STUDY ON AIR INGRESS PROCESSES DURING A DEPRESSURIZATION ACCIDENT OF - - PowerPoint PPT Presentation

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

STUDY ON AIR INGRESS PROCESSES DURING A DEPRESSURIZATION ACCIDENT OF VHTR

International Conference on High Temperature Reactor Technology HTR 2018, October 8-10, 2018, Warsaw, Poland

Tetsuaki TAKEDA

Graduate School of Engineering. Dept. of Mechanical Engineering University of Yamanashi

Contents

• 1. Introduction
• 2. Experimental apparatus and numerical model
• 3. Experimental and numerical results and discussion
• 4. Conclusions

National university corporation UNIVERSITY OF YAMANASHI

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Back ground and objective

1. Pipe rupture at connecting pipe between RPV and gas turbine. 2. Helium gas blows off from the RPV. 3. Pressure in the reactor equalized to the one in the containment or confinement vessel. 4. Buoyancy force produce by the temperature difference between inside and outside passage in the RPV. 5. Natural circulation of air will produce. (depend on temperature profile or geometrical condition) 6. Graphite of reactor component will react with ingress air.

Schematic diagram of GTHTR300C Designed by JAEA 2

Air ingress scenario in the case of the horizontal pipe break The objective of this study are to research gas mixing process and to develop the prevention technology of air ingress. It is necessary to prevent air Ingress or oxidation of graphite at pipe rupture accident of Very High Temperature Reactor. Even if the pipe rupture accident occurs, ingress of air can prevent by injecting helium gas.

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Back ground

It is necessary to prevent air Ingress or

• xidation of graphite at pipe rupture

accident of High Temperature Reactor

• 1. Pipe rupture at connecting pipe between

RPV and gas turbine.

• 2. Helium gas blows off from RPV.
• 3. Pressure in the reactor equalized to the one

in the containment or confinement vessel.

• 4. Buoyancy force produce by the temperature

difference between inside and outside passage in the RPV.

• 5. Natural circulation of air will produce.

(depend on temperature profile or geometrical condition)

• 6. Graphite of reactor component will react

with ingress air. Schematic diagram of GTHTR300C Designed by JAEA

3

Air Ingress Scenario in the case of the horizontal pipe break

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Schematic drawing of the HTTR and model of coolant passages

• A hot leg consists of an inner passage of a coaxial duct, a high-temperature outlet duct, a high-

temperature plenum and fuel cooling channels.

• A cold leg consists of an annular passage of the coaxial duct, a bottom cover and an annular passage

between the reactor pressure vessel and permanent reflector.

• As the hot and cold legs are connected at the top space, they make a kind of reverse U-shaped tube.

4

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Schematic drawing of the GTHTR300C and model of coolant passages

5

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

The experiment has been carried

• ut

to research mixing process of two component gases and onset time of natural circulation of air. When one side wall is heated and the other side wall is cooled in a vertical slot, a localized natural convection is generated. Heavy gas will diffuse into the both vertical slots at the same time, and then time elapsed, natural circulation through the passage will be generated finally.

Influence that localized natural convection in the vertical channels with different temperature exerts on onset time of natural circulation Purpose of this study

Molecular diffusion Light gas Heavy gas

To investigate an

• nset

time

• f

natural circulation and a mixing process

• f

two component gases by using 3D numerical analysis.

The flow regime of this localized natural convection is ranging from conduction regime to boundary layer regime.

6

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

7

• The apparatus consists of two vertical slots,

connecting passage and storage tank.

• Vertical slots and storage tank were

separated by partition plate.

• The left side vertical slot consists of a heated

wall and a cooled wall.

• The right side vertical slot consists of two

cooled walls.

Experimental apparatus

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Experimental procedure

• 1. Heavy gas is filled with apparatus.
• 2. Partition plate close.
• 3. Light gas inject from top of the

apparatus.

• 4. Two vertical wall of the left hand

side slot is heated and cooled.

• 5. When the steady state was

established, the partition plate

• pen.
• 6. Experiment starts.

Heavy gas Light gas

8

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Analytical domain X : 548, Y : 398, Z : 846 mm Analytical grid x: 122, y: 20, z: 118 (total cells : 266680) Boundary condition The outside of the heated wall and the cooled wall assumed an adiabatic wall. The

• ther

walls assumed natural convection heat transfer boundary condition. Others Standard density : ρ0  Buoyancy : (ρ-ρ0)gV

PHOENICS: three-dimensional CFD code

9

Numerical model

X Z

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Analytical domain

X : 548, Y : 398, Z : 846 mm

Analytical grid x: 122, y: 20, z: 118 (total cells : 266680) Boundary condition

The outside of the heated wall and the cooled wall assumed an adiabatic wall. The

• ther

walls assumed natural convection heat transfer boundary condition.

Others

Standard density : ρ0 →Buoyancy : (ρ-ρ0)gV

10

Numerical model

X Z

PHOENICS：three‐dimensional CFD code

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Numerical method:

• 1. Steady state calculation:

Natural convection was generated by temperature difference between the vertical walls.

• 2. Unsteady state calculation:

Partition plate was opened at 0 sec. Calculation step ：0.01sec/step・・・ 0~10sec 0.05sec/step ・・・ 10sec~ Initial condition of steady state: Heavier and lighter gases were filled. The vertical walls of the left side slot was heated and cooled. Initial condition of unsteady state: The result obtained by steady state calculation.

Numerical method

11 Gases Density [kg/m 3] (20ºC, 1atm) Helium (He) 0.164 Neon (Ne) 0.838 Nitrogen (N2) 1.17 Argon (Ar) 1.64 Two component gases (light-heavy) Diffusion coefficient [cm2/s] N2-Ar 0.20 Ne-Ar 0.32 He-N2 0.68 He-Ar 0.73

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Temperature difference between wall [K] 30，50，70，100

12

Two component gases and temperature difference

Two component gases (light-heavy) Diffusion coefficient [cm2/s] N2-Ar 0.20 Ne-Ar 0.32 He-N2 0.68 He-Ar 0.73 Gases Density [kg/m 3] (20ºC, 1atm) Helium (He) 0.164 Neon (Ne) 0.838 Nitrogen (N2) 1.17 Argon (Ar) 1.64

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Numerical result (change of gas temperature)

(9) (8) 50 50 90 75 (7) (6) 50 50 (5) (4) 50 50 (3) (2) 50 50 (1) (12) 200 (11) 200 (10)

Z

(He-Ar, ΔT=100K)

Z[mm] Z[mm]

13

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Z[mm] Z[mm]

(He-Ar, ΔT=100K)

14

Numerical result (change of gas temperature)

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Z[mm]

(He-Ar, ΔT=100K)

15 X

Numerical result (change of velocity)

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Comparison between experiment and numerical analysis

Z[mm] Z[mm] Z[mm] Z[mm]

experiment numerical analysis 15℃ 5℃

(He-Ar, ΔT=100K)

16

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

(a)30min (b)52min (d)54min Localized natural convection generated in the left side slot. Just before natural circulation generated. Natural circulation generated through the passage.

Velocity [m/s]

(He-Ar, ΔT=100K)

17

Numerical result (distribution of gas velocity)

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

(a)30min (b)52min (d)54min Mixing by molecular diffusion was promoted by the localized natural convection in the left side slot. Natural circulation generated through the passage and mole fraction became uniform.

Mole fraction(Ar)

(He-Ar, ΔT=100K)

18

Numerical result (distribution of mole fraction)

Just before natural circulation generated.

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

ΔT [K] He/Ar [min] He/N₂ [min] Ne/Ar [min] N₂/Ar [min] 30 90 105 140 180 50 75 76 130 140 70 60 72 100 90 100 55 60 85 70 ΔT [K] He/Ar [min] He/N₂ [min] Ne/Ar [min] N₂/Ar [min] 30 84 103 128 164 50 76 80 115 130 70 65 77 109 97 100 52 63 93 76 ΔT [K] He/Ar [%] He/N₂[%] Ne/Ar [%] N₂/Ar [%] 30

• 6.67
• 2.22
• 8.33
• 8.98

50 1.78 5.26

• 11.5
• 6.90

70 8.33 6.48 8.50 7.59 100 3.64 4.44 9.41 8.57 Numerical analysis Experiment Difference between experiment and numerical analysis －：faster than experiment

① ① ② ② ③ ③ ④ ④

19

Comparison between experiment and numerical analysis Onset time of natural circulation

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Gr number （in the left slot） ΔT [K] He/Ar (×10⁴) He/N₂ (×10⁴) Ne/Ar (×10⁴) N₂/Ar (×10⁴) 30 0.045 ～ 0.95 0.055 ～ 0.84 0.44 ～ 2.1 2.7 ～ 3.2 50 0.068 ～ 1.4 0.076 ～ 1.3 0.75 ～ 2.7 4.1 ～ 5.0 70 0.079 ～ 1.7 0.090 ～ 1.5 0.84 ～ 2.9 4.8 ～ 6.0 100 0.10 ～ 2.2 0.11 ～ 1.8 1.2 ～ 4.4 6.8 ～ 11 ΔT [K] He/Ar [min] He/N₂ [min] Ne/Ar [min] N₂/Ar [min] 30 84 103 128 164 50 76 80 115 130 70 65 77 109 97 100 57 63 93 76 Generation time （analysis）

• Onset time of natural circulation became short with increasing

temperature difference. Onset time of natural circulation became short because Gr number increased. Natural convection became strong and mixing of gases was promoted.

Density ratio 1/10 1.4/10 7/10 5/10

20

Relationship between onset time of natural circulation and Gr number

• Gr number increases with increasing temperature difference.

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

ΔT [K] He/Ar [min] He/N₂ [min] Ne/Ar [min] N₂/Ar [min] 30 84 103 128 164 50 76 80 115 130 70 65 77 109 97 100 57 63 93 76

Generation time （analysis） Diffusion coefficient

ΔT [K] He/Ar [cm/s²] He/N₂ [cm/s²] Ne/Ar [cm/s²] N₂/Ar [cm/s²] 30 0.742 0.678 0.325 0.205 50 0.779 0.712 0.326 0.211 70 0.833 0.763 0.355 0.228 100 0.888 0.824 0.374 0.234

21

① ② ① ① ① ② ② ② ③ ③ ④ ④ ④ ④ ③ ③

Relationship between onset time and diffusion coefficient

• When temperature difference was 30 and 50 K, onset time became

short with increasing diffusion coefficient.

• Onset time of N₂/Ar and that of Ne/Ar were reversed at 70 and 100 K.

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

ΔT [K] He/Ar [min] He/N₂ [min] Ne/Ar [min] N₂/Ar [min] 30 84 103 128 164 50 76 80 115 130 70 65 77 109 97 100 57 63 93 76

Generation time （analysis） Diffusion coefficient

ΔT [K] He/Ar [cm/s²] He/N₂ [cm/s²] Ne/Ar [cm/s²] N₂/Ar [cm/s²] 30 0.742 0.678 0.325 0.205 50 0.779 0.712 0.326 0.211 70 0.833 0.763 0.355 0.228 100 0.888 0.824 0.374 0.234

This is because Gr number of N₂/Ar was larger than that of Ne/Ar, so natural convection became strong.

ΔT [K] Ne/Ar (×10⁴) N₂/Ar (×10⁴) 30 0.44 ～ 2.1 2.7 ～ 3.2 50 0.75 ～ 2.7 4.1 ～ 5.0 70 0.84 ～ 2.9 4.8 ～ 6.0 100 1.2 ～ 4.4 6.8 ～ 11

Gr number

When temperature difference is small, onset time depended mainly on diffusion coefficient. When temperature difference is large , onset time depended not only

• n diffusion coefficient but also on localized natural convection.

① ② ① ① ① ② ② ② ③ ③ ④ ④ ④ ④ ③ ③

Relationship between onset time and diffusion coefficient

• When temperature difference was 30 and 50 K, onset time became
• Onset time of N₂/Ar and that of Ne/Ar were reversed at 70 and 100 K.

short with increasing diffusion coefficient.

22

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

60 120 180 240 300 360 50 100 150 Onset time of natural circulation [min] Heated side wall temperature [C]

Nitrogen/Argon 7/10 Neon/Argon 5/10 Helium/Nitrogen 1.5/10 Helium/Argon 1/10

Influence that combination of two component mixed gas exerts on onset time of natural circulation

Figure shows the onset time of natural circulation against of the wall temperature

• f the high-temperature side passage.

23

Gr number increase Diffusion coefficient increase

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

3 parallel vertical channels having different temperature

24

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Influence that temperature difference of 3 parallel vertical channels exerts

• n onset time of natural circulation

120 180 240 300 360 100 150 200 250 300 350 Onset time of natural circulation [min] Average temperature of the pipe wall [C]

• 3ch. same temperature
• 2ch. same and 1ch. low temperature
• 2ch. same and 1ch. high temperature
• 3ch. different temperature

162-162-162 200-199-209 243-242-243 281-279-280 284-283-283 323-321-322 164-115-121 164-162-108 164-162-85 205-130-135 205-133-131 245-145-142 204-202-132 244-167-167 202-200-167 245-241-118 243-202-169 244-241-170 285-174-174 284-181-213 283-207-216 284-281-133 324-211-209 281-280-211 323-243-210 246-319-245 322-281-248 319-317-250 282-240-209

25

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Influence that graphite oxidation on the reverse U-shaped vertical channel exerts on onset time of natural circulation

60 120 180 240 300 400 500 600 700 800 900 Onset time of natural circulation [min] Average temperature [C] Helium/Air with oxidation Helium/Nitrogen without oxidation 26

Onset time of natural circulation was not so much affected with graphite oxidation.

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Onset time of natural circulation against wall temperature of the high-temperature side

• f the reverse U-shaped passages (height of heated part is less than 1m)

1 10 100 1000 1 10 100 1000 Onset time of natural circulation [min] Average temperature of the channel wall [C]

Helium/Air with oxidation Helium/Nitrogen without oxidation Nitrogen/Argon 7/10 Neon/Argon 5/10 Helium/Nitrogen 1.5/10 Helium/Argon 1/10

• 3ch. same temperature
• 2ch. same and 1ch. low temperature
• 2ch. same and 1ch. high temperature
• 3ch. different temperature

Helium/Nitrogen 1.5/10 Helium/Air 1.3/10 Nitrogen/Argon 7/10

Figure shows the onset time of natural circulation obtained by 3 apparatus. Three apparatus are the reverse U-shaped tube, three parallel channels, and vertical parallel walls. The height of the heated part is less than 1m. When not only the localized natural convection but also natural circulation was generated, the onset time of natural circulation becomes short.

27

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Experimental apparatus simulated HTTR

Temperature difference Length of vertical passage

28

The onset time of natural circulation becomes long when the phenomenon is governed mainly by molecular diffusion. When natural circulation was generated, the onset time of natural circulation becomes short.

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Influence that temperature difference in the vertical graphite passage exerts on

• nset time of natural circulation

48 96 144 192 240 288 300 400 500 600 700 800 900 1000 Onset time of natural circulation [h] Average temperature of channel wall [C] Helium/Air with oxidation Different temperature 71.4 Different temperature 74.6 Different temperature 81.3 Different temperature 80.6 Different temperature 84.7 29

When natural circulation was generated, the

• nset time of natural circulation becomes short.

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Influence that vertical length of the passage under the core exerts on

• nset time of natural circulation

1 10 100 1000 10000 100000 1 10 100 1000 Onset time of natural circulation [min] Average temperature of the channel wall [C]

Helium/Air with oxidation Different temperature Helium/Air with oxidation Helium/Nitrogen without oxidation Nitrogen/Argon 7/10 Neon/Argon 5/10 Helium/Nitrogen 1.5/10 Helium/Argon 1/10

• 3ch. same temperature
• 2ch. same and 1ch. low temperature
• 2ch. same and 1ch. high temperature
• 3ch. different temperature

Helium/Nitrogen 1.5/10 Helium/Air 1.3/10 Nitrogen/Argon 7/10

30

The onset time of natural circulation becomes long when the phenomenon is governed mainly by molecular diffusion.

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Proposed system in the case of GTHTR300 (JAEA’s idea)

• Ref. Yan, X. L. et al. (2008). "A study of air ingress and its prevention in HTGR, " Nucl.

Technology, 163, No.3, pp.401‐415.

31

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Experimental procedure (He injecting)

1. Heavy gas is filled with apparatus. 2. Two vertical wall of the left hand side slot is heated and cooled. 3. Natural circulation flow are

• produced. When the steady state

is established, helium gas injects from the top of the apparatus. 4. Experiment starts. 5. Natural circulation will be stopped. 6. After the time elapsed, natural circulation will be reproduced suddenly. He Heated wall Cooled wall Air

In order to investigate of preventing natural circulation flow by injecting helium gas, an experiment has been done as follows.

32

Helium gas

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

5 10 15 20 25 30 35 40 45 2 3 4 5 6 7 8 9 10 11 Onset time of natural circulation [min] Helium gas injection volume [%] ⊿30K ⊿50K ⊿70K ⊿100K

Experimental results regarding re-onset of NC

Temperature difference decrease increase

Re-onset time of natural circulation increased with increasing the injection volume of He

In order to prevent natural circulation flow, the amount of injecting He increased with increasing temperature difference

33

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

1800 3600 5400 7200 9000 2 4 6 8 10 12 14 Stop time of natural circulation [s] Volume rate [%]

40K 60K 80K

Relationship between elapsed time and injection volume

Temperature difference

34

Helium gas

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Air Ingress Scenario in the case of the horizontal pipe break

• As the buoyancy force will be

small, the natural circulation flow will not produce under the condition of this density distribution.

• Thus, air will transport to the

core by mainly molecular diffusion.

• If the localized natural

convection occur inside the channel, it is difficult to estimate not only the density change of gas mixture but also the onset time of natural circulation through the reactor.

• After the time elapses, the

natural circulation may occur suddenly.

• After the pipe ruptures, air will flow into the bottom part of the RPV by the counter-current

flow.

• The density stratified fluid layer will be formed.
• Buoyancy force will produce between the hot and cold legs.

Vessel Heater

Air He

He Air

He Air

Diffusion Natural Convection

35

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

36

h₁ d₁ h₂ d₂ v₂ Sectional view of GTHTR300

Scaling of GTHTR300 and experimental apparatus GTHTR300/Apparatus Aspect ratio of pressure vessel Height/Outer diameter [H1/D1] Aspect ratio of outer cylinder Height/Outer diameter [h1/d1]

1.86 / 1.53

Aspect ratio of core Height/Outer diameter [H2/D2] Aspect ratio of Inner cylinder Height/Outer diameter [h2/d2]

1.50 / 1.43

Diameter ratio of primary double coaxial pipe [D3/D4] Diameter ratio of horizontal double coaxial pipe [d3/d4]

1.30 / 1.50

Volume ratio of pressure vessel & core [V1/V2] Volume ratio of outer cylinder & inner cylinder [v1/v2]

5.00 / 4.60

H₁ D₁ H₂ D₂ V₁ V₂ Experimental apparatus d₃ d₄ D₃ D₄ v₁

Experimental apparatus simulating GTHTR300 series

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

37 139.8 139.8mm 190m 190mm 283.4 283.4mm 350m 350mm 410m 410mm Co Coolin ling water water inle inlet Heat insulation Inner nner cylind cylinder Co Coolin ling water water outle

• utlet

Gas Gas sa samplin ling and and suppl pply port port Cartri Cartridg dge hea heater er Outer Outer cylind cylinder (W (Water cool cooling ing jac jacket en enter tered) Outer Outer pipe pipe of

• f horiz

horizontal do doubl uble pipe pipe Inner pipe of horizontal double pipe Ball Ball valv alve Thermocoupl uple 255m 255mm 100m 100mm Fl Flow me meter

Experimental apparatus

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Results of temperature profiles in the apparatus

38 Temperature difference is 15 to 55 K Approach to steady state

(38) (36) (32) (35) (31)

(31) (35) (38) (32) (36)

420 mm 100 100 mm mm

(32) (31) (38) (35) (36)

Valve opened

x z x y

Temperature fluctuated from 2 to 10 K

Localized natural convection will be generated at the top of inner cylinder

Heat input: 324 [W] (10995 [W/m²])

Temperature fluctuated from 1 to 2 K

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Results of molar fraction change in the apparatus

39

(1) (4) (8) (11) (15) (16)

He Air He Air

x z

Heat input: 324 [W] (10995 [W/m²])

(15) (11) (4) (8) (1) (16)

Air ingress Counter-current flow Dotted line shows an estimated curve because a detectable range of molar fraction of helium is lower than 50%. Blow-down of helium & air Natural circulation of air Valve opened

Blow-down of helium gas by a counter-current flow. Blow-down of helium gas that exists under the broken part of the pipe. Blow-down of helium and air. (mixture gas) Onset of natural circulation of air.

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Results of inlet velocity change at the horizontal pipe

40

Heat input: 324 [W] (10995 [W/m²])

Counter-current flow Blow-down of helium & air Natural circulation of air

(1)

He Air He Air

x z

Valve opened

Blow-down of helium gas by a counter- current flow. Blow-down of helium gas that exists under the broken part of the pipe. Blow-down of helium and air. (mixture gas) Onset of natural circulation of air. (15) (11) (4) (8) (1) (16)

Air ingress Counter-current flow Dotted line shows an estimated curve because a detectable range of molar fraction of helium is lower than 50%. Blow-down of helium & air Natural circulation of air Valve opened

International Conference on High Temperature Reactor Technology, HTR 2018, Oct. 8-10, 2018, Warsaw, Poland, HTR2018-180

Conclusion

• The onset time of the natural circulation depended more on molecular diffusion

than the strength of localized natural convection when the temperature difference was small.

• On the other hand, the onset time of natural circulation depended not only on

molecular diffusion but also on localized natural convection when the temperature difference between two vertical walls was large.

• These flow characteristics will be the same as those of phenomena generated in

the passage between a permanent reflector and a pressure vessel wall of the GTHTR-300C.

• In order to prevent a large amount of air ingress into the reactor by injecting helium

gas, we are planning to analyze the method for preventing air ingress by helium canister during a depressurization accident in the GTHTR-300C system.

• In addition, we are now doing the experiment by the double coaxial cylindrical

apparatus and also are planning to carry out 3-D numerical analysis of air ingress when the horizontal primary pipe ruptured.

41