ML19220B581
| ML19220B581 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 01/25/1978 |
| From: | Metropolitan Edison Co |
| To: | |
| Shared Package | |
| ML19220B579 | List: |
| References | |
| NUDOCS 7904270055 | |
| Download: ML19220B581 (19) | |
Text
SAFETY EVALUATION REPORT
'ecnnical Justification for Installino and Testino Tube Sleeves In Once-Through Steam Generator 1.0 Introduction The purpose of this report is to justify the installation of a limited number of internal sleeves in the tubes of a "Once-Through Steam Genera-tor" (OTSG).
This test installation will allow verification of the per-formances of the sleeves under operating conditions.
Sleeves will be in-spected using fiber optics and eddy current methods during refueling out-It is planned that sleeves will be removed at the end of the next two a<es.
fuel cycles for a destructive examination. Any recommendation to allow sleeves to continue in service will be based upon favorable inspection results and the results of long term corrosion testing in the B&W model boiler.
2.0 Background
A number of tube leaks and tube abnormalities have been identified in the OTSG's at the Oconee Nuclear Power Station.
In response, The Babcock &
Wilcox Company undertook a sleeving development program involving mathe-matical analyses and laboratory testing.
The objective of the program was to develop sleeve design and installation methods which would reduce the tube dynamic stresses in the region of indicated tube abnormalities.
Accordingly, sleeves were developed which reduce the stresses in the upper-most (16th) span of the steam generator.
The sleeves do not function as a primary or secondary pressure boundary but only as a stiffening device to reduce dynamic stresses.
80~180 790427oosyg
- 3. 0-Descriction The OTSG tube is Inconel 600 with a.625" outer diameter and a minimum wall of.034" The sleeves (See Figure 1) are 18 and 10 1/2 inch long tubes of Inconel 600 with a.545 inch outer diareter and a.431 inch inner diameter.
The sleeves, one which extends 9 1/2" below the secondary side of the upper tubesheet and the other which spans a support plate (See Figure
- 2) are secured to the OTSG tube by one inch long expanded regions.
These regions are expanded such that the tube 0.D. is increased 0.002 to 0.006 inchas.
Proper expansion is verified by sleeve I.D. measurements accounting for the sleeve-to-tube installation clearance. Most sleeves are secured with two expanded regions.
Two of the eighteen inch sleeves are secured with four expanded regions (See Figure 3).
B&W has developed a hydraulic method of producing the expanded zones which secure the sleeve into the OTSG tube.
The tube / sleeve expansions are accomplished with high pressure fluid and a polymer tube seal.
As an added precaution against loose parts, a retaining collar (See Figure
- 4) will be added to both ends of each sleeved 0TSG tube. The retaining collar consists of a short piece of Inconel 600 tubing which is welded into the ends of the OTSG tube.
It is designed to confine the sleeve within the OTSG tube in the unlikely event that the sleeve expansion should relax or for any other reason the sleeve should become loose in the tube.
4.0 Design Verification The performance of the sleeves was verified using a combination of analytical and experimental techniques. A NASTRAN computer model of the OTSG tube and a sleeve at the upper tubesheet was used to determine the optimum ~ sleeve length for reducing tube stresses.
The optimum sleeve configuration was then subjected to 16boratory testing to confirm the analytical results.80-181 4.1 Analytical The optimum effective sleeve length for a sleeve at the upper tubesheet, as determined with a NASTRAN computer model, is an 8" extension of the sleeve (measured tc the centerline of the expansion) below the secondary side of the tubesheet.
This configuration resulted in a theoretical reduction in the OTSG tube stresses of 48% at the tubesheet and 24% in the mid-portion of the top span.
The theoretical fundamental response frequency is increased by the 8" sleeve from 50.8 Hz to 58.0 Hz.
A favorable comparison with test results are shown in Figure 6.
Due to the long length of the actual 0TSG tubes (apprcximately 56'),
it was necessary to perform the laboratory vibration testing with less than a full length representation.
Consequently, the dynamic response of models containing differing number of tube spans (support plate to support plate) were evaluated with the NASTRAN model.
The results in-dicated that a 5 span model was sufficient to dynamically represent the full length OTSG tube.
4.2 Laboratory Testino The laboratory testing was divided into several phases addressing various performance parameters.
The testing included:
1.
Proof testing of the sleeves in the 5 span OTSG tubes.
2.
T' al and vibratory relaxation testing of the expanded sleeve /
Joint.
3.
Short and long term corrosion testing of the sleeve and tube assembly.
4.
Wear testing of the sleeve and tube assembly.
The proof testing was conducted to detenine the relative improvement in the vibration characteristics of an OTSG tube due to the addition of 80-182 a sleeve.
The test apparatus (See Figure 5) consisted of a 5 span length of OTSG tube supported in a split-block simulated tubesheet.
Four pieces of 0TSG support plate stock were attached to a " strong back" cor.sisting of a large "I" beam. An electro-dynamic shaker was used to drive the upper tube span.
The test consisted of sweeping the tube with a constant force sinusoidal input and recording the peak deflection and the strains at several critical locations versus exciting frequency. This test was performed on unsleeved 0TSG tubes and repeated using the same tubes with the optumum sleeve configura-tion (See Figure 6) installed.
From the resulting data, the peak strain, peak deflection, critical natural frequency, and damping coefficient were determined.
The results indicate that the sleeve reduces the peak tube strain in the uppennost span by 25 - 40% and increases the critical response frequency by approximately 2 Hz.
The peak deflection is reduced by approximately 20 - 35% and the damping is not significantly altered.
Additionally, a test was performed on an unsleeved 0TSG tube and repeated with sleeves installed as shown on Figure 7.
The results indicate that the peak tube strains in the uppermost span drop 10% and the strain at the 15th support plate dropped by 32%. Although the strains increased 5% in the span belcw the support plate sleeve, the strain level was less than an unsleeved upper span.
The retaining collar was proof tested to detennine the strength of the attaching weld.
This was necessary because the weld configuration did not readily lend itself to analysis.
The testing confinned that the weld was stronger than the original tube.
The tube was then analytically
'shown to be capable of withstanding any load which the sleeve could apply subsequent to its becoming loose in the tube.
Consequently, the retaining collar was demonstrated capable of confining the sleeve to the OTSG tube and preventing it from becoming a loose part in the un-likely event that the expansions should fiil to hold the sleeve.
The tube / sleeve expanded joints were tested for integrity by applying an axial load to actual expanded samples. The initiation level for relative motions was approxiestely 400 pounds for the hydraulic expansion. A minimum load of approximately 2200 pounds was attained prior to total separation with all samples. Additional pull testing was performed on samples which had been subjected to 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> at an elevated temperature of 600 F and vibrated at the critical response frequency for 107 cycles.
This test was conducted to detemine the effects of thermal and vibratory stress relief on the tube / sleeve joint. The results indicated no deoradation of the joint integrity.
An accelerated corrosion test (electro-chemical) was performed on material samples frcm the expanded region of both the tube and the sleeve.
This test involves exposing the samples to a 150 mV electrical potential in a 550 F bath of 10*. sodium hydroxide solution.
The samples are then compared to control sample - a specific heat of Inconel which is known to be resistant to caustic stress corrosion cracking.
Consequently, a qualitative result is obtained regarding the corrosion resistance of the test sample. The results indicate that the hydraulic expansions do not reduce the corrosion resistance of either the sleeves or OTSG tubes.
The sleeve / tube assemblies were subjected to an accelerated wear test.
This test involved shaking a sleeved tube at its critical natural fre-7 quency for 10 cycles. A conservative vibration amplitude of.140" (peak to peak) vis selected based upon available in-service test data.80-184
.3-
The tube was then sectioned and the joint contact areas were examined witn both an optical microscope and a scanning electron microscope for indications of wear.
The results revealed no indications of wear.
To assist in providing corrosion data which leads the operating plant, B&W has installed several sleeves iato two satellite test fixturesof the model coilers at ARC.
Each of these satellite test fixtures contains a single OTSG tube with typical primary chemistry and flow and bathed in typical secondary chemistry. Operation of the satellites is governed by the operation of the model boiler in use for other testing.
It is B&W's intention to operate the satellites at all times when the model boiler is in operation.
5.0 Functional Significance The effect of the sleeve upon the local thermal perfonnance of an 0TSG tube has been evaluated. The results indicate that the primary coolant flow rate in the sleeved tube will be reduced less than 12"..
The resulting steam temperature surrounding the sleeved tube will be reduced by approxi-mately 10 F.
Consequently, the test sleeves will have a negligible effect upon the OTSG thermal performance.
The remaining means by which the sleeve could influence the functional capabi-lity of the steam generator is through tube degradation resulting from wear or corrosion of the tube / sleeve joint. This has been previously investigated through accelerated testing (See Section 4.2) and all existing results indi-cate that the sleeve will not cause OTSG tube degradation.
6.0 Non-Destructive Examination of an OTSG tun / Sleeve Assembly Special eddy current probes (See Figure 8) have been designed and constructed g - g g3
for the inspection of sleeved tubes.
The standard inspection probe, differential annular type, nas been reconstructed with a smaller diameter in order for it to pass through the smaller sleeve I.D.
Other special proces, radial absolute and differential pancake, have been constructed to complement the differential annular probe examinations.
Prior to sleeve installation, an eddy current examination of the entire tube length will be completed with the standard differential annular probe.
This examination will provide a baseline inspection documenting the tube condition just prior to sleeve installation.
After the sleeve installation, a second inspection of the entire tube length will be per-formed with the reconstructed (reduced diameter) differential annular probe.
This first special probe examination will provide an adequate examination of the sleeve / tube assembly with some degradation of results at the expanded joints.
At the tube / sleeve expanded joints, the change in diameter creates a signal requiring special probes and test techniques.
The special radial absolute and differential pancake probes will be used to interrogate the expanded regions in detail.
The inspections with the standard probe will be done at the optimum testing frequency of 400 KHz and with the three special probes at their optimum frequencies of 50 and 100 KHz.
An extensive inspection program of examining calibration samples was undertaken to characterize the phase angles generated by defects at various regions of the tube / sleeve assembly.
Figures 9 and 10 show the types of calibration samples abricated for the calibration program.
Cal 1 (Figure 9) is the defect simulation used to develope curves of Phase Angle vs. Depth of Degradation that is applicable to the unexpanded portions of the sleeve.
By sliding out the sleeve portion of this simulation, calibra-tion of the unsleeved portions of the tube can be accomplished.
A
_7_
80-1SG
comparison of this machined simulation with that required by ASME Section XI, Rules for Inservice Inspection of Nulcear Pcwer Plant components in as follows:
Percent ASME Fabricated Wall Standard Simulation Through-0.052" hole 0.040" hole wall I) 80 5/64" FBH 1/8" FBH 60 7/64" FBH 40 3/16" FSH 3/16" FSH 20 4-3/16 FBH 90 1-3/16 FBH apart 20 1/16" wide groove 360 around I.D.
10 1/8" wide groove 360 around I.D.
Except for the 80% simulated defect, all defects are of equal or smaller volume of removed metal than the ASME standard.
Although the 60% standard was not simulated, the conservatism of the other simulations will allow probe calibra-tion to levels of detectability at least equal to that required by the ASME code.
H-Exp-l (Figure 10) is the defect simulation used to develop signal characterization of the expanded portions of the sleeve / tube assembly.
These simulations contain defects located in four regions of the expansion:
1.
Middle of the expansion on the tube 0.D.
2.
Transition of the expansion on the tube 0.0.
3.
Middle of the expansion on the sleeve 0.D.
4.
Transition of the expansion on the sleeve 0.D.
Fabrication of separate simulation aosemblies with different depths of defects allcw the development of the signal characterization for the expanded areas.
Y^Yl
- 1) Flat Bottom Hole Actual inspections of iv stalled sleeve / tube assemblies will begin with the calibration of the eddy current equipment.
Calibration will be performed with the Cal 1 simulation.
The probe will be inserted into a clean sleeved tube section and balanced at 100 KHz.
The probe will be drawn'in through the simulations and calibrated.
Also char cteristic signals will be recorded by drawing the probe through H-Exp-l simulations.
A manual controled in-spection of the tubes follows and then repeated at 50 KHz.
All data will be recorded on magnetic tape.
Any indications will be interpreted using the curves developed from the simulation assemblies.
7.0 Safety Sicnificance The OTSG tube sleeve will have no detrimental effect upon the safe. operation of the plant.
This is assured by the following:
1.
Extensive pre-installation testing which indicates that the sleeve does not degrade the OTSG tube through wear or corrosion effects.
2.
Ongoing corrosion testing to minimize the risk of a long-term corrosion problem going undetected.
3.
Joint integrity tests which confirm that the sleeve / tube joint is suffi-ciently strong to restrain the sleeve in position during OTSG operation.
4.
A sleeve capturing collar which will be welded into both ends of all sleeved OTSG tubes and has been demonstrated capable of confining the sleeve within its-tube in the unlikely event that the expanded sleeve / tube joints should fail.
5.
Eddy current probes were developed as a part of the sleeve development program to inspect the sleeve and 0TSG tubing.
These probes will be used during the baseline and subsequent examinations of the sleeves and tubes.
Consequently, extensive evidence exists that tube sleeving will reduce opera-ting stresses in the critical apper spans of the OTSG tubes without jeopar-dizing the safe operation of the steam generator.
[j() jgggg
-g-
P
?
s
/
A
~
i 4
1"
, s.
.s.
J
~
/
s
~
~
s s
/
s s
/
k
.545 0.D.x.431 I.C.
A s
(SLEEVE) s s
s s
s
.625 0.D. x 0.34 MIN YlALL s
(OTSG TLEE) s g
?
4:
. 0 0 2.
, O O G s
y o.D. EXPANSION N
~
s
_ y
~
c e'
+-
FIGURE 1
gg.ygg
\\
~
m e
s S
a w
W Po 6
a
~2
(
(
~
sN' XNNNNNNN NNNNNNNNN O
>0 "c) 3 J
f~
t 1
L 4
J g
g xN'. "NNNNNNN 4NN' NNNNNNN c.
[
U LJ o
a Q.
1 m
M E
C.
A $2
.y a
k n
w M
S
\\
- is c
N ah A, s
n'N
. c 5E
^
r G U R E 2.
!m" 80-1e0 W8 l
'QNEf '
L'M/ @
OFF LANEW OFF 2 A N@ OFF LANC& Opp LAMEQ
'g 70BE TUBE TUBE TuS E T U B E'
-21 77 23 17 'i7 f fo-fo gr-114 47-1of
) UPP6R H
(IT1 5 NCE~T (n'
I Q
l 3J L
T T
@l R
t a me>
L it A
4 4
!N A
f A !
A A
,A A
U U
l
'u U.
U L'
.hl Il d Lh bl.li y\\
L l-y j
i' I/
I/t
/
/
J/o
{d n
A n
Lc]
d)
La c
INGrRue. Err A
13 : MC'E D
Y.
A("TE,1 R
// U/t 5 i-w f
9 ig a
.60PPO RT c
i w.- p 4 3 w
+
1 a
c bz ad G,,
( 4J 1 i
em L
x
.. ),w Pd-rf-77-2 I
i i
e
\\
J PRIMARY SICE CF TW E SHEET I
7
- I.D.
i 16 z _,
a A
sl/
e, p\\
\\n'
/\\
N/
1
~~
g,, -
A N-
/\\
s sp
/
\\
\\
7 N /
]
s
/
( SLEEVE RETAINING.
CCLLAR s
/
/
/
9 Q/
OTSG TG E
^
FIGURE 4 80 1sg h-h
I \\-
_g...3
{ g\\
==
Q t
.T... b
~
/
v
\\.;
s 9
.j
\\
s o
y a
"i gI g/
.o r
i Eg
./,.<
'\\
(,
\\\\
^.
5
\\\\
ff DJ n'
- 1
\\
Os
\\
\\
g, g
,i
\\
\\
., \\
~
1.,
s.
\\
go o.
\\
g<
~
af
-}
o
\\
~
1.
a r
g e,
\\
U
}N
\\
\\
o e.
j el
'y
.s' f
~
~. ~..
p"'
y
's
\\
g g
g S '.
l-
~
_ q
/
p l
f,0 l l ( q;. h
?'
'f M
g'g
-1) 9L
~
3
<n p
l Ef
- C e
t5
\\
o
.q r
e
\\
^
gj
\\
\\g f.
gj e
.*e 8
.sg ry
.~
'.,4 L-y
\\
cf
.s
\\
a a
.'\\
\\
j pa mi-tea
..N P;Gu q 8* q
=s,
SMEVE AT TUBE SHEET
[6[6 TUBE
/~
l 2
3 SHEET 3
3 c
O p
RESULTS g
ANALYTICAL TESTING p
STRAIN AT TUBESHEET f'
48%
25 - 40%
~
, i, u
y
\\
i I
25 - 35%
I STRAIN AT HID SPAN C
24%
2
.n p
i h'
STRAIN AT SUPPORT PLATE
[3_
t 11%
" 20 - 30%
s
.y DISPLACEMENT 29%
y 7 20 - 35%
l i
J NATURAL FREQUENCY 7 Hz l 2 Hz'
-~
~
o g
i
SLEEVE AT 15th SUPPORT PLATE TUBE 2
1~ 64 SHEET /m - ++ --
N g-3
[
TEST RESULTS O
q SLEEVE 1 & 2 O
STRAIN AT TUBE SHEET d
35%
i 1r
(
STRAIN AT 16TH SPAN 6
4 10%
2 P1 q
STRAIN AT ISTH SUPPORT 63
~
p 32%
STRAIN AT 15TH SPAN I4 E
5%
l DISPLACEMENT l' %
1r 6
m.
-C NA.TURAL FREQUENCY g
2%
M (C
M I
.r~^-'"~
'?'*'
~
O
' ' ~
f FIGLDE B.
EDDY CURRENT PROBES L
D
' STANDARD & RECONSTRUCTED DIFFERENTIAL ANNULAR PROBE
.h O
['
s k
(Ol o
l.
RADIAL ABSOLUTE PROBE I
,O I
wwu cc C
ie O
.(.
CC cn
_ DIFFERENTIAL PANCAKE PROBE 4
f i
di
X o
.]
o
<b si.
e a3 W i
a CE3 6 m
o N -
in o
cw s I
t
-<x
. ao owe mwe g
g C
<t i
l' O
w
'l w
J J
?
i.
o o
H
=
g 2a
=
I y
2J
- C o J
- C o. = J
- >- 4 4
-e C
W 4
H o>3 o63 g
o o
m as w W
N o
i e
a as W y
m >.-
g P
o 4
W
~
as W ew C
w
. J o 4,O e
J z
1 W
i e
e m
D J
r
,2'-
l w
W I
e o
a 3
' 1, I
.W 3
< a. _Cn a, o
2
,e c
o a
= * - - *
-C o J
- - - =
o GHX
' 1 W
8 a3 W d
o i
o
.m 40 H i
N o
1,
_, M
.oWn
. _a,e b<
C W C%
A N
\\
8 m
w w
Q Q
i 4
J LaJ =C db W aC
> 3 i
> 3 o
o i
oL M
ow i
E o M o o
o.-
s.
_c _C' Co i
m I
i80~197 i
~
Wese
a 3
~
67 L
P R
OTSG. TUBE & SLEEVE EXPANDED EDDY CURRENT SAMPLES O
g 1"
H TYP "A"
"B" "C"
"D" Ji A
l
,. -.- - - - - - -. nnn-n a-r-J1,
..=_ _x-Gl
.f
- ~~
~_
-j e
s.
l
~
fLA- _ mmmc suw e wwG h%wew,ksr>W4 th,arswvesra:-CRg',a 6
- l-6 6
6
-6
=
SAMPLE NO. 1
.1874, 40% THRU DEFECTS AT A,B,C & D g
NO. 2
.1874 '60% T11RU DEFECTS AT A,0,C & D 9
NO. 3
.1874, 80% THRU DEFECTS AT A,B,C & D w
te 2
1 i
.i