Creep Rupture Damage of

Nb-Containing Catalyst Tubes

Creep stress on 24Cr/24Ni-Nb catalyst tube was analyzed by computer simulation using actual
operation and design condition data and were compared with that on HK40 catalyst tube. The creep
rupture damage was also assessed by destructive testing.

Takaaki Mohri, Toshikazu Shibasaki, and Katsuaki Takemura

Chiyoda Corp., Yokohama 230, Japan

Introduction dition data to simulate the creep damage and to com-

pare it with the destructive examination results.

I
n early 1960, HK40 material was used worldwide
for catalyst tubes as first-generation material. Destructive Examination
Since then, much creep damage occurred on the
HK40 catalyst tubes, as reported in the 1970s at the Destructive examination results for Plant A
AIChE Ammonia Safety Symposium.
As a substitute material for HK40 material, Chiyoda The two sets of the 24Cr/24Ni-Nb catalyst tubes
selected in 1974 24Cr/24Ni-Nb material as second- were taken from a hydrogen reformer furnace after 18
generation material as standard catalyst tube material years operation to investigate the aging of the materi-
for the steam reformer furnace. As of July 1996, a total al. Table 1 shows the summary of design conditions
of 2,289 pieces of 24Cr/24Ni-Nb catalyst tubes have for the tubes.
been installed for 22 sets of steam reformer furnaces. Table 2 lists the test items and notations for the test
The oldest 24Cr/24Ni-Nb catalyst tubes for the coupons for the destructive examination, and Figure 1
hydrogen reformer built in 1975 were replaced with also shows the detail notations for the tested tube.
new 25Cr/35Ni-Nb-Ti catalyst tubes due to the termi- Results of the destructive examination showed
nation of the tube life as assessed by the destructive aligned voids in the base metal at the inner side tube
examination results performed by Chiyoda. wall and isolated voids at the outer side tube wall
On the other hand, 24Cr/24Ni-Nb catalyst tubes respectively, as shown in Photo 1. These creep voids
after usage of 15 years were also destructively investi- were distributed over all of the areas observed by
gated. Creep stress analysis by computer program was microscope, from the inner side to the outer side of the
carried out using actual operation data and design con- tube wall.

AMMONIA TECHNICAL MANUAL 50 1997

 Table \. Summary of Service Conditions

Equipment Steam Reformer Furnace for H2Plant

Material 24CR-24Ni-Nb

Used Periods 18 Years

Dimensions Outer Diameter: 152.0 mm

Wall Thickness: 12.0 mm
Inside Surface is Machined

Design Conditions Temperature: 893°C

Pressure: 31.5 kgf/mmA2G
Operating uonditions INLE l Temperature: 520~u
Pressure: 22.9 kgf/cmA2G
OUTLET Temperature: 820"C
Pressure: 18.4 kgf/cmA2G

Table 2. Test Items and Notations of Test Coupons

Test Item 0.1 1C 1U 1M 1L 1.2 2U 2M 2L 2.3 3U 3M 3L 3.4

Dimensions O O O O O O 0 O O O

PT O O O O

Chemical Analysis O O O O

Macro. Examination O O O 0 O O O 0 O O O O O O

Micro. Examination O O O 0 O O O O O O O O O O

Hv Measurement O O O O O 0 O O O O O O O O

Tension Test O O O O O O O

Creep Rupture Test O

AMMONIA TECHNICAL MANUAL 51 1997

 Ill I i
1C 1U IM 1.2
I I t
2U 2M 2L 2.3
i
3U 3M
•
3L
i
3.4
IE

Figure 1. Location of sampled test coupons.

1
\
\
8 V
\
\
6

\
\ »

« "

\Ç

v°
\
Z c

î
19 20 21 22 23 24 25
\ 26

LMP( 0=17.09)

Figure 2. Larson Miller stress rupture curve for

used 24CR/24Ni-Nb catalyst tube after 20 years of
operation.
- fv 1 | |\i |
^i
\i
i\
i\ 1 A
! M
*Ki —i—~~
\l

'
IM t
\i Si
K \ i
K|
n !

.0 Ls 1
1
K Nv i
S r —L^ 1 [\
i iiN' ?i \
\i
^T —,
È '•
Î s
t
| [
*-|— Hoop Sires; i? \!• \ S
to* \-
0 1.0kg/mm,2
L \ i\ï \
_j±_i
-
o
a
1 .2kg/mm
1.4kg/mm
'
f- K I
\l u
i '
N,

- D 1.6kgftnm,Z i. KI 1

A I.jjkg/tana 1 ik \
Photo 1. Microstructure of aligned voids found in 3L, A 2.0kg/mnu
1 \ !~-

M
A

as polished (xlOO). T i i
750 GC 0 aSO 900 950

Temperature, deg.C

Figure 3. Rupture time vs. skin temperature for

various hoop stress calulated from the Larson
Miller curve for used 24CR/24Ni-Nb catalyst tube.

AMMONIA TECHNICAL MANUAL 52 1997

 at 2 year intervals, it is clear that the decision
Table 3. Summary of Service Conditions was correct for the tube replacement to be made
before 2 years had passed, since from a safety
point of view replacement of the tubes at the
Equipment Steam Reformer Furnace for H2 time of rupture is not allowed.
Plant It was assumed that the residual tube life for
24Cr/24Ni-Nb catalyst tubes is close to the
Material 24Cr-24Mi-Nb retubing time when creep voids are found in the
tube wall thickness from the inside of the tube
Used Periods 15 years wall to the outside of the tube wall, even if no
creep fissure is detected in the tube wall thick-
Dimensions Outer Diameter: 152.0 mm ness.
Wall Thickness: 12.2 mm The other remarkable point is creep deforma-
Inside surface is machined tion. Figure 4 shows the bulging in inside diam-
eter (ID) for the 24Cr/24Ni-Nb catalyst tube
Design Conditions Temperature: 895°C along the tube length, and Figure 5 shows the
Pressure: 27.0 kgf/mmA2G relationship between bulging in ID and opera-
Operating Conditions INLET Temperature: 520°C tion time to be obtained from the destructive
examination results performed during 15 years.
Pressure: 23.0 kgf/cmA2G
The bulging in ID clearly increases in propor-
tion to the service period as shown in Figures 4
OUTLET Temperature: 820'C
Pressure: 1 9.5 kgf/cmA2G
Table 4. Creep Stress Analysis Conditions for
HK40 Catalyst Tube
All of the 24Cr/24Ni-Nb catalyst tubes had
to be replaced with new 25Cr/35Ni-Nb-Ti cat- 1. Plant NH3 Plant
alyst tubes after 2 years of operation based on
the results of destructive examinations. Creep 2. Tube OD 152mm
rupture tests were conducted for the tube taken
out at retubing to confirm whether the assess- 3. Tube ID 104mm
ment made was correct or not.
A total of 17 test samples were tested at vari- 4. Wall Thickness 24 mm
ous states of creep rupture and the Larson
Miller Stress Rupture Curve was made as 5. Effective Length 12mm
shown in Figure 2. Based on the Larson Miller
Curve, estimated rupture times were calculated 6. Material HK40
for various hoop stresses against the tube
metal temperature, as shown in Figure 3. 7. Operating Press. 35 Kg/cm2Q
Figure 3 shows that if the hoop stress occurs
at around 1.4 kg/mm2 (1.2 kg/mm2 of pressure 8. Operation Time 104,000h
stress + 0.2 kg/mm2 of residual thermal stress)
at a tube metal temperature of around 900°C, 9. Operating h per year 8,000 h/Y
the tube life will be terminated after operation
of less than 20,000 h. 10. No. of SD 1
In the case of this hydrogen reformer, as the
annual shutdown for maintenance is scheduled 11. Design Temp. 900°C

AMMONIA TECHNICAL MANUAL 53 1997

and 5. This is because the 24Cr/24Ni-Nb catalyst tube by high temperature oxidation.
wall thickness is thinner, and the ductility of the mate- The outside surface around the D portion for tube
rial after aging is higher than those of the HK40 cata- No. 1 was examined by liquid penetrant inspection
lyst tube. (FT) and many indications were detected as shown in
It is understood that if creep deformation is detected, Photo 2.
the residual tube life until retubing can be estimated. Photo 3 shows the result of PT for the cross section
around the D portion for tube No. 1.
Destructive examination results for Plant B It is .clear that the tube life seems to be almost termi-
nated, since creep fissures and creep voids were
During the annual scheduled shutdown of a hydro- detected through the tube wall thickness, even though
gen reformer furnace after 15 years operation, outside no leakage was seen on the tube during operation.
diameter (OD) of all the 24Cr/24Ni-Nb catalyst tubes It is also found that the creep fissures will start from
were measured at 500 mm pitch mainly for outlet por- the outer surface for this tube in view of the PT.
tions of the tubes in accordance with the safety main- It should also be noted that the inner side tube wall
tenance manual. shows equiaxed grain structure as opposed to showing
Resulting from the measurements which were made columnar grain structure at the outer side tube wall.
for the tube OD, some unusual bulging in OD were This means that the equiaxed grain structure for the
detected and detailed radiographie examinations (RT) inner side tube wall has the effect of causing late pro-
were performed for all tubes. gressing of creep fissures, as shown in our previous
Both bulged tubes and defective tubes detected by article presented at AIChE annual meeting in 1979
RT were taken out and new spare tubes were installed. (Kawai et al., 1980).
Remaining tubes were replaced with the new
25Cr/35Ni-Nb-Ti catalyst tubes at the next annual Creep Stress Analysis for Used 24O/24NÏ-
scheduled shutdown, although no leakage trouble Nb Catalyst Tube
occurred during the operation.
Figure 6 shows the results of measurements for OD The creep stress analysis used for 24Cr/24Ni-Nb
to be detected as a heavy bulging tube. The largest catalyst tube for Plant B were carried out by a comput-
bulging for OD was 3.6% and located at the outlet er program using the following assumption:
portion of the tube as marked by D in Figure 6. (1) The tube life was calculated on the basis of both
Figure 7 shows the results of measurements for the pressure and thermal stresses. The dead weight of the
tube wall thickness by ultrasonic testing (UT) corre- catalyst tubes and the stresses from bending or from
sponding to the measurement points in Figure 6. It external forces to catalyst tubes were ignored.
also shows that the tube wall thickness was reduced by (2) Creep deformations in each stress relaxation
high-temperature oxidation at the outlet portion of the stage were calculated using a strain hardening hypoth-
tube. The thinnest portion was detected at the D por- esis.
tion, which was the largest bulging point and estimat- (3) The cumulative life fraction was calculated on
ed as the highest tube metal temperature portion. the basis of the Robinson linear life fraction damage
Figure 8 shows the calculated bulging in the ID cal- rale as follows:
culated from the OD and its tube wall thickness at the
measured points for the outlet portions. The largest Life Fraction %= S dt/trxWO
bulging in the ID was 4.2% at the 550 mm portion
from the bottom weld line for tube No. 1. It is neces-
sary to calculate bulging in ID because measured where dt is the time increment, and tr is the rapture
bulging in OD does not always show real bulging, time at a given temperature and stress. A fissure is
since the measured OD is smaller than the real bulging defined as the point where the life fraction reaches
in OD due to the reduction of the tube wall thickness 100%.
(4) No load operation: Immediately after the

AMMONIA TECHNICAL MANUAL 54 1997

 1.5
20 years
• 10 years
• 6 years

1.5

\ 5 10 15 20

Operation Time, year

o 20 40 60 80 100
Inlet
Figure 5. Bulging in ID at max. tube metal temp,
Tube Length, {%) Outlet
portion for used 24Cr/24Ni-Nb catalyst tube.

Figure 4. Bulging in ID for used 24Cr/24Ni-Nb

catalyst tube.

Figure 6. Results of measurement of OD for used 24Cr/24Ni-Nb catalyst tube.

AMMONIA TECHNICAL MANUAL 55 1997

 ï
E- * —w
N
1 r

a == TT* 1
1 -1 ! Tube Wall Thick., mm
t 13.0 5 1 3.0
Tube Wall Thick., mm
1 2.6 5 1 2.9
1 N 2 1 3.0 6 3.0 2 3.0 6 1 3.0
1 -2 -1 A
3 1 3.0 7 3.2 3 2.9 7 1 2.5
4 1 2.9 8 3.1 4 3.0 8 1 2.6
1 1 2.6 5 2.9 1 2.0 S 1 2.2
c
c« K - - W-E 2 1 3.0 6 3.0 2 2.3 G 1 2.2

px
«C
1-2 B
\ "V '' 3 i l 3.0 7 '2.9 3 1 2.4 7 1 2.3
, 4 1 3.0 8 2.8 4 1 2.5 8 1 2.0
1 1 2.5 5 2.8 1 2.4 S 1 2.2
1
| 2 1 3.0 6 2.7 2 2.3 6 1 2.4
2-1 2-1 C
3 1 3.0 7 2.5 3 2.4 7 1 2.5
S r*n

1 4
1
1 3.0
1 2.8
8
5
3.0

2.5
4

1
2.2
2.2
8
5
1 2.G
1 2.3

[1 MA LUJ.
C 2 1 2.7 G a.O 2 2.1 G 1 2.2
C
t*
i
•J^. — 2-e 0
^r 3 1 2.6 7 2.9 5 2.2 7 1 2.0
\L S3 4 1 2-7 g 3.0 4 2.3 g 1 2.0

Figure 7. Results of measurement of wall thickness for used 24Cr/24Ni-Nb catalyst tube.

—o— Tube No.1

—u— TubeNo.2
—•— TubeNo.3_
—*-TubeNo.4~
-*-TubeNo.5
-^TubeNo.6

Bulging in I.D., (%)

Figure 8. Bulging in ID for used 24Cr/24Ni-Nb Photo 2. Results of liquid penetrant examination
catalyst tube. for tube No. 1.

AMMONIA TECHNICAL MANUAL 56 1997

 Rupture Time : 115,451 Hr

O—UfiEWftL IHCKNESSjnm—00

Photo 3. Creep damage at portion D of tube No. 1. Figure 9. Calculated damage distribution
throughout the tube wall for 24Cr/24Ni-Nb cata-
lyst tube in actual operation.

Rupture Time : 195,236 Hr

2 4 6 8 10 12

0—TUBE W«i THCKtëSS^im—CD

4 8 12

D—TUBE WAI THCKNESSmm—OD

Figure 10. Calculated creep deformation for Figure 11. Calculated damage distribution
24Cr/24Ni-Nb catalyst tube at actual operation. throughout the tube wall for 24Cr/24Ni-Nb cata-
lyst tube at design operation.

AMMONIA TECHNICAL MANUAL 57 1997

 3. j
8

5 E
6 fl HT
81
s X
7 * 005.02 Hr w *
»005»
s** i
^ 13.A
I 2 «==^; ><- 2 h^ ~-
"^C" i 1 ^7
—•—^
^L. \ N
i • ^ -—^ û ^aooc j*
-^ |
\JL aoo HT
x ?
\

i "
4 "^

i " VJL
00 Hl

-4

-i -
h"" -6

-a
"«i "j 4 8 0 ~°0 Î 4 6 S 10 12 0 2 4 6 a M 12

0—1« TO K3«ei mi — 0) 0 — MEWLWUeSLim — (D 0 — Œ WU. W»eS m> — »

Normal Operation(a) No Load Operation(b) Normal Operation(c)

Figure 12. Stress distribution for 24Cr/24Ni-Nb catalyst tube at design operation.

TMT H- 60 deg.C, Rupture Time = 71 ,51 8 Hr

TMT + 40 deg.C, Rupture Time = 97,740 Hr
110

100

± Nv

oft 90

ao
h£
r ^^
70 \

^.
60
9 30 940 950 960 970

Tube Metd Temperatue, deg.C

Figure 13. Predicted rupture time at ID and OD for Photo 4. Creep damage observed in HK40 catalyst
24CR/24Ni-Nb catalyst tube showing the effect of tube after 44,000 h operation,
tube wall temperature.

AMMONIA TECHNICAL MANUAL 58 1997

 reformer furnace is switched from normal to shut-
down operation, it is followed by high temperature
steaming after the feed is cut off. In this operation,
the catalyst tubes are exposed to a low temperature
Ruptur« lim« » 99.084 Hri
difference in the tube wall because a reforming
reaction does not take place in the catalyst tubes. A
similar operation takes place hi reverse during the
startup operation. This operation is called a no load
operation and assumes that there is no temperature
difference in the tube wall. The operation for
reducing the oxidized catalysts using steam and
hydrogen is also defined as a no load operation. In
our creep stress analysis, we assumed that no such
load operations would take place during the shut-
down and startup operations.
-TUBE W«L THOOCSSUtim CD

Results of creep stress analysis for 24Cr/24Ni-

Nb catalyst tube for Plant B

Figure 9 shows the result of the creep stress

Figure 14. Calculated damage distribution through- analysis on 24Cr/24Ni-Nb catalyst tube for plant B,
out tube wall for HK40 catalyst tube in 1 sd/year based on actual operation data.
operation at 900°C tube metal temperature. The rupture time is calculated as 115,451 h,

"0 « 8 0 16 » 2« 0 « S 12 IB » 24 0 » ! 12 16 a 24

0 — ME WU. »0»€S im — CO 0 — UK WU. MOES mn — 0)

D — IBE WU. WOKS mm — <D

Normal Operation(a) No Load Operation(b) Normal Operation(c)

Figure 15. Stress distribution for HK40 catalyst tube in 1 SD/year operation at 900°C tube metal
temperature.

AMMONIA TECHNICAL MANUAL 59 1997

which correlates well with the destructive examination tion, no temperature difference occurs hi the tube wall.
results of Photo 3 which shows that the tube life nearly Reverse stress distribution, such as a compressive
reached termination after 15 years of operation stress of -2.9 kg/mm2 at the inside wall and tensile
(115,500h). stress of 4.2 kg/mm2 at the outside wall, occurred due
The feature of the life fraction obtained is that no to a rebound in the creep deformation as shown in
fissure is calculated after 97,401 h service, although Figure 12b. This reverse stress is uniformly relaxed to
the life fraction is calculated from 90% at the inside tensile stresses for both the inside and the outside wall
wall to 80% at the outside wall. during a 5 h period under high temperature.
Figure 10 shows the calculated creep deformation Figure 13 shows predicted rupture time at ID and
for the tube. The creep deformation at the inside wall OD for the 24Cr/24Ni-Nb catalyst tube showing the
is calculated as 4.5% at rupture time of 115,451 h, effect of tube wall temperature.
which is also nearly the same value as the destructive When the feedstock is again supplied, new thermal
examination results of 4.2% as shown in Figure 8. stresses are added to the relaxed reverse stress, and
It should be noted that the creep deformation at the new high stress of 5 kg/mm2 occurs on the inside tube
inside wall is greater than that at the outside wall. wall.
These results correlate with the destructive examina- To study the effect of tube metal temperature to the
tion results. tube life, the tube metal temperature was increased
This is because the peak stresses at the inside wall from the design temperature of 895°C to +40°C and
are repeated at each resumption of operations. +60°C. The calculation results show that raising tube
To compare with the design case, the tube life was metal temperature shortened the tube life considerably,
calculated at design operating conditions as shown in as shown in the calculated rupture time of 97,740 h for
Table 3. The rapture time was calculated as 195,236 h, tube metal temperature of +40°C, and 71,518 h for
which is longer than that of actual operation data. +60°C.
It was considered that the actual tube lives were It should be noted that the fissure starts from the
shortened mainly by those higher tube metal tempera- outside wall when the tube metal temperature is
tures than the design explained later. increased higher than the design. This coincides with
Figure 11 shows calculated damage distribution actual creep damage phenomena shown in Photos 2
throughout the tube wall for 24Cr/24Ni-Nb catalyst and 3.
tube at design operation. It should also be noted that the time for rupture from
Figure 12 shows the stress relaxation history for the the outside wall to the inside wall is calculated as
24Cr/24Ni-Nb catalyst tube during normal and no load being only around 3,000 h of service life.
operations.
When feedstock is supplied to the tube, thermal Results
stress occurs in the tube wall due to the thermal gradi-
ent from a large input of reforming heat into the tube. Comparison with creep damage of HK40 catalyst
As shown in Figure 12a, tensile hoop stress of 5 tube
kg/mm2 at the inside wall and the compressive hoop
stress of -2 kg/mm2 at the outside wall are distributed Photo 4 shows typical creep damage observed in the
along the tube wall as the sum of both internal pres- HK40 catalyst tube after 44,000 h of operation. The
sure and thermal stress. creep fissures occur at the inside tube wall area and
This initial stress is relaxed to 1 kg/mm2 of tensile many creep voids are distributed along the middle tube
hoop stress at the inside wall and 1 kg/mm2 of tensile wall area. In the area close to the outer surface, we
hoop stress at the outside wall by creep deformation at found an absence of creep void.
8,000 h of service. In the case of the 24Cr/24Ni-Nb catalyst tube, the
When the feedstock is cut off during normal opera- creep fissures are found on the outer side tube wall,

AMMONIA TECHNICAL MANUAL 60 1997

while creep voids are distributed along the inner side wall thickness and good ductility after aging.
tube wall, which shows different creep damage aspects These features for the 24Cr/24Ni-Nb catalyst tube
from that of HK40 catalyst tube. lead the creep damage to be different against the
The results of the creep stress analysis revealed that HK40 catalyst tube, since when the thermal stress
the calculated results for HK40 catalyst tube which becomes lower, the stress due to the internal pressure
was presented at the 1992 AIChE Ammonia Safety will be dominant for the 24Cr/24Ni-Nb catalyst tube
Symposium (Mohri et al., 1993) show different results life and the creep voids will appear uniformly through
compared with that of the 24Cr/24Ni-Nb catalyst tube the tube wall when closing to the life termination
shown below. without creep fissure.
Figure 14 shows the calculated damage distribution Figures 18 and 19 show creep damage correspond-
for the HK40 catalyst tube at the design conditions ing to the predicted residual tube life for both the
specified by Table 4. First, the creep damage for the HK40 and the 24Cr/24Ni-Nb catalyst tube. These fig-
HK40 catalyst tube is quite different from that of the ures are based on Figure 14 for the HK40 catalyst
24Cr/24Ni-Nb catalyst tube, since in the case of the tube, and Figure 11 for the 24Cr/24Ni-Nb catalyst
HK40 catalyst tube, the creep fissure reach to the tube.
medium wall after 80,599 h operation time corre- In the case of the 24Cr/24Ni-Nb catalyst tube, when
sponds to 18% residual life for the HK40 catalyst tube creep voids appeared in the tube wall from the inside
as shown in Figure 14. In comparison with that, no fis- to the outside surface, the residual tube life will be
sure is found at 97,401 h operation time corresponding estimated at 23%.
to 15% residual life for the 24Cr/24Ni-Nb catalyst This is quite different from that of the HK40 catalyst
tube as shown in Figure 9. tube showing the residual tube life will be only 2%
Second, although the creep deformation is only when the creep voids reach the outside surface.
around 0.5% from the inside wall to the outside wall Also, when the creep fissures are found at both the
for the HK40 catalyst tube at rupture, as shown in inside and outside surfaces, the residual tube life will
Figure 16, the creep deformation calculated for be estimated at only 0.3% for the 24Cr/24Ni-Nb cata-
24Cr/24Ni-Nb catalyst tube is 4.5% max. at the inside lyst tube.
wall as shown in Figure 8. These different creep damage features for the
Third, stress levels at resumption of operation for 24Cr/24Ni-Nb catalyst tube resulted from its thinner
the HK40 catalyst tube are calculated 4 or 5 times wall thickness and ductility after aging.
greater than those of the 24Cr/24Ni-Nb catalyst tube,
as shown in Figure 15 for the HK40 catalyst tube and Conclusion
Figure 12 for the 24Cr/24Ni-Nb catalyst tube.
Finally, it is remarkable that the creep fissure starts The features of creep damage for the 24Cr/24Ni-Nb
from the inside wall for the HK40 catalyst tube as catalyst tube are summarized below:
shown in Figure 14 and 17, in comparison with start- (1) The bulging hi the ID is one of an effective esti-
ing from outside wall for the 24Cr/24Ni-Nb catalyst mation procedure for remaining tube life. For instance,
tube while increasing the tube metal temperature more when the bulging in the ID reaches to around 2%, the
than shown hi the design of Figure 13. Most notably, life will be close to the retubing, as shown in Figure 5.
the propagation time for creep fissure through the tube The dimensional inspection for the tube OD and the
wall is more than 30,000 h for the HK40 catalyst tube, tube wall thickness are essential for obtaining bulging
as compared with around 3,000 h for that of the hi ID.
24Cr/24Ni-Nb catalyst tube. (2) When creep voids are found in all of the wall
It is considered that in the case of the 24Cr/24Ni-Nb thickness from the inside to the outside surface
catalyst tube, the tube wall thickness can be designed through destructive examination, the remaining tube
thinner than that of the HK40 catalyst tube and the life will also reach to the tune to consider the retubing,
thermal stress can be lowered as a result of thinner even if no creep fissure is detected in the tube wall.

AMMONIA TECHNICAL MANUAL 61 1997

 80

70

x___
* v)
I§
Fi
40
o.
er
99,084 Hf 30

20
4 8 12 16 20
10
890 900 910 920 930 940
D—HJBE WW. IHCXMSS/nm—00 Tube Metal Temperature deg.C

Figure 16* Calculated creep deformation for Figure 17. Predicted rupture time at ID and OD
HK40 catalyst tube after 99,084 h of rupture time. for HK40 catalyst tube showing effect of tube
metal temperature.

Residual Tube Life, %

Residual Tube Life, %

2 — .---...••:.•••:... :"-~
0.3

20 Fissures •-.i-.o------ . . • - • . ' . ^3~3 Fissures

23

50
Aligned Voids
::.'
." ;X"~' •'-"•"•
-,.î ...'»~ CTÎS"V1 Alinnorl Void«;
"..*•' •
60
37 . I—,™ . • . •
Voids Voids
80 AC
» • «
." .

90 50 *.

I.D. Side O.D. Slda

Figure 18. Predicted residual tube life for HK40 Figure 19. Predicted residual tube life for
catalyst tube. 24Cr/24Ni-Nb catalyst tube.

AMMONIA TECHNICAL MANUAL 62 1997

 (3) When the tube metal temperature is increased "Effect of Macrostracture on Catalyst Tube
higher than the design, creep fissure will start from the Damage," Ammonia Plant Safety & Related
outside surface, which shows that the tube life will Facilities, Vol. 20, AIChE, New York (1980).
almost have reached its end. Mohri, T., K. Takemura, and T. Shibasaki,
(4) When creep voids are found in the inner side "Application of Advanced Material for Catalyst
tube wall, the residual tube life will be around 50%. Tubes for Steam Reformers," Ammonia Plant Safety
& Related Facilities, Vol. 33, AIChE, New York
Literature Cited (1993).

Kawai, T., K. Takemura, T. Shibasaki, and T. Mohri,

AMMONIA TECHNICAL MANUAL 63 1997