then pull the electronics mo dule away from device b ox, to unplug it from the internal connector. ­ Optical plate with MASS PSU can b e removed to check the optics or the p osition of the exit pupil. First, slacken 2 screws M2.5 and remove the optical plate cover. When unscrewing 3 M2.5 screws completely, supp ort the plate by hand. To provide access to the optics inside of the main device b ox, the cover 01B must b e removed. To do this: detach the electronics mo dule first, unscrew 4 M3 screws -- 2 which fasten the cover to the device base (near the viewer) and 2 which fasten the cover to the b ottom tie. With some effort remove cover away in the Y-direction.

24


Then, the optical b ench where most of the optics is installed, can b e removed from the main b eam. To do this, from inside of the main b eam, unscrew completely 4 screws which are fastened the b ench to mainb eam. Flip the mirror in the viewer-on p osition to detach the mirror semi-axis from the gro ove in the mirror knob. Pull gently the optical b ench out in the Y-direction. Further disassembly is not recommended. If it is really needed, consult the designers for additional recommendations.

2.3.2

Disassembly of the electronics mo dule

Disassembling the electronics mo dule includes several steps, which must b e done sequentially. To remove the PCB of p ower and auxiliary electronics, one must: ­ ­ ­ ­ unscrew 3 M2.5 screws which fasten the electronics cover 08I and remove this cover; unscrew completely 2 M2.5 screws from the plate 08J of DB9 line connector; unscrew 3 M2 screws that fasten the PCB itself; if it is necessary to remove the PCB completely, unsolder the HV yellow cable and disconnect the blue cable.

To change the PMTs or repair the counting electronics, do the following: ­ unscrew 4 M2 screws from the PMT housing cover 08B and remove this cover; ­ unscrew completely 2 M2 screws from the counters PCB (connector side) and unscrew from the base the long M2 screw with teflon tub e; ­ disconnect this PCB and turn it by 180 ; ­ unscrew 3 M2 and 2 M1.6 screws from the amplifiers PCB; ­ with the help of a thin screwdriver (< 1.5 mm), b egin to unscrew 2 M1.6 screws through the holes that are nearly opp osite to the connectors edge of the PCB, simultaneously pulling up the PCB itself; ­ when these screws are detached from the PMT housing, fold the PCB very carefully, pulling the PMTs out of the housing.

2.3.3

Assembly

Assembly is done in reversed order. A few recommendations may b e useful for this pro cess. ­ When mounting the optical b ench back to the main b eam, pay attention to the p osition of the gro ove in the axis of the viewer knob. The mirror half-axis must ho ok into this gro ove. Be sure that the b ench lies correctly in the b eam b efore tightening finally the 4 screws. ­ When installing the cover b ox back to the device, do not damage the rubb er cord which is glued in the gro oves of the b ottom tie and the device base. Also, check that the connector on the optical b ench is correctly inserted in the corresp onding hole of the cover. Be sure not to leave a slot b etween the upp er edge of the cover and the device base. ­ When fastening the PCB, b e sure that the PCB is laid correctly and tightly. ­ When attaching the electronics mo dule, b e careful to insert the connector pins correctly into the matching connector on the optical b ench. ­ When installing the holder with Fabry lens, do not reverse it. The more convex surface of the Fabry lens must face the telescop e. 25


Chapter 3

Alignments
3.1 Preliminary alignments

Preliminary MASS optics alignments are p erformed during device assembly. These alignments include a correct placement and tilt of the optical elements to provide light pass through MASS channels. Alignment p ossibilities were describ ed ab ove in Sec. 2.2. To align the optics, one will need to prepare some additional to ols: a kind of the optical test b ench, the laser light source, and a telescop e mo del (see b elow). The optics test b ench may b e arbitrary but providing enough rigidity and the source-toMASS distance of the order of 0.5 ­ 1 meter. The attachment of the device to the b ench must provide the p ossibility to adjust the p osition of the light source (laser b eam) with resp ect to the device in two directions. The semiconductor laser of no more than 3 mW p ower is set on the opp osite end of the b ench. The variable resistor of a few KOhm is recommended to b e connected sequentially with the laser to adjust laser b eam intensity. The laser supp ort must also allow the slight corrections by angle. In addition, one needs the weak negative lens to attach to the laser to make the slightly divergent b eam. It is needed to illuminate homogeneously the entrance pupil of the mo del telescop e. The latter is attached to the device instead of the Fabry lens holder and consists of the go o d-quality ob jective lens (fo cal length ab out 50 mm) and the pupil diaphragm of the size ab out 5 mm set in front of the ob jective at the distance equal to the lens fo cal distance. The telescop e fo cusing should b e p ossible.

3.1.1

MASS PSU alignment

Before installing the PSU, rotate its segments around rotation axis according to Sec. 1.3.1. Namely, if seen from the PSU base side, the rotation angle is defined by gro ove in the base of each segment. Note that the D­segment base is closest to the PSU handler and the A­segment base is upp ermost. The segments must b e rotated with resp ect to the YZ­plane by following angles: D -- at 45 and C -- at 15 counterclo ckwise; B -- at 15 and A -- at 45 clo ckwise. Do not tighten the PSU cover plate. Switch on the laser and direct its b eam into the field ap erture. Provide that laser b eam falls on the PSU installed. If the laser b eam is wide enough (but no lens is installed in front of the

26


laser), all four segments of the PSU will b e illuminated. Otherwise, firstly p oint the b eam on the largest D­segment. One can see the reflected b eam sp ots near the re-imaging mirrors. Correct the segments orientation will direct the b eams onto the mirror centers. With help of a thin screwdriver or awl, rotate the first D­segment in such a p osition that the reflected b eam falls closest to the center of D­mirror. After this, try to align the segment C similarly, fixing the D­segment orientation with help of another screwdriver. Then pro ceed with B and A in the same fashion. It is evident that, having the widest b eam, the D­segment is most critical in alignment. After finishing the tuning of rotation angles of segments, tighten the PSU cover plate. In principle, it is p ossible to use a sp ecial mask with the marked mirror centers put atop of the mirrors, but normally the laser b eam sp ots are sharply seen at mirror surfaces. After doing these rotating alignments, try to set the b eams closer to the mirror centers using tilts of the PSU holder with help of 3 screws which fasten it to optical plate. Normally, a combination of the prop er tilt of the PSU supp ort and appropriate rotation angles of the segments provide the reflected b eams falling close enough to the re-imaging mirror centers that no light is lost somewhere in the further path due to vignetting. There is no individual alignment for each re-imaging mirror. The supp ort of those mirrors as a whole can b e tilted slightly in two directions. This p ermits to align a little the p osition of segment images built by re-imaging mirrors RA, RB, RC, and RD on the PMT photo catho des. To check the correct p osition of the images, a sp ecial mask can b e used. The exact p ositions of the PMT photo catho des with resp ect to electronics mo dule reference plane are shown in Fig. 3.1. The manufacturing accuracy is sufficient to provide right image centering on the PMT photo catho des.

01B 01D

6.5

9.5
Photocathode

Figure 3.1: Photo catho des p osition with resp ect to the reference plate of the b ottom tie.

Pay a sp ecial attention to D­segment image, b ecause it is the largest one. To check the pupil images as they will b e on photo catho des, attach the mo del telescop e to the device. With help of a negative lens, pro duce the divergent laser b eam, fo cus the mo del telescop e. In the dark ro om, the images of segments are seen on the pap er mask placed in the plane of PMTs photo catho des. Also, the images can b e observed directly using a magnifying lens when PSU is illuminated by any scattered light. The describ ed alignments are normally done in the lab oratory once after the device optics assembly. The rest alignments related to the installation of the device on the feeding optics are describ ed b elow.

27


3.2
3.2.1

Device alignments at the telescop e
Fabry lens p osition

First alignments after the device attachment to telescop e are convenient to do with the transparent plastic mask, where the size and p osition of the exit pupil are drawn on the front side. The mask is mounted definitively instead of the optical plate 04A. The mask p ermits to do lateral alignments of the Fabry lens and fo cusing the lens. Also, electronics mo dule must b e detached. Illuminate well the entrance ap erture of the telescop e or p oint telescop e at the bright ob ject such as a white wall. The image of the entrance ap erture of the telescop e pupil can b e viewed directly on the mask or with help of a magnifying lens. Put this image in the center of the marked circle, rotating the alignment screws of LF lateral shift.

K
17.6 17.4 17.2 17.0 16.8 16.6 16.4 16.2 16.0 34 36 38

b

E

K
17.2 17.0 16.8 16.6 16.4 16.2 16.0 15.8 15.6 c

b

E

3000 2000 1000 0 -1000 -2000 -3000

2000 1000 0 -1000 -2000 -3000 -4000 113 115 117 119

40 -da'

Figure 3.2: System magnification K (solid line) and entrance pupil plane p osition b E (dashed line) dep endence on the Fabry lens fo cusing (left) and transfo cator lens p osition c (right). -da is a distance b etween the device fo cal plane and LF. c is a distance b etween the device fo cal plane and TF. Black line -- for nominal fo cal lengths (60 mm and -18 mm), blue -- at 1.5% less, red -- at 1.5% greater. Place some flat opaque ob ject (mask) with a sharp edge (e.g. a pap er strip e) into the plane of the entrance pupil (top end of the telescop e tub e). Observe the image of this mask in the plane of exit pupil. If the Fabry lens is fo cused well, the pupil image with a mask shadow will b e seen sharply. Otherwise, remove the side cover 01B and optical b ench 01E. After this, it is p ossible to rotate Fabry lens holder in needed direction. The LF shift can b e estimated with help of the Fig. 3.2. For this, move the mask away from the telescop e top at ab out 0.5 m. If the mask image sharpness will improve, then LF is lo cated to o far from the fo cal plane of the instrument. Rotate LF holder counterclo ckwise at 2 revolutions. Rep eat the pro cedure again to reach the correct LF fo cusing. 28


Then check the correct p osition of the exit pupil again. When Fabry lens is fo cused and laterally aligned, remove the plastic mask, install the optical plate and the optical b ench. To check MASS channels, lo ok at PSU through PMT holes with help of a lens. Segment images must b e uniformly illuminated without any vignetting.

3.2.2

Viewer alignment

When a star image is lo cated in the center of ap erture (it can b e checked again by a partial illumination of the telescop e entrance) the viewer can b e aligned, to o. Lo ok in the viewer. To see an illuminated central hole, attach the electronics mo dule. If the star image is offset from the center of the illuminated circle, one needs to detach the side cover and lo osen the CC glass holder fixing screws. Using a magnifying lens for controlling the star image in the glass hole, move the holder until star drops in the center of the hole. Very precise alignment is not needed here. When done, tighten the screws to fix the glass holder to the viewer mirror supp ort and mount the side cover 01B at the device (see Sect. 2.3.3). Lo oking in the viewer, check fo cus. Viewer fo cusing is made by eye-piece shifting in o cular tub e.

29


Chapter 4

Critical parameters determination
4.1 System magnification

As it follows from Sect. 1.2.1, a system magnification dep ends on parameters of the telescop e optics as well as parameters of the MASS optics. The system magnification transforms the physical dimensions of the pupil segmentation elements into the sizes of the instrument annular entrance ap ertures which are included in the theoretical formulae. Therefore, exact system magnification is needed for MASS correct work. The measurement of the system magnification has to b e p erformed in a dark ro om. All the alignments and a real telescop e fo cusing must b e done b efore. Remove the electronics mo dule. Put some strong light source in front of the channel D hole in the b ottom tie 01D (the hole closest to the connectors side) to get the segmentator fully illuminated. The size of the source must b e not less than 4 mm, which is the D-segment image size on the photo catho de created by the re-imaging mirror. The lo cation of this source has to coincide with the D-segment image place; light b eam from the source must b e directed in the Channel D re-imaging mirror. If the source is not large enough, displace it from the p osition of the D-channel photo catho de. Note, that angle b etween device axis and direction to the re-imaging mirror is ab out 15 in the YZ plane. The segment D image is built in the entrance pupil plane of the MASS + telescop e system. It has a blue-green color due to the selective reflection by re-imaging mirror. The edges of the segment D image are easily examined with a magnifying lens. Put a transparent ruler or other precise measurement to ol in the plane of the entrance pupil. Use a magnifying lens to see simultaneously the D-segment image edges and the ruler clearly. Measure the outer and inner diameter of the image. Alternatively, some semi-transparent screen (pap er) may b e placed in the plane of the entrance pupil and, if the light source is bright enough and well collimated (like LED flash), one can see directly the D-segment image. Mark the edges of the image on the screen. Then, measure the picture by any ruler. While examining the edges of the D­segment image, make sure that there is no vignetting in the system (edges are equally sharp, vertical size is equal to the horizontal size). Similarly, rep eat the same pro cedure putting the light source in channel C. This helps to control the magnification obtained by measuring D­segment. Note, nevertheless, that the less image size which is measured, the less precision of the magnification is obtained.

30


The magnification of the system is obtained by division of the measured sizes by the corresp onding physical diameters of the segment D and C of the segmentator (see Table 1.3). If the image is well-fo cused, the precision of image size measurement of the order of 0.5 mm is easily achievable and is more than enough for our purp ose. Check the matching of these estimations, compute the mean magnification value and put it in the device.cfg file.

4.2
4.2.1

MASS detectors parameters
PMT optimal voltage and discrimination determination

In order to cho ose the working p oint (optimal HV level common for all PMTs, and individual discrimination levels), one needs to conduct the counting characteristics registration. Counting characteristics are recorded using the Detector Counting measurement function of the Turbina program (Menu Tools) (see [11]). Since the fluxes from the control light differ much in different channels, at least two levels of the control light are recommended to set in the sequence, to have the curves with the plateau fluxes from 300 to 1000 pulse/ms. With lower signal level, the precision of the non-Poisson parameter is degraded, with bright light, the strong non-linearity is already encountered. An additional control light level equal to zero (0) must b e set in a sequence to get the dark current characteristics. The grid of high voltage levels covers normally the range 550 to 950 V with a 50 V step. While fine-tuning the settings subsequently, the step and range may b e lowered. The discrimination threshold level is tuned within a range from 0.3 to 0.9 mV with a step 0.1 mV. These input parameters for the measurement are set in turbina.cfg file in the Section Operations Subsection Detectors counting measurement. The accumulation time of each p oint should b e long enough for the reliable estimate of non-p oissonity. The estimate of the precision of its determinations is:
2 p

=

1 2 (1 + ), N F

(1)

where N is a total numb er of micro-exp osures, F is a mean count p er micro-exp osure. In practice, to achieve the relative precision of p ab out 0.5% one needs the accumulation time more than 100 s at high fluxes. This implies ab out 2­4 hours pro cess for the total cycle of measurements. Since the drifts and temp erature dep endences are p ossible, the rep etitive measurements for checking the working p oint stability are necessary. These measurements can b e done with a narrower range of input parameters to economize time. The dark current characteristics are aimed to determine the range of the HV level and discrimination thresholds where the dyno de or pulse amplifier noise is negligible. For making the light characteristics, one needs to measure the relations of b oth flux and non-Poisson parameter p on the HV level U . An example of such relations is given in Fig. 4.1. From this figure it follows that the high voltage must b e not lower than 800 V. The counting characteristics b ecome flat enough, fluxes dep end weakly on the discrimination threshold and the non-Poisson parameter approaches the value ab out unity only ab ove this value. Note, nevertheless, that for the threshold of 0.9 mV the HV has to b e not less than 900 V to provide a low slop e of the HV dep endence. On the other hand, the HV of 800 V is quite enough for the threshold of 0.5 mV. 31


Channel B, 11/11/03, t=+10
N, ps/ms 2000 p 1,8 1,6 1,4 1000 1,2 500 1,0 0,8

1500

0

600

700

800

900 U, V

600

700

800

900 U, V

Figure 4.1: Light counting functions. Left: Flux dep endence on the high voltage for 7 threshold levels (0.3, 0.4, 0.5, 0.6, 0.7, 0.8 and 0.9 mV), lowest curve corresp onds to the 0.9 mV level. Right: Non-Poisson parameter as a function of Voltage for 7 threshold levels, here lowest curve corresp onds to the 0.3 mV threshold. It is b etter to use HV as low as p ossible. Since the HV value is common for all the PMTs, joint analysis must b e done. Doing this, keep in mind that non-Poisson parameter is most critical for PMT in the channel A. Note, that the upp er limit for PMT R7400 is 950 V and this value should not b e selected for a long term usage. An additional constraint is the over-light protection. Note here, that since the relation of an average ano de current on the high voltage supply is quite steep, the safety limit of the over-light system (counted in pulses p er second) decreases significantly when the HV level grows.

4.2.2

Non-linearity and Non-p oissonity determination

In order to treat correctly the photon statistics and compute the correct scintillation indices, one has to know the non-linearity parameter and the non-Poisson parameter p. Correct value of the parameter is critical at high fluxes (in C and D-channels), while an exact value of the parameter p is needed at low fluxes (in A and B-channels). Both these parameters are derived from the dep endence of p on the light flux F which needs to b e sp ecially obtained. To get the p­F relation (see previous section), one can use the sp ecial function Detector Statistics measurement in the Turbina program, placed in menu item Tools. The measurements of flux F and non-Poisson p values are made with currently set values of the discrimination thresholds of counters and high voltage level. The grid of the control light relative intensities is supplied dense enough to get the needed precision of the output parameters. Some fifty values from 0.0 to 1.0 with a step 0.02 are recommended. The duration of one p oint 32


1,1
Channel A

1,1
Channel B

1,0 0,9 0,8 0,7 0

1,0 0,9 0,8 0,7 0

2000

4000

6000

2000

4000

6000

Figure 4.2: Dep endence of non-Poisson parameter p on flux F for A and B channels. Line is linearly fitting the measured p oints. measurement is determined by the formula (1) and may b e of the order of 40 sec or more. These input parameters for the measurement are set in turbina.cfg file in the Section Operations Subsection Detectors statistics measurement. The typical relation of the non-p oissonity p on the average flux in channels A and B is shown in Fig. 4.2. It is clear that this relation is practically linear. It should b e noticed meanwhile that the b etter fit is obtained with a quadratic approximation of the relation. Use the least-square metho d to get the linear regression co efficients (the handy graph-plotting program xmgrace provides such a p ossibility as many others). The crossing p oint of a line fit with the p-axis (constant term in regression) determines the parameter p. The line slop e in the p oint of zero flux is equal to -3 where the non-linearity is expressed in milliseconds if the flux F is counted in pulses p er milliseconds. The slop e of fitting line in the Fig. 4.2 corresp onds to a non-linearity parameter ab out 12 ns.

33


Appendix A

Optical parts specifications
A.1
Des. TF

The sp ecifications for MASS purchased optical elements
Part and parameters Transfo cator lens Singlet LENS Fo cal length: -18 mm Diameter: 9.0+0.0 mm -0.1 Fabry lens Fo cal length: 125 mm +0 Diameter: 25.0-0.0 mm .2 Viewer lenses Fo cal length: 50 mm Diameter: 18 mm Kellner eyepiece Fo cal length: 12 mm 1 Barrel diameter: 1 4 inches Manufacturer Edmund Optics Sto ck name/numb er PCV9x-18MgF2 TS NT45-381 ACH25x125MgF2 TS NT32-492 ACH18x50MgF2 TS NT32-913 -- Total q-ty 1 Rem. 1

LF

Edmund Optics

1

1

V1,2

Edmund Optics

2

1

K

Any

1

1. See sp ecification at www.edmundoptic.com

34


A.2

The sp ecifications for MASS sp ecial optical elements manufactured by the contractor
Part and parameters Removable mirror Size: 12 â 18 mm Substrate: BK7 glass Thickness: 2 mm Surface Accuracy: /4 MASS mirrors Diameter: 12.8 mm Curvature radius: 102 mm Substrate: BK7 glass Thickness: 3 mm Surface Accuracy: /4 Segmentator Diameters: see Tab. 1.3 Curvature radius: 250 mm Substrate Material: Hard bronze Surface Accuracy: /4 Circle reticle Central hole 2.3 mm Thickness: 1.0 mm Diameter: 13.0+0.0 mm -0.2 Sp ectral filter Diameter: 11 mm Surface Accuracy: /2 OP6 Part numb er OP4 Ref. op4.dwg Total q-ty 1 Rem. 1

Des. MV

2 OP1 op1.dwg 4 1

RA-D

2 md10a.dwg md10b.dwg md10c.dwg md10d.dwg op6.dwg 1 1

PSU

2 1

CC

SF

OP5

op5.dwg

1 2

1. Coating: Protected aluminum, R avg. > 87% 2. Surface Quality: 40 ­ 60 scratch and dig over central 95% of surface

35


Appendix B

List of mechanical parts
Table contains the list of mechanical parts which are needed for production of the MASS device. The parts are grouped in assembly units. Remarks "S", "R", "C" are the assigned ranking estimations of the part work-consuming -- simple, rotation-symmetry and complex. Needed fasteners and standard items are not included in this table. Des. AS01 MD01A MD01B MD01C MD01D MD01E MD01F MD01G AS02 MD02A MD02B MD02C MD02D MD02E MD02F MD02G AS03 MD03A MD03B MD03C MD03D AS04 MD04A MD04B MD04C Part Box Device base Box cover Main beam Bottom tie Optics bench PSU cover Connector support Fabry lens unit Fabry lens holder Shifted square nut Locking nut No.1 Mount ring Cam axis Cam Cam finger Viewer Ocular socket Viewer tube Viewer flange Locking nut No.2 Pupil segmentation unit PSU support MASS segmentator holder Cover plate HA HA Steel 1 1 1 C C S HA HA HA HA 1 1 1 2 R R R R HA HA HA HA Steel HA Steel 1 1 1 1 2 2 2 R R S R S S S HA HA HA HA HA HA HA 1 1 1 1 1 1 1 C S C C C S S Material Q-ty Rem.

36


List of mechanical parts. Continuation
Des. AS05 MD05A MD05B MD05C MD05D MD05E MD05F MD05G MD05H MD05I MD05J AS06 MD06A MD06B MD06C MD06C AS07 MD07A MD07B MD07C MD07D AS08 MD08A MD08B MD08C MD08D MD08E MD08F MD08G MD08H MD08I AS08 MD09A MD09B MD09C MD09D Part Viewer mirror unit Support Mirror holder Clamping cramp Cross cover plate Right axis Knob axis Switching knob Bushing Left axis Bushing nut Fo cal unit Field aperture Aperture support Central blind Filter holder MASS optics holders Mirrors sockets Mirrors cover plater Motor cramp Worm pivot Electronics b ox PMT house PMT cover Teflon spacer Shutter blade Shutter cramp Shutter lever Shutter axis Electronics frame Electronics cover Optical interface Interface tube Transfocator holder Holder support Locking nut No 3 HA HA HA HA 1 1 1 1 R R R R HA HA TF Steel Steel Steel Steel HA CA 1 1 1 2 1 1 1 1 1 C S S S S S S C S HA HA HA HA 1 1 1 1 C S C S Steel HA HA HA 1 1 1 1 R R R R HA HA HA HA Steel Steel HA Steel Steel Steel 1 1 1 1 1 1 1 1 1 1 C C S R R R R S R R Material Q-ty Rem.

37