."

"7' -; ~ .. - -.

J •

',- .~

NASA-CR-176818 19860016285

A Reproduced Copy OF

i Reproduced for NASA

by the.

NASA Scientific and Technical Information Facility

'. : .... q .: "":iG ,1',1,' ' .... ( ,_ .~,~

FFNo 672 Aug 65 . 111111111111111111111111111111111111111111111-

NF01199 ,'.

https://ntrs.nasa.gov/search.jsp?R=19860016285 2018-06-26T04:41:37+00:00Z

I .. , " !

... ;.., .' - .\ .......

"-~ ... .!-

_" J

! . !

'''.

Final Report for

NASA Grant NAG 1-303

Development of aNew Instrument for

Direct Skin Friction Measurenlents

Prepared for

Experimental Techniques Branch NASA Langley Research Center Hampton, Virginia 23665-5225

by:

A. D. Val~i!i, J. M. Wu Gas Dynamics Division

51B6

The University of Tennessee Space InsUtute "i til e,l1"-C/vil Or Tullahoma, TN 37388 ''''' n

March, 1986

UTSI 86-02

(NASA-CF-176~18) DEVELCPHENT C~ A NED 106-25757 INSTBUHhUI FOb CIriel ~KIN F&lCIION nlllsun:r';EI;1S Fin.ll Repot:t (TE.nz;c::see Univ. Space Inst., !U 11 dhcma.) '=2 f He A\13/MF A01 Unclas

CSCL 14U GJ/35 ij354U

'.

.......

TABLE OF CONTENTS

SUMMARY .

I. INTRODUCTION

II. DESCRIPTION OF THE INSTRUMENT Principles of Operation . . . Strain Gage Sensing Technique

. • . • . 1

· . 2

• -1

· '. 5 Optical Sensing Technique. ..... . . . . . 6 Vibration Effects . . .. . . Calibration Procedure

III. MEASUREMENT FACILITIES Unitary Plan Tunnel O.3-m Transonic Cryogenic Tunnel ARL Mach 3 Tunne! ...... .

IV. MEASUREMENTS AND DISCUSSIONS

V. CONCLUSiONS AND RECOMMENDATIONS

- 1 -

· . 6 7

. . . . 7 . . • . . . 7

· . 7 . . . . . . . 8

· 8

12

....,---------. --."..,---._:-.::,-: • .:' ............ Jw3- .. -.::< ~ '-: .. ·I"~·~"'-'· ~ ------~:w...IW\.l.:.;.~i":..:';;.."! ,.. .. '~-. ______ .... ___ -...;..,t ___ ._ ~_~~~ ________ .

. i ~ ! l

I r,

I ~

SUMMARY

A new instrument has been developed for direct measurement of wan shear stress generated by flows. This device is described in this report with typical measurement results. This instrument is simple and symmetric in design with optional small moving mass and no internal friction. Features employed in the design of this instrument eliminate most of the difficulties associated with the traditional floating element balances. This device is basically small and can be made in various sizes. Vibration problems asso­ciated with the Hoating element skin friction balances have been found -to be minimized due to the design symmetry and optional damping provided. The unique design of this instrument eliminates or reduces the errors as­sociated with conventional floating-element devices: such as errors due to gaps, pressure gradient, acceleration, heat transfer and temperature change. The instrument is equipped with various sensing systems and the output signal is a linear function of the wall shear stress. Dynamic measurements could be made in a limited range and measurements in liquids could be. performed readily. Measurement made in three different tunnels show ex­cellent agreement with data obtained by the floating clement devices and other techniques.

- 1 -

----....,.~',~

. ',. . .~~-. " ;

..... ~""'-....... ----

Throughout the years the importance of viscous effects in aerodYliumics has motivated a great number of both theoretical and experimenta.l in\'(~s­tigations in the field of boundary layer flows. particular attention has been given to the frictional forces introduced by these effects on the surface of moving bodies. For the laminar boundary layer the theoretical evaluations of the frictional forces has reached a well advanced stage. But for turbulent . boundary layer there is not a generally accepted theory to predict accu­rately the frictional forces. In particular the skin friction associated with the turbulent boundary layer presents an important problem today. Correct skin friction measurement not only can serve to offer vital information but also can be used to check the accuracy of theoretical modeling techniques.

Direct measurement of skin friction was originally an essential step in setting-up basic skin friction laws. Direct measurements of skin friction by I{empf (Ref. 1) in 1929 and others (Ref. 2, 3, 4, 5 & 6) later forllled the basis for the generally accepted skin friction estimation for incompre~;sible flow. However, because of our limited understanding of turbulent flows there has been the need to extend direct skin frict-ion measut'ements to compress­ible flows. In many circumstances there exists no theory, even experimental measurements are not simple. For example, in flows with strong pressure gradients and unsteady flows, accurate skin frict.ion measurements arc dif­ficult or impossible.

All aerodynamic experimentalists are well awa e of difficulties a5~")ci­atcd with the direct measurement of skin friction. Most of the existing skin friction data have been votained through indirect method!> such as Preston tube, surface hot film, boundary layer velocity profile measurements, etc., inf:tead of direct measurements. There are many difficulties and limitations associated with the application of these indirect techniques and III 0 reOVl'r, these are approximate approa.ches.

The know\edgt! of skin friction distriblltioti on PlOclds at high Hpynolds numbers is ('xtremdy helpful fol' the accurate prediction of iLl'rodYIlHllIic

loading and perfor'mance, delt'I'ntinalioll of transitilHt point.. regioll of 11m\" separatioll and other flow characteristics. Because of tlH~ limitatioll:; assod­ated with the prediction techniques, direct measure;ll(,lIts of skin friet iOIl is the onty accurate way to identify the shear stress kl;tding at critical points

-2-

-'-

-.

I J

i f· I .

on realistic bodies.

In the last decade tremendous progress in the computational aerody­namics has been accomplished. Often the accuracy of numerical models are tested against experimental results especially in regions of stronggm­dicnts in the flow. Obviously reliable and accurate experimental data are investments of high gain. It is therefore necessary to improve the experi­mental level of the aerodynamic data in rcgions of insufficient or inaccurate data availability. The significance of a simple and functional instrument, for direct measurement of local shear stresses, is evident.

Most of the conventional direct skin friction force measuring dcvices are very similar in concept and fundamentals of operation.

There are some more elaborate designs in order to compensate for some of the commonly known difficulties. K. C. Winter (Ref. 7) described somc of the problems associated with the direct skin friction measurement devices. The Naval Ordinance Laboratory (NOL) balance (Ref. 6) is one of the most accurate skin friction measuring devices. The NOL balance is based on the same principle of a one time commercially available J\:istler's balaJl(~e (Ref. 7). Our skin frictio~ transducer was designed and manufactured as an alternative to eliminate the troubles we encountered with the NOL balance .

. It has proved to be a very capable instrument with great potentials as· discussed later. Detailed description am\. advantages of the belt instrument is given in the next g. ::tion, and in the p..lper by Vakili and Wu (Ref. 8). An important feature of this transducer is its ease of handling and simpiicity of calibration and measurements.

It is obviously desirable to have a small skin friction transducer so that one or more of them can be embedded onto a model surface for direct local skin friction nlcasurcments. Similar to pressure, temperature, velocity data etc., skin friction data on a model has significant applications. So far the' problem has been thc la<;k of a rea.~onably small size transducer. We now believe that our dc\!ice would <.'ventllally fulfill the requirements. .

At the pr('sent our instrument. is relatively slIlall I" wide, 1.·1~" long. 1.5" deep (2.5·t em x :1.76 cm x :U~ I em) compared to other direct skin frict.ion instruments. However, tlsing opt.ical tedllliqIH'S for the det<'ction of the strain 01' deflection of the flexures would enable fU1'ther reductioll of

-3-

--=:~.'i~

*5aif:~·~-'..Gm""c_,,~,

I -. I

the total size. So far tlwre has been no need for real miniaturhmt.ion of our device. Miniaturization would result in a small area of the instrument to be used to measure the wall shear stresses. This in turn results in smaller forces, therefore, high precision sensing techniques (especially optica.l) arc needed to measure small loads and maintain sufficient accuracy and need to be investigated further for applications in (the miniature size of) this transducer.

Principles of Operation

The Belt Skin Friction Balance (BSFB) consists of a flexible belt (tape) wrapped continuously ~nd tightly over two cylinders separated by a small distance. The cylinders are mounted on frictionless flexures which are sup­ported at the ends in a solid housing. A portion of the belt replaces the surface over which the fluid flows. The belt can be replaced with belts of different mat.erials and surface finishes, for measurements of surfcce shear stress on the corresponding surface environment.

The shear stresses applied by the fluid to the belt causes the cylinders to rotate (the amplitude of rotation is designed to be less than three de­grees). The incremental rotation of the web is directly reIated to the shear stress 13?plied to the belt by the fluid. Strain gauges arc installed on both sides of the flexure web to measure the amount of deflection. Each flexure, having no bacIda'3h and requiring no lubrication, is double end supported and made from a pail' of stainless steel flat, cross-springs supporting rotat­ing sleevcs. Figure 1 shows a schematic diagram of the inn"cr"details of the BSFB instrument, this figure is. also demonstrating the principle of oper­ation of the BSFD. Figure 2 shows the details of a more compact version of the same instrumcnt with dimension~ of: 1" wide, 1,48" long and 1.5" high. Different sizc and range instruments arc required for acctlrate rnea­SU1'CIlH.'llt of the shear stress loads encountered in "arioll:; nows to provide good sensitivity and resolution.

For ('ach particular ill!'trUIll('llt the SWrIH'SS of the f)<'XIII'e wd)s are' selected stich that, the all~lIlal' deflection for tlte ('xpected maximum force exert('ti Oil the area of the helt exposed to the fluid is (lot more than three degrees. The commerieally available flexures arc manufactured to provi<1r.!

- -I _.

-.

\.

r ( r

f ,- I

t [ l

I

frictionless pivots for limited angular travel which a.utomatically prevents them from being damaged by overloading.

Because of the necessary movement of the belt, it is essential that there be a small gap (clearance) around the belt. 'This corresponds to the gaps around the floating element of the common balances. Since the belt moves on cylindrical (circular) path t.here is no change in the clearance dimensions created during measurements. There is no. evidence that the presence of these gaps influence the accuracy of the transducer. There is also no particular requirements on the gap dimensions, except that the belt must move freely without contact with the top surface and sides to ensure frictionless operation. The presence of the gaps allow for pressure equilibrium on both sides of the belt's exposed surface. This is of particular importance for measurements in flows with transient behaviour or with pressure gradient. In this respect a finite gap is realistically desirable to allow instantaneous equilibrium.

The gaps created by the necessary clearance between each cylinder and the top surface also depend on the manufacturing and the assembling preci­sion. These gaps have been kept to a minimum in the present instruments. It is interesting to note that for small deflections of the flexures no change in these gaps is made.

Strain Gage Sensing Technique

Electrical resistance strain gages are used frequently to measure strains on mechanical elements in form of electrical signal. This t(!Chnique is well developed and quite accurate measurements are being made using wide varieties of strain gages. In the present transducer, foil t.ype strain gages have b~en installed on the flexure webs in order to convert the wall shear stresses on the belt surface to electrical signal. There arc two st.roain gage ·bridges installed one on each flexure, with the two bridg<,s cOIHlC'cted in parallel. This provides fOf reduced dbturbance effects which lIlay he resulted by rotation of the two flexures in opposite directions.

EVl'1l thou~~h st.rain gages arc widl'ly lIs('d, tht're art' SOliI<' limitations associated with their usc. [n general. envirollrtH'lltal cliallg<'s, suet. as tlte ambient temperature effects the rt'sistance of strain gag<' which in t IIrn pro­duce erroneous straiH gage output not necessarily rdaled to the designed

-5-

'r'" ','

~ "

t:

".

f ! L

f

-I! "

conditions. To minimize this, normally strain gage bridg(~s are tcmp(!rature compensated for a reasonable range of temperature variations. However, it is also common practice to calibrate the temperature dred on the whole. transducer, since full compensation fot' a very \yide temperature range i~ frequently almost impossible.

We have ahlO developed an optical sensing technique to replace the strain gauges and to eliminate some of the associated problems (like tem­perature effect). Using fiber optics it is possible to directly detect the trans­lation of the belt due to the (shear stress) forces applied to the m~astlring surface. The optical arrangement is shown in Figure 3 a,b. The amount of light backscattered from a special geometry reflective surface is related to the relative position of that surface. In a bench experiment, infrared light was directed on to the reflective surface and the back-scattered light was sensed by a light sensitive transistor. The voltage output was proportional to the reflected light which is in turn proportional to the displacement of the surface reflecting the light (Rcf. 8).

The local optical technique is incorporated into the instrument froill the opposite side of the measuring surface, therefore no form of interfel'­ences is expected in the flow. There also exist.s the possibility of displace­·ment detection remotely, through interlcrometric techniques, which is yet to be investigated. Using these optical techniques shall make a significant reduction of the instt clment size readily possible.

Vibration Errcct~

The effect of vibration on the instrument is minute, however, becalJ~e of the amplification of the signal the vibration effect is also amplified parall<'l to the force signal. nllt, since the vibration ('[feet is symmetrical it call be separated from the (I'orce) signal by avcraging or liItprilig lll<' oUlPUt. (Ih-f. 9). There was no need for d<Ullpillg for all of thc 1II('aSIlI"('IIl(,llts l'('porl (·d here. In this I'('!'pect it is not Iwces!'ary to pl'Ovid(' damping ('x<"<'pt ill v<'I'y harsh envirollments. .

To reduce the vibration sensitivity, damping can h~! provided by ... i~("olls oil in the damping gap provided for this purpO!iC (Figure 1). The result of

- 6-

, ._---------_.----" . ~

~" <: ~..: .. ; ..... ~~ .... ~' •. :,' --------------.,-_. ,-

'.

,.:.. ,

( I t I i , l

! t~

damping provided in this manner is extremely satisfactory. The viscosity .. of damping oil may change due to temperature changes. Therefore, it is recommended to use an oil to provide a proper viscosity for the conditions of mea"urements. The oil damping tcchnique may be replaced by magnetic or other means. This is essentiai, since at certain measurement conditions (i.e., for low temperatures) oil cannot be used as a viscous {luid.

Calibration Procedure

Because of the symmetrical def:(Jn of the instrument, it could be used at any orientation. To calibrate the instrument it is held such that the sensing surface is vertical, and dead weights are hung at one end of a very thin nylon thread that is attached to the sensing surface. The output of the transducer is recorded for different dead weights in order of increasing and dccreasing weights to make sure of the perfect linearity of the instrument. A sample calibration curve for this transducer is shown in Figure 4. The slope of the calibmtion line depends on the flexure stiffness and the sensitivity of the strain gages, which are design parameters and are built into the instrument.

III. MEASUREMENT FACIL!TIES

1. Unitary Plan Wind Tunnel

Tre Langley Unitary Plan Wind Tunnel is a closed-circuit, variabl~­pressure Facility with two variable Mach number test sections which ar.e approximately 1.22m in height and width and 2.13m in length. The Ma.ch llumbe. ranges of the two sections are 1.46 to 2.80 and 2.30 to 4.6.'3. D(~t.ail

description of this tunnel is given in Reference iO. The nominal test section unit Reynolds number is 6.5 x 106 per met~r. The test section stagna­tion pressure and tcmperature are controlled to vary the Reynolds number. Present measurcmcnts . were performed in the first test section, with the lowcr Mach numbcl" rangc, at a. nominal ~{ach number of 2.16 for various Reynolds numbers.

This tunnel has a. two dimensional tcst section :W x nocl1/. with lhe operational parameters as Iist<,'d in table below. (t.he data on this chart is takell from Itcfcr<"r1ce 1 t. The 20 cm sidc \vall is :-;iou('d and the ()() cm

- 7 -

f __ -_ .... --...,·_----r·-·· ~t- ---~ .' '.

-~---.-.~;, .. ;;.. .... -:.~-":,~ .. ;-)",---- -'_.'-" -_._------_._--

. ~ .

'~

r r I I

wall is solid. The solid wall was used as a flat plate on which the various transducers were measured. More discussion on the waH is given in the next section.

Table I

Parameter Range of the O.3m TCT

e Continuous Runninl~ • 20 X aOem slotted tClJt section G Operating total pressure 1.1 to G.12 atm. • Operating total temperature 77.4 to 340°I{ ct Mach number 0.02 to 0.94

• Re/m 0.4 to 400 million

3. ARL Mach 3 Tunnel

The Aeronautical Research Lab (ARL) tunnel is located in the Wright­Patterson Air Force Base in Dayton, Ohio. It is an intermittent (blow-down) tunnel with no total temperature control. At nominal tunnel Mach number of 3.0 the unit Reynolds number can be carried from 20 X 106 to 95 X 106 per foot. The test section is 8" x 8.2" x 24". The measuremen~.s were made at average stagnation temperature of 480° R and at three nominal stagnation pressures of 85, 240, 400, 505 and 560 psia. The corresponding free strcun Reynolds numhers were 18.6 X 106 to 93 X L06 per foot. More details on the ARL Mach 3 t.unnel could be found in Reference 12.

IV. MEASUREMENTS AND DISCUSSIONS

Tests in the Unitary Plan Wind Tunnel of NASA Langley were the first attempt to conduct systematic measurements by six different transducers, simultaneously. Previous measurements in this tunnel hy .f. ~1. Allen (Ref. 13) using various techniques had re:mll.ed in a h~rge rmmher of skin friction data sp~cilically at "t.he slime measurements location ·'.lId lin\\' condit iOlls. A window in the tunnel wall had lwen rcplaC'{'d with .'~ :;oli<l plat(' Oil which different. transducers were Hush mounted. Figure 5 is a photo takl'1l rWIIl

thi~ plate and shows the relative position of the dilrercnl transducers. TII('f(' were three floating dement balances, two sllrfac(~ hot film sensors and ollr belt skin friction transducer. The tests were sponsored by the NASA lItD

-3-

I

to evaluate the surface hot film sensors against the floating clemcnt bal~ ances, which in turn had (noating) elements of different sizes and :materials. Therefore: the measuremcnt.s by the BSFB was on a piggy back basis arid had no influence on the run conditions.

Many runs were made at a nominal Mach number of 2.16 for variolls Reynolds numbers Rei It = 0.5 X 106 to 11 X 106 , at Tt = 125°F. The mea­surement results were plottcd on the graph containing the data by Allcn (Ref. 13) as shown in Figure 6. Allen has used the Van-Driest's technique to convert the data to incompressible values in order to compare the mea­surements to that of an incompressible flat plate. The agreements between the measurements made by our BSFB and Allen's data appear to be quitc guod for the high Reynolds number range.

At [ow Reynolds numbers there appear to be some scatter in all of the data. This is attributed to two possibilities: 1) The tunnel flow may not bc quite uniform and steady, 2) The load corresponding· to the low Reynolds number falls in the lower design range of our transducer. In addition to . the above two reasons, we experienced another source of error. This hap­pened due to evacuation of the tunnel which sucked small particles, from the seating day around our transducer, into this instrument and resulted in a small zero shift. At low Reynolds numbers the small zero shift influcnced the data more. The contamination wa;, identified only when the transduce'r war removed to be inspected for damates causing a zero shift. It is desirable to repeat these .measurements, if possible, in thefulure .

. Variations in Reynolds numbers were accomplished by changing the total pressure in the unitary plan tunnel. The strip ·chart recording made during transition from one total pressure to another indicates tlw corre­sponding transient readings by our transducer as shown in Figure 7. Since there were intentionally no damping provided, one can see the small (SYIII­metric) influence of tu[}nd vibration 011 the lralll'ciucpr OUtpl1t.

Similarly meahlrem(;ilt~ were made with the BSFB illsl.rullwl.11. ill III<' O.:!om cryogellic t.ullnel of the NASA Lallgley RC~l'ardl C(·lIt(,I·. TIt(' O.:~III

cryogenic tunnel 1I::.s a very wide range of H.eynolds IltlllllH'rs cL'\ stal('d ill Table 1. The tUl·Hld total temperature was varied over a wide rallg(' of T = 300° [( to T = 100° f(. The op('rating t(,lllperatllrt'rallg(~ of t.he tun­nel during these measurements \VeL'; beyond the ra.nge for which lhc BSFB

-9-t

t.r---------.-'~ .

/

.,' ~,

---"

instruments strain gages were temperature compensated. Therefore, the instrument was calibrated at various temperatures for which measurements were to be made. These calibrations represented linear functions of output vs. input for each temperature.

The data shown in Figure 8 v"ere also measured during a special test designed to exercise a number of new instrumentation tcchniques in a high pressure, transonic, cryogenic environment. To insure a minimum of distur­bance during these tests, there was no model mounted in the tunnel, and extra time was allowed for data acquisition to allow repeat points and to minimize temperature gradients. Since the skin friction measurements were to be taken on the tunnel sidewall, special care was also used in fitting the sidewall components to minimize boundary layer disturbances. However, the data in Figure 8 exhibits a rough wall trend, i.e., very little change in C, with increasing Reynolds number. It is instructive to consider the distributed wall roughness height which might predict the observed data .

The phenomena of rough flat-plate flow is discussed by White (Ref. 14) where an approximation for C, in the fully rough regime is presented as

C, = (2.87 + 1.581og (~) ) -2.5

where x is a characteristic length and ( is roughness height. A wind tunnel waH is not a perfect representation of a flat plate since there is no leading edge altd the turbulent boundary layer is preceded by a run of high pressure gradient flow in the contraction section. Thus, in order to properly utilize the above expression, it is necessary to establish a virtual origin for the turbulent boundary layer which will define a characteristic length, x, and a meaningful ratio of length to roughness height, x/f..

Reference 15, prescnts measured values of boundary ·laycl', displacc­ment, and momentum' thickness for the O.3-m TCT as a function of free­stream Mach and HeYllolds number. This information allows the determi­nation of IlclJ whidl together with the relatioll

Ilc6 = 0, W( RCr.)<>I7

from reference 1-1, pel'mils calculation of /lex, and finally the calculatioll of x from

-10-

"

. '

r , .

. '.

I "

..

Substitution of this value into the expression for C f now yields a value for each distributed roughness height selected. (Note that this formulation for x reintroduces a Reynolds number depenc1ence int.o the exprc:;sion for C f ). Curves representing a range of values for roughness hejgh~ arc plotted in Figure 9 , and values in the range from 0.005 to 0.02 mm (0.0002 to 0.0008 inches) are typical of the data. -

Even though joints between component r~arts are being rept"csented by a distributed sand type roughness in this <'.nalysis, it is ob\"bus that small disturbances will producp. the trends shown by the data in Figur"!!t. Also, when considering that this data was taken over large changes in tcmper~~ure and pressure, which would cause small changes in the relative positions of the components, it is not su:-prising that the lowe"t temperatures lie in a band unto themselves. Finally, the classic approximation for smooth wail skin friction

C, = O.027/(Rex )1/7

is shown, and is seen to approach the measured data at the lower Reynolds numbers which were obtained at conditions near a.tmospheric pressure al:a ambient temperature.

Measurements in the ARL Mach t.h~ce tunnel were performed in coop­eration with the Experimental Eugineering BraLch, on a piggy-bac:'- hasis. Ther.-r.as been an ongoing developmcllt project on a noaling clement di­rect ::;kin friction measuring balance for the past three years. The h,llan-:e that wa'; U:::!G was the same NASA balance used in the other tunlll·ls and develop.;>:'! ~y T::hcng (Ref. 16), with a heater built ~ntv the balance to warm up various internal components and to reduce pot<'ntial dalr.ag<: and thermal drift. The BSFB tr? sducer was the same one used in the previoas measurements with no modifications except that it was cleaned thorollr,hiy.

The two transducers were installed on the tunnel floor at the down­strcam of t.he 1I0707.le wher~ the wall represent approximately a rIat plate. The tunnel flow parallu!ters and the transducers output were sampled thwlIgh­Oll~ each I"lm by the data acquisition complitl':. Tlh~ lral\~dun'rs OIl',pU!

along with the the :-;ignal from an aCCcier(l(lwt<'r 1ll(,lIl1lr'<i 011 lh\~ ttllllld

wall were rccord<,d Oil a galvanometer type strip d'a.rt recor<il'r a.~ SIHo('<iS

up to 80 illches per second. There were thermocot!i>b, inslall<'d il1~!:I,~ ('ach transd~lcer which were also recorded by the data .1CC\,lisilion syst<'H1 ami tIll

-11-

''F:!:I------­.,.-:. "

1~ --_ .... _-------_._--

_ ~ ~ ... __ :_~ ..... ;.:,..i-;.,.t..Im.~j:·~:i~·.1 ~-: . . \ ... _--- '. \. • .. -' "-•.•. c,;....:-: •.• ' .• J. .. ,_.~ ._ •

/

.' ." :

!;.

"

'.,.

r i , r i

, .. I

",--,

i ! t

I

.-}"--/

.: ,

" .. " .:;

:. ~r

strip chart recorder. This was to identify the oJ.H!mling temperalur(' of each instrument for possible corrections.

Typical run times were in m-der of 30 seconds during which the lun­nel total temperature dropped ne""rly 30°C. The modified NASA balance required a one to ten minute warm-up time (by the heater) in bctw(.'Cn the runs. Figure 10 shows a plot of the data measured in the ART. tllnnd for the two transducers. The agreement appears to be reasonably good with the NASA balance slightly higher even when compared with the Van Driest (Ref. 15) nat plate theory for compressible flow. Each data point plotted represent at least twenty data points.

In order to investigate the influence of temperature dr~p (heat transfer) on the performance of the two transducers, the ARL tunnel was operated during one run, with the tunnel stC1gnation preSt;ur~ P, = 240 r·sia, rOl' 50. seconds. A comparison of the skin friction coefficients measured bj' the two transducers and the theoretical prediction by Van Driest (Ref. li, 18) for various temperature ratios is sho"m in Figure 11. As has been observed be­fore, the data by the BSFB appear to lie closei' to the flat plate theoretical line. Figure 12 shows a typical section of the strip-chart recorder indicating the trends in the two transducer signals and the vibration signal by the ac­celerometer transducer. It is noteworthy to compare the effect of tite tunnel startin~t shock as well as the stopping shock on the two signal. II appears that th~ BSFll transducer has a better dynamic' response, as expected.

IV. CONCLUSIONS AND RECOMMENDATIONS.

A l'elativcly small belt skin friction balance instrument has been de­veloped for direr.t Ine~urernent of the wall shear stresses (~xert('d by flu­ids on submerged moving surfaces. This tran~duc('r consists of a flexihl(· belt wrapped continuously and tightly over two partial cylinders which are mount.ed OIl frictionk"s HexlIl'('s. A portion of the 1){·lt rq>lac('s a small arc'a of the body O\'er which qlC fluid flows.

Due t.o it.~ d(·sign ~illlplicit.y, most of problt'llls associalc'c1 wit II dired flH'(lS\Ir<'/Ilc'lIt of ~;kill friction by I.he floating <'1<'1111'111 Imlam'('s arc' millilllize'c1 or eliminated.

Mt'flstlrcments have uecll made ill various wind t.lIIInels alld compal"-

- 12 -

isons have heen made with available data amI ot.hcr direct Ul<'ClSIJrcmenl and theoretical techniqlJ(':s. In general, excellent agreement is observed he­tween mea<;urements made wit.h the present balance and data available for appwxir&1ate flat plate £lows.

An optical displacement detection technique has bren incorporatcd into a protltype instrument successfully. This technique is very promisin~~ and can result in a significant increase in sensitivity and may eliminate the temperature effects. Optical techniques will also enable reductions in overall dimen"ions of the belt skin friction balance. Further work is recommcndcd to optimize the optical components to be incorporated inte the belt skin friction balance.

- 13-

.,.

, ~ ~ 1

~

.~

j "J .,

~,~ . " ,~

:1 f

1 J :J ,!

·1 :";

:! .. i

, '. . -. '.~ '!

. r

\ .

\ \

\

\ \

\

1. Kempf, Gunther, "Neue Ergebnisse der Widerstandsforschung," Hagen, 10: 234-253, 1929 .

2. Schultz-Grunow, F., "Neues Reibungs-Widerstandsge5etz fur Glatte Platten," Luftfahrtforschungj 17:230-246, 1940.

3. Dhawan, Satish, "Direct Measurement of Skin Friction," Ph.D. Disser­tation, California Instit~te of Technology, Pasadena, 1951.

4. Schut~s, W. E., W. H. Hartwing and J. E. \Veiler, "Turbulent Bound­ary Layer Skin Fdction Measurement. on Smooth, Thermally Insulated Flat Plate at Supersonic Speeds," DRL-364, The University of Texas, Austin, Tcx?..5, 1952 ..

5. Weiler, J. E. and W. H. Hartwing, "The Direct Determination of Local Skin Friction Coefficient," DRL-295" The University of Texas, Austin 1952.

6. Vak.ili, A.D., "Study of Direct Skir Friction I-t,feasurements and De­sign of Nozzle for Upper Transonic; ..vind Tunnel," M. S. Thesis, The University of Ten ,essee Space InstitLte, August, 1976.

7. Winter, Ie G., "An outline of the Techniques Available of the Measure-· mcnt of Skin Friction in Tllrbul~nt Boundal'y Layers," Prog. Aerospace Sci. 1977, Vol. 18, pp. 1-57.

8. Vakili, A. D. and .J. M. Wu, '~Direct Measurement of Skin Friclion with a New Instrument," Proceedings of the rut. Symp. of F!uief Cont rot and Measurem<.'nt, Tokyo 1985, paper no. F'I:~, FLUC.O~lE 85, P<.'rgamon Press, 1985. •

9. Vakili, A. O. and .1. ~l. Wu, "A New Instrument for Direct ~kasurt·­ment of Wail Shear Stress," Proc(!(·dings, 28th International (list rt III It'll­

. tation Symposium, rnstrument Society of American pl"Oceedin~s, Vol. 19, part one, pp. H7';'152, ~lay 19H2 .

. - 14 -

I f" / 1

t

r

l I I

I ... t

, .

... -

lO. Jackson, Charlie M. etal. "Description and Calibration of the Langl('y Unitary Plan Wind Tunnel," NASA TP- W05, 1931.

11. Murthy, A. V., C. B. Johnson, Eo J. Ray, and P. L. Lawing, "Re­cent Sidewall Boundary-Layer [Il\'estigatiolls with Suction in the Lan­gley O.3-m Transonic Cryogenic Tunnel," presented at the AL\A 20th Aerospace Sciences Meeting, Orlando, Florida, J,,:n 11-14, 1982.

12. Fiore, A. W. et.al., "Design and Calibration of the ARL Mach 3 High Reynolds Number Facility/' Tit 75-0012, 1975.

13. Allen, J. M. "Evaluation of Compressible-Flow Preston Tube Calihra­t!on," NASA TND - 7190, 1973.

14. \Vhite, Frank M., "Fluid Mechanics," McGraw Hill Book Company, 1974.

15. Lawing, P. L., Johnson, C. B., "Summary of Test Techniques Used in the NASA-Langley O.3-m Transonic Cryogenic Tunnel," presented at the AIAA 14th Aerodynamics Testing Conference. March 5-"{, 19H6, \Vest Palm Beach, Florida, paper No. 8G-0745-CP.

16. T( heng, P., "Development of a Servo Transducer for Din~ct Skin- Friction Mpasurement." Research Report prepared under Grant l"SC I Oi9 for Langley Research Center, NASA. School of Engim.'Cring, Old Dominion University, Norfolk, Virginia, 1979.

17. Van D['iest, E.R., "The Problem of Aerodynamic Heating," Aero. As­pects Session, Nat~onal Summer ~'lecting, JAS, Los Angeles, .Jlllle 19.16.

18. Hopkins, E .. J., [(eener, E. R., and Louie, P. T., "Direct Measllf(~II\('I\t.s of TurlHllent. Skill Friction Oil a. Nonadia.batic Flat Platt' at Mach ~1I111-ber 0.5 and C~mparison with Eight Theories," NASA TND-;)O;;), 1H7D.

- 15 -.

~ ... -'

"I I ,

\1

1

~ ~ I :1

1 -, I ~

\

'\ -J . , -.. \

I -~to:l ~'"

'~"'. "., _______ .. _._-~ ______ ,.---- __ ,. ______ .. _ .. _ ... __ ._. ___ -----------""""1

Belt

Flow c::==:>

Damping Gap

Damping Block

Figure 1; Belt Skin Friction Balance, Schematic Showing Principles of Operation.

..

I li , ,.

r f f

t i

All Dimensions Arc Inches

Fs1~ 0.53

-t A

~-B I

T 1.5 1.2

l~~~ ~L j.-l.48 ·-i

Section A - A ~-O.821

Section B - B

Sections of the Modified Skin Friction Transducer

, ! Figure 2. Dimensions and details of the compact version of the Belt Skin Friction Balance.

I ~ I

!

I I

I

'-

: I .

.'

/

I

,.' I r

. i I

) !

Figure 2. Continuded

f I

I t,

r ,

f i I l

i i I

;-,, __ , t '- f

. -- , . I

!

t r i

I I t

t /i , .

//.

i I

/t .. i

!

Flow

Bifuricnted Optical Fibers

n. Interface of Optical Fibers in. the instrument.

sH . Light

end

Light Detector ::=~=I~ ~

Gource end

Bifuricated Optical Fibers

Belt

b. Detail of the Optical Sensing technique.

a

Figure 3. Fiber Optics Sensing Hechanism for the UT",;I BSFB.

.' Ir .

-- ,.

"

I /" j

I

.<,

. t /1 . ~

I t f,

/

./

~ -~--- -- -... --

I

I I

'\ •

'\ ~. I

I " .1

L

71tl0.00 f..

i-

i-

f-f2l

i-

.1-

i-

I-500.0 0

400.0 Counts (mv)

0

ra

.200.0 m

10121.0 fZj

i-

l-

i-

I-

:-i-

f..

f-

-- . i-

i-

i-

... ~

J-

Skin Friction Trans~ucer ~ 1 Calibration

V-/

/ V

//

I/ 1. ;JQ)'

I I I , I I , t ,

2. 00 . 3. 00 ' t~eight (gm)

--

.LJ t , ! ,

4 .. 00 5.00

Figure 4. Typical Calibration of the BSFll.

. .

\ I j

I ..

• ,-<

/ /

......

~---....

," ,." - .... ~ .. -

Skin Friction Transducer g 2 . Calibration

700.00'~ ____ -r ______ ~~ __ ~ ____ ~~ ____ ~

to~nt!i (mv)

1.00

Figure 4. Continued

2.00 . 3.00 Ueight (gms.)

4.00 5.00

. _____ .~_ ... 1 .',._

-""-,

,I !

FigureS. Measurement configuration of the DTSI transducer on the Unitary Tunnel Wall.

~ ~I 'j

I ! i i i

.~ .. ~

',.; ....•

"

xl03

Cf

5

4

3

2

1 O.l:

"

H CII 2.16 CD

A UT£· Transducer ! o UTSI Transducer I D UTSI Transaucar II

Othzr s~olu are maaaureaenta eode by J. K. Allen ~~tb floating element balances (Ref.13)

ill b.

I.

Figure 6. Direct Hall Shear Stress Coafficient (incompressible) Measured in the NASA Langley Unitary Wind Tunnel.

~ " .

,,!

I ,. .,. 'j

\

~ .

·" :J

" .., .. , ... "

- ~~?..a.i1.~t1:'~i if '5···;'~~· ~; :oZ", \:'ZS::;:·;·~~~"":'~~~l~·!?' ~~;:;r.::&J.' ." ... -w~~, ,.I h~1~1/LLl:I.1U1 .. ~.~;.£;;;;;' i;.,·jtz:li~~!.~ ~" lot 6" .r: ¢·t :::sI'!.a... ..... ~e:::~WiJ.g:. .• :"¥t:::5'.:;:;;;:t~:a 1

. _ .. 1. i. I ..' g ... "t." ... N'" :::: I~::'::J:::: ,:':':' i :':::1 ::':' .T.:.·~ .~ -:- ::,:.=- I .... 1:.:..:~ T. - .-. i -. 8.... --- .-. - -----, _ ...... ___ . -l .. --t._.. ' , .. -... I - - .. ' I • -.. _... .... i. ...._ -' '!" ''''1 .. .... - ..... , ... - I···· , =.-= • :::':::'. '.' I·:': .. · '-... ':-: .. 1 _ ... i '.' : .. ::. " ". :,":-l=~:: !_'=~::' ~I :~::=J =~: : ::::=.:1::.::: i .::_. : =: ~::~! .. :" i -.~ J :; :'. '..-.: !. ":::~'F'~'~~ ~I :. 'L-_._~ __ .1.. __ ._ I L" I," •. -.- .. , __ .. ,- - ..... ----... Ii' -I'" . I = :=j:-_ r :C::~= ':' - i'--' .. - -+--- . ...... . --"-, . . =0 _......:-: .. '1"::: , : .:~'! u .: •• I 1--" ... __ ._1 _··;·--f- 1- ...... I··· . C"' • r -...... =- --r--"--' _ . .:.-. __ .j---== ... == _ .. ,,,._-. t-·--· ~~:=:.... .. ·--- .. t·-·-- t' .. --. ..- ... ___ I J I--"r-- '--' 1 -- _. - 1"-- _. "---" .- -_. . ... ,.. . .. _. __ ._--(._----_._- --_ .. _. _ ... _--- --....... - ... _ .. ---..... "-=I=:~:--=': . ='. ". .:p-rg.: I~ ""1ft • 11 x 106 .. '" - -_ ... -_ ... ' _u" - i·::':':· :c::: f'::: !.::

f.--- ~_._ - Q _. · .. ·1· _.--- . 4-f" .... c: -- -.-~.-- ... a 2427 PH - ----- .. _ .. -,. --=-=-:,~ . .:.-..... F'---' ---- 1--~. __ ._.... . __ . __ ml Cr.' -3 ~ ':'=-- .. -' --- -.. -:::~ .:.::.:-.. -_ .. -- . -.- :: =:..,-.l=='f:-.:::::t=j==t==l" R f 1.26 " 10 . g -_. ·--1-- -·:·::r·=·I=I·::·· , __ .... , __ ... _.!--==- --'=r . --. I" 1250 F . ~. '. -- --- ... =/ ... _., I

-- '.' It ... :--=+~ -: ·--:·~~~-=~·~·~,·~~~~T~-~ ~:::·:·I·~~:~ =r 1- . +---. ·J.r.-:-:-:l--·J· __ .,._.1-___ --I !'Ej 2-- --==-r I ,

- --~--f--' . ,..-,;::, "+:'~~:~~:'~~"'~""'~;':'!":-':~''''''''\/lr;;;:n-~~\~r~' - '---i--t--- - - -+ .. '·,;..fr·=i .... "".,.......vwl'"'r.\.-~'T"'"" ..... v ..... ...,~ -.~. .}.~ •. ~ --I-- I-- R 'f 6 = r- ,--- '--0 --r-~: .. \'"Y'.-.. -r-J:.... .. --.~ - !-- --- _.- '(;":1")~-' .''''- ..... - el t .. 7 X 10 _. -_._' -:. -'-" .. __ . ..&1:. -g -.. "f-~ -1·----· --,. Q"" ::a 1545 PSF -3 -. -- - " _ -- .--. -- ._-.- ----'. '.':'t>::.- 1-.- f • X - -'1=- .- ~~- - ·c"-" ..... -- f-._. C .. 1 31 10

~.~ .~ .-;;.~ ~~r 'I~" - _.-. --- .-. ~3-~"r~- It - 1250 F

~L-' - ~ t=§B- ~'-i~" ... _-,_.-.-.... '--'-.

-' L_ -~ ... -~ .-. ...: ---'--~-- -==t==--- ."""",~~~~,..--... ~ -.-. -1--- --" _ 0 .. +---1'-- ·f . -_. -- .. -. -- -6t-- ---t-_· .. ~ . _.- """;~~~·'P"l'f~~ === ..... __ .J -.'.-~ -+-_. ,-1--,'._- _ .. ____ . __ ~.~l:: ~~ - .. 1.- .. -_·t-·-··l·· .. · 1 .. - ·-·t· . - -i~ -l- , -_. '-- ----. ~~_-~- --}-_ .. -.-_. '---'1-'· .-.-~.~ ~ '-.~ . '--. - ~. -. -.. ~.--~-= ~ =--~~--,~-~I~-~~~\=~·i~~~:.:: _. --,'-- .... .._ ... _- ,--_., .. -'1-' .. __ . --.. _ .. r--' .... -- .1---r-·--- -. __ . - _. -', - -- 00 __ ' ..... I ... ::-.:.::== r==: .:::=",_:'-:: .. :===r-=:.':-,':.::::-r--.:-:=-= -=:: .. :='. ==. '='::'f=--::~ -.: :~'F':=. :-.:' .!'.-:-::-.' -:'. 1 .. :: '.

-' j--- --- ·----I·--";I-·-l.--_· r=::-i--· 1---'- I -'j'---'- j .. --- ·-·-!----t .. --.!-..... ~ ... ::.-::. ;-_ .. --.. t:-r·-=-:-+I-·-'--"1-- -_. -- . __ ...... __ ._1... •. _.. - - .. t· __ · --'1 _ ..... - - /.--- "1----' . -_._ .. -----1. _.- . - --I·-~-d-':'" . 1-- . '_-:-:-,'-:.--:= ~..::::' .'.-:" :~~'.~OO -:,:,::"1-:-:-:: :i·:-::-:-:_-' C-:'=-.- j'=:-:'!':'::-." '=:.:L.~' .::::':"'i.-:::-~ :~.:.-- !'::::.:: ~::_:-.:'. '_:'!.":;,u '-". , ...... : ..,. , , ~ ,,,. . I rr!;;' I - ~* J.in:c 0 ~;o

o;S ;;or-

Figure 7. Transip.nt Oscillograph Rccording of the BSFB with Variations in the Tunnel Operating Conditions.

.0." c:::» >0 r- fit ::i_ ..((1)

Cf x

::igurl" 8.

·.

10 8 6

4 10

3 3

2

a

'i I ~ , I

1 ? -

M = 0.7 cg

T = 300 K 0

T = 220 K . l:l

T= 160 K c

T = 100 I< 0

t ~ ~ <> ~

~ ~<> <>0 a

.t ' ...

M = 0.4 co

0

£1.

fj

~

I I I I I I I . I I r I' I , '-.lJ 3 5 10 20 30' 50 . . 100

-4 Rea x 10

~r.).ll She::.r Stress H~a.;u .. ements in the NASA Langley O.3-m Trar.sc :ic Cryogenic Tunnel at Different Total Temperetures.

~~ .:.' . !:

~ :" ~~~ "'1 .•

;.

l 'j ~r:J ~.11

i·~ ~~J

~ q "::~

.1

J rJ I~]

·s '.~ ... ,., :4 ;

.j

t~

fl -= '1

•

lli{~J:,:~;2:~~~::i~:~~iii~6·;'i·:;;;;~~~~::·\I"i1·~~;~~iXJ~y{/i..i-;1f:·.~~·,~:\;;;:~·.~,~:;:\~:S';;:·;;':~~~·;':.~~"~~~L~~~' i;:'/.J;,·~·':i'!j -:··~;·;·J;.;J;;.J~;';k;+AA4r.;::~~~~~,~:~~~~t;.;k;f'ta~;;"ki:;,'S).f* 'L~4·;/wU. " .. ,'- ... " , . ",,', J J. ". , "

I I ! I

I

..

. ~~ . ....... .... '", .. ~ ._ .. ~ ... ~"':-'-"-_'._' _ _ ..... , .. ,._.~~.~,. __ ,_._._., .. , .. _" ~ ~_.".ft._._ ....... 7-'·_ •• _. __ ..... _ ,_ .. , __ ,. __ •. ~ • ,'_ ....... --....~

M = 0.7 M = 0.4 00 co

T = 300 K 0 fj)

10 T=220K II A

8 T = 160 K 0 a 6 T = 100 K <> ~

3 4 Cf ~ 10 3 r ' A

2

I 1 2 3 5 10 20 30 50 100

-11 't Rea x 10

Figure 9. Wall Shear Stress measured on the O.3-m IeT side wall and comparison with predictions (Reference 14).

-, ,

' .. '-

." '.

, .

"I

", o o o ...

SKIN FRIr.TION BALANCE TEST UTSI &- NASA

1.150 ----r----r----r---1----'----,----, I I I I I I I I I I I I I I I I I I I I I I I I I I I I

l.flO ----~----~----~---~----~----~----~ 1 1 I _ Van Dried II 1 I I I I + UiSI Dolence I I I I I II flASh Delanet I I I I I I I

1 .... 0 ____ L ____ L ____ L ___ ~ ____ J ____ J ____ J

5% I I I I I I I I I , I I I I I I I I I" I 1

)( 1.30

+1: :': :: : : ----r----r----r---~----,----,----,

.... I 1 I I 1 I t o I I , I t I I

I I I II I 1 I 1 I I I i I

1.20 L ____ L ____ L ___ ~ ____ ~ ____ ~ ____ J

1 I I I I I I I 1 1 1 II ~ Iii I I ~1 1 r 1 1 I

1.10 -tl I I I 1 I' ----r---'----,----'-~--l

I 1 I

I I I~ C I I

I I 1 1+ I ht I ~I" --~--:-----:---.--:-----i-~-- 5~

I t I I I i---- I

I.CO

1 I 1 I I 1 + I I 1 I 1 1 I I

O.GO +---I-----t---4--...,-I-----+-1 ·---II----4~ 2.00 4.00 6.00 13.00 10.0r 12.00 14.00 16.00

Rc theta x 10u-4

Figure 10. Direct Shear Stress Measurements in the ARL Ha'ch 3 Tunnel.

-.,

., f

..... - '

,.

SKIN FRICTION BALANCE TEST - UTSI & NASA

g-3 IIRNF !J OCT 85

I.ISO ----r----r----r---'----,----'~---1 I I I I I I I I I . I I . ..

I I I 1 1 I I 1.130 ----~----~----~-~-~----~----~----~

I I I - Van Drielll II I I I I I + UTSI Baiancil I I I I I l! NASA Sotonee I I I I I I I

1.40 ____ L ____ L ____ L ___ ~ ____ J ____ J_~ __ J

I "1 I I I I I I I I I , I I

X 1.30 :r I I I I I I I i1E 5% I I I I' I I I

---~----~----~---~----~--~-~----~ ~.

o I I I I I I I I I I I I I r I , ,., I , I r I I I I I I I

1.20 L ____ L ____ ~ ___ ~ ____ ~ ____ ~ ____ J

I I I I I I I II I ·1 I I I I I I I I r~ I 1 1 1 1 II- I I I I I

1.10 r----r---'----,----'--~-l·

I 1.00 ----~----~----~---. I I .

I I I . I I . I

o.ao +-- --!!---+I-----tli-----t----t----i-----t 2.00 4.00 G.OO 0.00 10.00 . 12.00 14.00 16.00

Re theta x 1 O$~-4

Figure 10. Continued

;. -:,-.: ... -.-::_-~---. ~~-'"'r'..:::_:;7:"'":~...,......"&,;o. ... __ ..u~-_. ______ '>-. __ ~ ____________ •• _~ __

.1 ,{ol /~~

/

1. i

.. i···

~ . I

SKIN FRICTION BALANCE TEST - UTS/& NASA

1.30

11"'3 HtlllF' 'OCT OS

-------,--------T--------r~------l I I I 1 I

• 1.25 -------,--------T--------r-------,

I') 0 • 0 ,.-

)C

... 0

1.20

1.10

I . I .,

1 , '. .

- Vall Drlll.St II + UTst BoraneG II HAS.'!. Dcl(tnc:~

I I I I

• • I .

-------~--------~--------~-------~ I I I , I· I I I I , r1 I·' I .. !l I .. ~ I . " . I ~ ., '.. I ~ I

--------------------------~-------I . I I I I I 5% I I I I I r I 1 .2 + I I '+ "' I _______ ~t----~--~ ~ ______ ~-------~ I __ ..ld- " ·1 : I .. r . I .. C:' 1 t I I I • . .1 .1 t I I I I

1.0:5 - - - ... - - - -1- - - - - - - - .. 1- - - - - - - - -I ..... - - - - - - - .... I I I I I I I I I I 1 I I I I I I I I I

1.00 l- I I -~ 3.40 3.45 3.50 3.55 3.GO

(T edge/T wall) x 1 o~; 1

Figure 11. l~all Shear Stress Heasurements and Comparison tdth theory in presence of Heat Transfer.

___ ••• _ . _______ ~ _______ • ___________ .. _____ ~ .. ____ "'_,-.""'.: ____ ~ .. _-..-.~ .• : ........ r"': •. _"".I't .. ~ __

I L n

11 I

" ! "~ I'

i; 1·\

.j

I:

f'!--"

i I

I I t j

.-------1

" i

../. ------

--'

" .- :.-

--.. "i

,.-

j

... : -~

-.! .

..

ORIGINAL PAOE (S OF POOR QUAUTV

.....

! -

---------'

~"!=;"7.~3- ~:~~~~~~.:=E=--3~~~:::;:~~~~=

•• ---- - ___ ::::::;ot;A;.}J ~f~~J:=:--~;. _~ .... = _____ ._... ___ ... :r. ..... _e::::t: . f_ 1. . ----..

---.. -.--.. =; \ NASA Balance ----.-.. - .... .. ... .~-""".r-en-' - """"" __

---_.- .. _-------_ .. --

-. Time

Figure 12. Oscillograph trace of the UTSI BSBF, NASA Blllance, and an accelerometer in th~ ARL 1-Ia Tunnel, during the Tunnel starting time.

,.

., I:. I 1 " I ~ ',:

~'" r-:~ / <~

· End of Document