TemperaturemeasurementbyTLC

while the TLC layer was firstly coated then follows by the black panting. This configuration is actually depending on the camera position. In present study, configuration A was adopted since the temperature changes on the endwall was captured from the tip side of test section.

This means the camera will be positioned from the top side of test section. Thus, the TLC layer must be coated on black painted acrylic plate in order to keep the visibility during the

measurement. Figure 28 illustrates how the measurement was carried out, including camera and light positions. The color change of the TLC was recorded with a digital video camera,

and recorded image data were captured by PC frame by frame, then converted from RGB

images into HSL (HuelSaturationlLightness) images using a software

(GraphicsConverter)(Hachiya, 2004). The accuracy of the measurement techniques was

based on Funazaki [30].

.

- O.63Cza

Il

inkt• 1

=>• i II .1

Leakage slot.

1

•L

.

- el- 1

-y

Lx

.

smput

lealcage slot Flexible e zz.al1

sectiQn-ELts

1 I l 1 1

.

.KK

.

Main in'

.

't

SS]k.

O.24C.

O.63Cvt

o.064C..

t

SecotKlary flow injection oo'

Lx

z

Figure 26 Temperature measurement region

Conrrgumtion 1

-0

/1

CCD

camera

1 LC laver Black paint layer Acrylic plate

Confrgure tion 2

Si(I?N.

CCD

camera

Acrylic plate

nc laver

Black paint layer

Figure 27

lighting

gÅé)...

NN

TLC coating configurations

CCD camera

lighting '/0i

, t

epe

1

Measurement area

Upstrearnleakage air

-, ' , ,

KAypc thennocouple

Dataloggcr

Menum chamber

Segmcnt lealcage air

-Figure 28

Persorul Compater

Temperature measurement system

2i6.3 TLCcalibration

The relationship between the temperature and measured Hue of the reflected light defines the calibration curve for the TLC. Figure 29 shows the calibration device that was used where the stainless sheet was attached on the acrylic plate with two cooper electrodes were fixed at the end. The TLC was coated onto the stainless sheet surface by configuration A to have a similar condition with the measurement. The voltage was applied through the electrodes and the color changes due the temperature different on the calibration plate were captured by the same camera used in the measurement. At the same time, the temperature distribution was measured by sixteen K-type thermocouples which were attached on the bottom side of the plate. In order to reduce the uncertainties, the calibration test was

conducted in place in the wind tunnel with the same lighting level and viewing angle used

during the data acquisition phase of the measurement as shown in Figure 30. The RGB

images captured during the calibration test which was converted to HSL images is shown in Figure 3 1 . Then, Eq. 1 1 was used to obtain the relationship between temperature and Hue

Ts Ps

-TL Å~(p- p, )+ T,

-PL

(11)

where,

P, Hue value at specified pixel position;

Ps, Pixel position ofthe low-temperature side thermocouple;

.PL, Pixel position ofthe high-temperature side thermocouple;

Ts, low side temperature measured at Ps; and TL, high side temperature measured at Pi..

Finally, the curve can be plotted as shown in range of 30-170 illustrating the most accurate post-processmg purpose.

Figure 32. Based on the Figure 32, the Hue characteristic has been selected for the data

Electrode 1

Electrode 2

16 positions of

thennocouples -,.

. ...,.. ,.,tr w""/ f"\s,""i{'i'L`nt,'i•e•,,. ,:.,'111/I.' ...i{li',;;"i

...,.i.,.,':•.••g•"" ."'l"dii•i'i"'i'ii'li'illlii'/'i'/iilil'l;:11/t-'lil,II/,, ' .•'"i''ec'i:'ii'ii'i'

•pt• •.. •

.

•"•-... ,g

J'it

e .."

- .///t./"Il ..l/.'t,t't'' .r /l.t.,lt,si. ... /t"

Acrylic plate Stainless sheet

(a) Schematic view TLC calibration plate

Electrodecoer

16-2VV ThermocouleK-e

-8'

9

8

^•15

-

-10

300

¹⁰

483 20

StainlessFoil

Under-coatandLiuidcstal

(b) TLC calibration plate with dimension

Figure29 TLCcalibrationplate

ade

b.

•Xk-g•' 's'

e

xS

•

(a) Actual measurement

(b) Calibration test

Figure 30 Position of calibration plate during test

(a)

RGB lmage

(b)

Hue image

Figure31 RGB lma

ge (a) converted to HUE

lma

^{ge (b)}

- v

L 9

e N

8.

E o

-3LO

30.0 29.0 28.0 27.0 26.0 25.0 24.0 23.0

""" i'-[i]"Maxtcmpetture: : : , ,

tl-)lil-t +ii-l-itl ldlltldtt

--•---

--li-1-i

,

---L r----=--r---T--J-i--r---T---1---"1---"---Jl---n---+----r---T-r---r---T---l

-""' l"'6I"A'i'e'ia'g'ed'fÅë"m"i5ett"-'e-'l-""'-l--"'"'l"-"--1"--'"'-;-L""'f"'---':"---'-l

III-dl tt

, i i i i i

...L..

i l I ,

,

i I i l

:

i

:

,

i i i l

,

i i

--.`...--..-...-...-.--..=-..--...--..-.

i

...i.--.-..A. pprp.!mation curve

l::

...i...J..--+-.4.--....J..++L-

t1-111

l-lt

^:

il ll

i-...L.--....L...i...i-..-..-J..

i ,

i

'i+

l/

i l ,

:

--i

:

l

,

l...J-..--.+J."--`...L..-....L..-..-.L-...,

.--... t....-..i---....J.-...`.j...-...J....-.=.i..`...L-...-..L--.---.L...i

:l

l Referenceirangq

l-i i i i i

l i i i i i

,

i i

, , d t , 1

i l i i l l

.L-...-l-..-.-..L...-.L-..--.J

l l

,

i

bl

-1 -i -1 ll ld ll dl dl -t d- lt il -- it tl li il t+ iL -l -- -l ll +I ll +i -i +l t- +1 bd

.-.---.----d

il 11 11 -4 11 -1 il -1 tl ip il l- i4 l- Ii li li ]1 +i ll ti ll dl li -1 dl li dl tl 11 -1 -d 11 -1 -4

co

-oe ts

E g e E

9 B

L

o

20 40 ⁶⁰

⁸⁰

Figure 32

approximated

100 120 140 l60 180 200 220 Hue

TLC calibration curve

Actual temperature rise

:

ll

t II t lt t lt 1 lt

l l l ITime

240 260

l t 1 t

tl 11 11

1 1 1 1 1 1 ' ' l t s

rj

Figure 33

Elapsed time (t)

Temperature rise curve by step series oftemperature gradual

2.6.4 Transientmethod

This study employed a transient method to determine heat transfer coefficient and film effectiveness from the time-varying temperature data of the surface. A brief description on the method is given in the following. Temporal change of the surface temperature on semi-infinite body T., subjected to step-like temperature change of the flow over the body with constant heat transfer coefficient, h is provided from the solution of one-dimensional heat conduction equation as

T,,•

Tg

i; -i-exp( ft2S )eofc( k" )

(12)

Where,

T, , initial temperature ofthe body;

Tg, flow temperature;

p, density;

c, specific heat; and Z, thermal conductivity.

In a real situation, since it is almost impossible to obtain a step-like temperature rise of the

fiow, as shown in Figure 33, Duhamel's theory can be applied to cope with a gradual

temperature rise. In this case the temperature rise is approximated by a series of steps with small temperature increase, which yields the expression for the time-varying wall temperature as

N

T. (t)- 7J]' =2U(t-Tj XTg..f-Tg..f-i) (13)

.f--1 Here,

u(t-T,)=i-exp(h2Sti-,zT')ierfch,V/7-:l-]}isJi (i3i)

When

applying the above-mentioned relationship to film cooling situation, the flow

temperature Tg in Eq. 13 should be replaced by adiabatic wall temperature, T..•. Since film cooling effectiveness q is defined as,

Taw - Tom

where T. and T2 are main and secondary flow temperatures. The adiabatic wall temperature can be written by

Taw=rpT2+(1-rp)Too (15)

Suppose that the film effectiveness remains constant even when the temperature rise ofthe secondary flow is approximated by a series of step-like temperature change, the following expression can be used for Ta"•

Ta".i =nyT2.J +(1 -O)Tco (16)

Therefore, Eq. (13) can be rewritten by substituting Eq. (17) into Tg,.i

?Vr

Tw (t)- Ti ='72U(t-TJ )t2u -T2uTi) (1 7)

.1'=1

Eq. 17 can be regarded as a non-linear equation with respect to two variables, i.e., effectiveness, ij and heat transfer coefficient h. Combination of two different temperatures for two different elapsed times t. and xb given by Eq. 17 yield

T. (t. )- T, 2i.i U(t" -li XT2i- T2 Jmi)

T,, (t,)- Ti :ii] u(t, - l, XT,,.f - T2,,•-i ) ,1'=1

film wall

(18)

This is an equation only with respect to h, from which h can be determined by solving Eq. 18 numerically. Then, film effectiveness can be given as follows,

Tw (ta )- 7]

2U(ta - ly' XT2.,i' - T2,f-i

)

J'=1

However, there are required several procedures before the h and n can be properly calculated by this method. Figure 34 illustrates the procedures ofoverall data post processing.

TemperatureA•Ieasurement TLC calib ra tion

Imagecapture Imagecapture

ConversionfromRGBformattoHue ConversionfiromRGBformattoHue

ConversionfromHuetotextformat ConversionfromHuetotextformat

Inputoftimerecord

Confumationofthepixel

ositionofHue

ConversionfromHuetotcmperature

Calculationefheattrartsfercoefficientand filmceolineffectiveness

Post-processingofresults

Figure 34 Data post processing procedures

ドキュメント内 Studies on Aero-Thermal Performances of Leakage Flows Injection from the Endwall Slot in Linear Cascade of High-Pressure Turbine (ページ 52-64)