ORNL-2658.txt

 

 

e ety e — i 1 . e St . 1

»

5

Contract No, W=7405-eng-26

REACTOR PROJECTS DIVISION

MEASUREMENTS THROUGH A HOT CELL WINDOW

USING OPTICAL TOOLING

A. A. Abbatiello

DATE ISSUED

APR 2 3 1958

 

.. OAK RIDGE NATIONAL LABORATORY
o Ock Ridge, Tennessee
operated by
UNION CARBIDE CORPORATION
for the
U.5. ATOMIC ENERGY COMMISSION

 

ORNL-2658

 

 

 
 

 

 

 

 
 

 

vy
 

 

 

1

MEASUREMENTS THROUGH A HOT CELL WINDOW USING OPTICAL TOOLING!
A. A, Abbatiello

ABSTRACT
.

Optical tooling was evaluated for the measurement of physical dimensions of radicactive parts

through hot cell windows. Instruments were set up outside a 4-ft-thick lead-gloss window. Al-

though the window was not specially selected, the readings were within 1.0% of the true dimension.
Use of a calibration chart of the window variations reduced the error to 10.1%. The method is

considered feasible and sufficiently fast for @ wide range of hot cell measurements.,

of the window's measurement qualities.

The reflections of a point light source from the lead-glass laminae form @ convenient indicator }

 

INTRODUCTION

The dimensional measurement of irradiated parts
has become one of the problems connected with
reactor development, These measurements are
important to determine strain, creep, and metal-
lurgical changes in parts which have become radio-
active and therefore inaccessible to ordinary

measuring tools. A number of measurement methods .

have been considered, one of which is optical
tooling working through hot cell windows. The

principal advantages of this system are: (1) parts .

may be measured without direct contact or con-
tamination, (2) convenient and comfortable working
areas can be utilized, (3) the cell interior is
entirely free for useful working space, and (4) ac-
ceptable accuracy "is attainable. Essentially,
optical tooling is measurement by means of o
telescope (called a transit square) which is capable
of moving in horizontal “and vertical directions
(Fig. ‘1). It is referenced from an independent
optical control line and may be aligned with the
object point located in space, Direct recdings
taken from a vernier scale mounted on the carriage
are subtracted to obtain the net distance between
two points. In the cases where measurements are

made through windows, a calibration - factor for .

optical distortion is then applied to obtain a cor-
rect reuding.'."Optifi:dl"tooli‘hg? has been in use in

 

‘Prepared for the 7th Hot Cell Conference.

2
W. L. Egy, ‘‘Optical Toeoling,'” The Tool Engr.
(April 1955). ¢

aircraft plants for years and is now being applied
in other industries requiring accurate location of
widely separated points, as on large assembly
iigs, heavy machinery fabrication, and shipbuilding.

The purpose of this test was to determine the
feasibility of taking accurate measurements with
an optical tooling system through a 4-ft-thick lead-
glass window. The locations and the effect on
the readings of optical defects in the window were
desired as well as the time required to take typical
measurements,

Optical tooling instruments were set up in front
of a new 4-ft-thick hot cell window. Although this
window was selected mainly because it was avail-
able, its optical properties proved adequate for
an acceptable range of accuracy. By calibrating a
specific window, the accuracy can be further in-
creased by applying the correction factors which
are determined by test. '

DESCRIPTION

Measurement by the use of optical tooling re-

quires that the points on the object to be measured

be visible in o direction perpendicular to the win-

~dow syrface thrb(:gh which it is viewed, and that
it be possible to turn the object so that the points

lie in a plane paraliel to the window. A cell

_gfiuipped with a lift and turntable would be most

useful for the larger parts which would be en-
countered in a typical reactor program, Smaller
parts might be adequately handied by manipulators,
or merely set up in front of the window aend prop-
erly oriented.

 
 

 

 

- ALIGNMENT TELEsc07

  

TOOLING BAR

TO OBTAIN DIMENSION "X" TAXE TWO VERNIER READINGS AND SUBTRACT.

UNCLASSIFIED
ORNL-LR-DWG 34274

  
 

Fig. 1. Optical Tooling Principles.

In order to provide the necessary range of meas-

urement, a transit-type instrument is held in a
plumb position and perpendicular to the window
surface, as shown in Fig. 2. Mechanical means
are provided to move the instrument in horizontal
and vertical directions to cover the complete
window area. An alignment scope generates a
reference line to which the 20-power transit scope
is set perpendicularly at the vertical plane and
elevation of interest.

- An,obticclly flat mirror, magnetically held on a

steel optical flat mounted on the wall at the side
of the window, provides the vertical reference

‘plane. from which the transit square and alignment

scope are positioned. This system is used to

establish a horizontal plane at each new elevation .

(see Fig, 2).

Fundamental Steps of Alignment

The following describes the basic steps in
setting up this optical tooling system:

1. The tooling bar is positioned parallel to the
front of the hot cell window in a convenient loca-
tion. The transit scope and alignment scope are
set respectively 90° from and parallel to the win-

dow (Fig. 2). .

2. The optically flat mirror mounting bar (straight-
edge) is placed in a vertical position beside the:

window and adjusted parallel with its front surface,
The optical mirror is held magnetically to the
optical flat, - | o

3. The transit scope is aligned perpendicular to
the plane of the optical flat {Fig. 3).

4, Without disturbing the previous setup, the’
alignment scope is positioned from the fransit -

*

C
 

 

 

.y

STEEL OPTICAL
T

ALIGNMENT
TELESCOPE

 

TOOLING BAR

VERTICAL
SCALE

 

._ : : a ,
N Yo/ ot ceLt

UNCLASSIFIED
ORNL-LR-DWG 34275

  
  
 
  
   
 
   
 
   
   
   

WINDOW

 

PRECISION
LEVEL

LEVELING

. ELEVATING
GEAR

Fig. 2. Onficdl Tooling for Hot Cell Test.

- scope and udiusted to bring it square by means 'of -

the built-in transit scope mirror, thus providing a
fixed reference line parallel to the front of the

~window.  This - control reference line is- ‘then
available for cll subsequent mecsurements at thuf, o
" level, : ' :

5, The transn square is moved to the flrst
position and aligned (Fig. 4).

6. The transit square is aligned with the part
(Fig. 5) and the vernier is read,

7. The transit square is moved to the second
posmon, allgned and read,

;8._,,A_; change to unother elevation is made by
repositioning the magnetically mounted wall mirror

B '(Fig'.' 2) along the steel optical flat, and repeating
- the - perpendicular setting of the transit scope to

this mirror. The alignment scope is then raised
and positioned from the transit scope, thus estab-
lishing a new reference line in a plane parallel to

 
 

 

 

 

Fig. 3. Tronsit Square Is Set with Opiicu"y Fliat Reference Surfuée.

 

 

UNCLASSIFIED
PHOTQ 414763

 

 

 
 

Fig. 4. Transit Square Is Adjusted Perpendicular to the Alignment Scopg.

o

UNCL ASSIFIED
PHOTO 41764

 

 

 
 

 

«

Fig. 5. Truns!t Square Is Aligned

 

with Grid Plate.

 

UNCLASSIFIED
PHOTO 41765

 

 

 

 
 

 

"

 

the front of the window, and at the elevation de-
sired for the new set of measurements.
Equipment Requii‘é& L
Commercially available optical tooling equip-
ment was set up for evaluation under practical hot
cell conditions. Vertical measurement was impro-
vised using a vertical height gage and a precision
level (Fig. 6). In a typical hot cell envisioned,
the test arrangement would be replaced with a
combination tooling bar as sketched in Fig. 2,
The purpose of the combination tooling bar is to
provide easier manipulation in the vertical plane
than is available with the standard tooling bar. It
has been estimated that the combination tooling
bar and all the associated equipment for one com-
plete setup would cost about $6000,

GENERAL PROCEDURE

System evaluation was done by setting up the

instruments at the hot cell window and taking -

readings of an accurately scribed plate at three
locations: (1) outside the cell, (2) with the plate
hung just inside the cell (to determine the wedge
displacement of the window), and (3) repeating the
readings with the plate at the most distant (about
9 f1) portion of the cell, to get the maximum angular
deviation, '

Typical Example of Measuring a Part

- A cylinder was placed in the cell ot a point
about 5 ft back from the cell window, and the
diameter was recorded using. the optical tooling
instruments, Table 1 shows a first and a second
try, with the cylinder moved to o new position for

the second test in order to view through a different
portion of the window. The cylinder was 6,800 +
0,002 .in.,-in diameter and had @ shiny surface,
which was not considered the ideal for best accu-
racy, but was used as an example of practical
system evaluation, A view of a part inside the
cell is shown in Fig. 7,

DISCUSSION

Although the instruments may appear complex,
they were surprisingly easy to operate, During
the calibration of the cell window, the complete
cycle of raising the tooling bar, precise leveling,
squaring with the optical wall mirror, taking 18
sets of readings on the horizontal scale using the
transit square, and also toking 18 corresponding
points vertically with the optical micrometer attach-
ment were made in about 1 hr. Operators were
trained rapidly in the use of these instruments,
During the course of this test, four different
assistonts were used, some for as little as one
day, and each was able to learn the technique and

The reproducibility of the results was determined
by taking a set of readings along the same hori-
zontal line on three different occasions when the
instruments had been removed and replaced. These
data, plotted in Fig. 8, show a dispersion of about
0.030 in, for length measurements of 20 to 30 in.,
which indicates a reproducibility of about 10.1%.

- Further improvement probably could be obtained

by using a more permanent setup with refinements

-such as floor plates having dowel bushings

mounted in the floor. This would have the added

advantage of making it easy to remove the instry-
“ments” for use at other installations, or to free

" Table 1. Measurement of a Cylinder

 

 

: :Fir'st Try Second Try
Right side of cylinder, vernier reading, in. .~ 48.959  44.488
" Left side of cylinder, vernier reading, in. 42,160 | 37.664
. ,Differqnce{ '(éylindér ;Vlii',.nm;eter zfilllc'ortect'ed.), in, 7 6.799 6.824
' _B-eéd-’.a:.s'q.-.:th'i..r_;_'obiec'f'w,u."_srot' 5 ft, _u#@‘/z -cdrr_ecfiqn : -0.022 |  —0.018
- (from a Vf_ub'le_pr‘epoired‘.for this winde);-_ i_h; : ’ :
Cdrriec'féd_dirarhe‘tér, )i'rrs. | 6777 | 6.806

Accuracy

6.777/6.800 = 99.7% 6.800/6.806 = 99.9%

 

 
 

   

 

UNCLASSIFIED
PHOTO 41862

 

'Fig. 6. Data Taken Using Vernier Height Gage, Precision Level, and Optical Micrometer,

   
 

 

 
 

S 5 | PHOTO 41864

 

 

 

 

'Fig. 7. Measurement of a Part Inside Cell.

 

 
 

 

 

 

NOTE:

UNCLASSIFIED
ORNL~-LR-DWG 34276

THE INSTRUMENTS HAD BEEN REMOVED AND REPLACED EACH DAY

GRID PLATE WAS 9ft 9in. BACK OF CELL WINDOW

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2 0.250 , —
w o . ‘ B 2
© 0.200 |
Z A OCT 4. A/??
o O OCT 7. A .
u b
w 0450 I~ ¢ ocT 9 -
a
w ‘ ‘ —
> 0.100
> 0.050 ==
2 0.
3
| o Az . |
o a 8 12 16 . 20 24 28 32

DISTANCE ALONG WINDOW AT LOWEST VISIBLE LINE (in.)

Fig. 8. Window'Calib_rution.‘

the window when other work is in progress, Work-
ing area may be provided directly .in front of the
window without interfering with the optical tooling,
since the distance over which the instruments
operate may -be increased considerably without
appreciable loss of accuracy. A permanently
mounted optical flat at the side of the window
would aid in maintaining a common reference plane
as the basic starting point.

No special effort was made to use a selected
window; the one tested merely happened to be
available, and therefore high optical accuracy was
not expected. It is of interest to note that a pin-
point light source revealed noticeable variations,
although the window is of acceptable accuracy.
If all laminae have parallel surfaces and are
assembled into a paraliel pack, the light-source
images would lie in o straight line when viewed
from any position. The use of a simple light
source appears useful as an aid when assembling
windows, because the alignment of each lamina
could be checked easily as it is being placed. .~

SUMMARY

1. The method proposed for linear measurements
through hot cell windows is considered feasible on

10

the basis of the equipment and methods used in a
hot cell test at ORNL, Views of the components
in use are shown in Figs. 3, 4, and 5,

2. The accuracy of measurements taken through
this particular window without correction is about
99% and 99.87% respectively for readings taken
93/4 ft and 6 in. inside the cell window.

3. By using the calibration chart produced for
this window during the test (Fig. 8), accuracy can
be improved to about 99.9% (or 10.1% variation)

- for  long cell distances, which approaches the

practical accuracy of routine shop measurements
in non-radioactive work using conventional methods,
As the work is brought closer to the window,
errors are proportionately reduced. |

4., The speed and accuracy of taking measure-
ments is high for hot cell work. It may be com-
pared to typical experience using a cathetometer
or a sutface plate and vernier height gage.

5. A pin-point light reflection test is a simple
method to evaluate a window for potential accu-
racy, and might be developed further to assist the
manufacturer in selectively assembling glass
sections for best precision, '

6. A zinc bromide or other liquid-filled window,
having the minimum humber of light-refracting sur-
faces, would be preferred for measuring purposes.

C
 

 

 

-}

el

“Since optical tooling is used perpendicular to the

viewing surface, chromatic aberration is not a
problem. Because of the lower density of zinc
bromide, however, greater thickness would be
necessary to obtain equivalent shielding.

7. Operators can be rapidly trained in the use
of these instruments,

8. Hot cells planned for accurate parts meas-
urements could use this measuring system by
calibrating a window or selecting one with suit-
able properties.

ACKNOWLEDGMENTS

The assistance of W. W. Alto, Chief Engineer
of the Brunson Instrument Company, is acknowl-
edged for the contribution of the idea of an opti-
cally flat reference surface from which to align

 

 

the transit scope for different elevations. A. N.
Brunson, President of the Brunson {nstrument
Company, Kansas City, Missouri, provided the
instruments which were loaned for this test,

The assistance of R. J. DeBakker in setting up
the equipment is acknowledged, as well as the
cooperation of D. E. Ferguson and C, P. Johnston
for the use of the cell. Credit is also due to
T. E. Crabtree, H. W. Hoover, C. K. McGlothlan,
and J. J. Platz for help in operating the equipment
and taking daota,

The complete review of this report and the many
helpful comments of D. B. Trauger are gratefully
acknowledged; also, the assistance and encourage-
ment of M. Bender, W. F. Boudreau, and F. R.
McQuilkin are appreciated.

1

 

 
 

 

 

 

  
 

 

 

 

INTERNAL DISTRIBUTION

- 1. C. E. Center 67.
2. Biology Library 68.

. 3. Health Physics Library 9.
4-5. Central Research Library 70.

6. Reactor Experimental 71.

Engineering Library 72.

- 7=26. Laboratory Records Department 73.

27. Laboratory Records, ORNL R.C. 74.

28. A. M. Weinberg 75.

29. L. B. Emlet (K-25) 76.

30. J. P. Murray (Y-12) 77.

31. J. A. Swartout 78.

32-36. A. A. Abbatiello 79.

37. S. E. Beall 80.

38. M. Bender 81.

39. D. S. Billington 82.

40. E. P. Blizard 83.

41. A.L.Boch 84.

42, C. J. Borkowski 85.

43. W. F. Boudreau 86.

44. G. E. Boyd 87.

45. E. J. Breeding 88.

46. R. B. Briggs 89.

47. W. E. Browning 90.

48. R. S. Carlsmith 9.

49. R. A, Charpie . 92,

50. R. Clark 93.

51. W. B. Cotirell 94.

52. G. A. Cristy 95.

- 53. F. L. Culler 9.
54. R. J. DeBakker 97.

. 55. S. E. Dismuke R
¢ 56. H. G. Duggan 99.
57. D.E. Ferguson . 100.

58. W. F. Ferguson g 101.-
59. A. P. Fraas 102,
60. E. A. Franco-Ferreira - 103,

61. E. J. Frederick 104.
' 62.- J. H Frye, Jr. 105.

 63. W.R. Grimes 106,

64. E.Guth - 107,

65. C.S. Harrill - 108.

-

66, W. R. Harwell

109.

EX TERNAL DIS TRIBUTION

110. Dmsmn of Resecrrch und Deve |opment, AEC ORO
1”-633. Given distribution as shown in TID-4500 (14th ed.) under Instruments category

H.
A.
A,
C.
w.
G.
C.
M.
J.
R.
R.
M.
H.
W.
E.
J.
F.
A.
K.
G.
M.
A.
P.
A.
E.
P.
F.
A
H
A
H
R
E
0.
M.
A.
E.
D.
C.
G.

ORNL.2658
Instruments

TiD-4500 (14th ed.)

W. Hoffman
Hollaender

S. Householder
P. Johnston

. Jordan

. Keilholtz

. Keim

. Kelley

. Lane

. Lindaver
Livingston
Lundin

. MacPhetson
. Manly

R. Mann

R. McNally
R. McQuilkin
J. Miller

Z, Morgan
Morris

L. Nelson

R. Olsen
Patriarca

M. Perry

E. Pierce

M. Reyling
Ring, Jr.

DO wpP—4 V=T

. F. Rupp

. W. Savage

. W. Savolainen
. E. Seagren

. P. Shields

. D. Shipley

Sisman

J. Skinner
H. Snell

H. Taylor -
B. Trauger
D. Watson

D. Whitman -

G. C. Williams -
C. E. Vinters -
ORNL ~ Y-12 Technical Library,

Document Reference Section