ML20234B774
| ML20234B774 | |
| Person / Time | |
|---|---|
| Site: | 05000000, North Anna |
| Issue date: | 07/07/1976 |
| From: | SUN SHIPBUILDING & DRY DOCK CO. (SUBS. SUN CO., INC.) |
| To: | |
| Shared Package | |
| ML20234A777 | List:
|
| References | |
| FOIA-87-40 NUDOCS 8707060203 | |
| Download: ML20234B774 (78) | |
Text
- -
- y..
.'y.
i Y'
f.,l.;' f \\ /b
)
P I
T THE SAFETY OF STEAM GENERATOR SUPPORT STRUCTURES FOR NORTH ANNA UNITS 1 & 2 BY SUN SHIPBUILDING & DRY DOCK COMPANY JULY 7, 1976 8707060203 870610 PDR FOIA THOMASB7-40 PDR
___.)
i 1
s o
.s-INDEX PAGE INTRODUCTION 1
SUMMARY
3 i
CONCLUSION 5
j BACKGROUND 6
BRITTLE CHARACTERISTICS OF STRUCTURE 8
CODES & PRACTICES 10 i
ADDITIONAL ANALYSIS 12 i
A-572 PROBLEMS 13 DISCUSSION 14 CONSULTANTS William S. Pellini 16 Richard Roberts 22 TABLES Tables 1 thru 5 FIGURES Figures 1 thru 13 APPENDIX Stakutis/Ragone Letter - 6/24/74 A-1 Method of Converting Impact Codes A-5 to Equivalent Shear @ 80*F Rebuttal to Vepco Comments - 6/11/76 A-lO Re. cord of Material from Bethlehem Steel Co.
A-12 Identification of Test Slabs - Lukens A-15 i
h A
1
.. a l
+
1 PRESENTATION TO ADVISORY COMMITTEE ON REACTOR SAFEGUARDS JULY 7, 1976 INTRODUCTION The question before the NRC and ACRS is:
Is there reasonable assurance that the steam generator and pump supports for North Anna 1 & 2 are safe from brittle fracture given the proposed method of operation and the data available?
To answer that question entails answers to the following:
I. What does reasonable assurance mean?
II. How easy is it technically to give this assurance?
III. Has sufficient data been provided to meet the technical standard in Question II?
The answers are given in this report as briefly as possible.
l l
l l
l l
~-
' ' - ~ ~ ~ ~
~~~~' ~~ ~ '~
t 3
=
SUMMARY
Applied to the question, "Is there reasonable assurance that these structures are safe frem brittle fracture?",
we reason as follows:
I.
What does reasonable assurar.ce mean?
Assurance against brittle fracture can be provided using the principles' of fracture mechanics.
Complete assurance against brittle fracture is given if the structure is operated in the plastic zone (Steel 1 Facing Exhibit).
For a large number of structures, a reasonable level of safety is assured with operation in the upper elastic-plastic region, provided a fracture control plan is used (Steel 2).
Operation to be safe in the lower elastic-plastic region or plane strain region (Steel 3) requires careful design, careful fabrication and extensive brittle fracture analysis and total elimination of flaws of critical size.
For structures as critical as these -- supporting a key element in the primary pressure barrier of a nuclear plant -- operation in the plane-strain or lower elastic-plastic region (Steel 3) cannot meet the " reasonable assurance" standards of the NRC when it is so feasible to operate in the plastic or upper elastic-plastic region.
II.
How easy is it technically to give this assurance?
This assurance can be given in two ways:
1 1)
Operate the structures at sufficiently elevated temperatures to assure that they are in the plastic region (like Steel 1).
2)
If it is desired to operate in the upper elastic-plastic region (like Steel 2), then a fracture mechanics analysis must be provided.
This must be based on toughness measurements from each heat, the weld metal and the heat affected zone (ASME Section III, Subsection NF).
Temperatures of about 250*F would probably meet this in both longitudinal and thru-thickness directions.
(Some additional test data will be needed.)
I 4
III.
Has sufficient data been provided to meet the technical standard in Question II?
No vepco and stone & Webster in their rebuttal of June 11, reject their own solution of raising the structure tempera-ture to operate in the plastic or elastic-plastic regions.
Also, their own data shows that in fact much of the structure will be operating in the plane strain or lower elastic-plastic regions (like Steel 3).
To operate in this region with an assurance of safety requires data on all heats, the heat affected zone, and weld metal.
Also required is a comprehensive brittle fracture analysis and a statistically valid nondestructive critical flaw detection program with an acceptable level of significance.
None of thesc have been provided.
In addition, recent analysis reveals that there are at least 20 to 30 feet of A-572 in the structure with abnormally low energy absorption on the upper shelf.
This beam could be in the most highly stressed area of
{
the structure.
i I
1
)
\\
e
5 i
e CONCLUSION h
public Webster have not provided t e f rmation on the steam generator ssurance of safety at the and St'one &
and NRC with sufficient in osupports to prove reasonab critical locations Vepco proposed temperatures.
ily ll A-572 beams inif they will respond satisfact The impact properties of athey should be replaced.
I must be determined to see itical beams have normalth the longi If not, to heating.
h After determining that all crresponse to temperature in imple heating of the structures and in both the shortest time thickness directions, then s If this solution is with suitable will provide Ehe assurance fety.
with the greatest margin of sathen redesign of the struc tures impractical, steel is required.
4
i 6
0 BACKGROUND The controversy over these structures revolves simply over what is sufficient data to reasonably assure their safety from brittle fracture.
This question is not one on which there should be much of a technical argument (as some other issues might be where the engineering field is not as well developed as fracture mechanics).
The argument rages because of the history of these structures which has placed Vepco and Stone & Webster in a very embarrassing position.
Without re-arguing all the points previously made, let's shnply enumerate the critical facts developed in earlier presentations.
1)
These structures were not designed to any fracture control plan.
Brittle fracture was not considered in the original design of these ' structures.
2)
The design of these structures imposes severe restraints resulting in large residual welding stresses, probably approaching the yield point of the material.
3)
The design of these structures uses extraordinarily heavy shapes as strength members in the short transverse, i.e.,
through-thickness dimension.
4)
The design of these structures is thus peculiarly i
susceptible to the creation of lamellar tears and l
restraint cracks as a result of both materials used and joint details used.
5)
Even after the potential for disaster which this design had built into it was highlighted by Sun Ship's discovery of lamellar tearing in 1971, by Dr. Stout's core samples in 1973, by Combustion Engineering's difficulties with lamellar tearing in 1974, and by Stone & Webster's own analysis, DC-81, in early 1974, Vepco and Stone & Webster obstinately pursued the same course.
6)
Instead of redesigning the structure pursuant to a rational fracture control plan, Vepco and Stone &
Webster wagered that the structures could be fabricated without defects -- a bet which Dr. Week (British Welding Institute) has described as " foolish".
(
Reference:
" Avoiding Failures in Welded Construction",
by Richard Weck)
G 8
e I
7 I
e 7)
We know from their own mouths why this decision was made -- in the belief, now known to be mistaken, that this would have the least impact en the schedule.
8)-
Now Vepco and Stone & Webster have had to justify their earlier misjudgments.
Their efforts to do so are centered on three principal elements.
These are:
I a)
Toughness tests of 2 or 3 of the 20 heats of
' steel used in the structures (while failing to I
reveal adverse data on one heat in their files).
b)
Derivation of critical flaw sizes based on assumed stresses and strain rate.
c)
Disputing the character of detected flaws, and assumption as to the efficacy of their non-destructive test program in detecting all flaws.
9)
Each of these elements has been shown to be, or on its face is, unsound.
On June 11, 1976, Vepco presented a written rebuttal to Sun Ship's conclusions.
The thrust of that rebuttal is communicated in the summary letter of transmittal, which states:
"In fact, the material Charpy tested in the l
supports exceeds the brittle fracture control criteria of that code (referring to Subsection NF) over most of the operating temperature range..."
When analyzed, this very forceful statement is stripped of significance.
The material tested was satisfactory over most, but not all of its operating temperature range.
What about "the material tested?"
Subsection NF requires testing of all heats, not 2 or 3 out of 20.
Subsection NF l
requires tests of the weld metal and heat affected zone.
l This material was not tested at all.
l Thus Vepco and Stone & Webster have tried to confuse dhe issue by avoiding doing what public policy demands, namely:
Characterize the structure's brittle properties.
Compare these properties with good engineering practice for structures whose service failure would have similar consequences.
Provide all the data to permit the NRC and ACRS to judge if the reasonable assurance required for nuclear safety has been met.
i
e 8
We shall attempt to do what they have failed to do based on the information available to us which is obviously limited.
Brittle Characteristics of Structure The discussion which follows deals only with base metal, since no data is available to us on heat affected zone or weld metal in these structures.
In Figure 1, we have compared the three common impact properties used -- Charpy energy, lateral expansion and percent shear for the one or two heats for which there is considerable data (Vepco to NRC - 5/14/76).
The point of this graph is to establish that no matter which criteria is' used the results are basically the same.
Table 1 illustrates the same point by showing that at all temperatures and even in the through thickness direction, the statistical mean and standard deviation of each test method compare closely.
Furthermore, any of the methods can tell you about where the metal is with respect to plane strain, elastic-plastic or plastic region.
The plane strain region is most brittle and dangerous, the plastic region the most ductile and safest.
In between, in the elastic-plastic region, is a gradual shift from brittle l
to ductile or plastic.
We have selected percent shear as the l
measure to use in discussing the steel in this structure because l
it directly indicat es the percent of shear or ductility in l
a sample.
It is the opposite of brittle.
A material that is 100% shear is 0% brittle and vice versa.
Thus Figure 1, with the appropriate 90% confidence levels, is a start in statistically defining the material properties of the steam generator supports.
However, the data came from only 1 or 2 heats out of 20.
Can we infer the properties of l
the entire structure?
We believe that at least an attempt can 1
be made.
The chemical composition of the various heats permit emperical estimates of the range of NDTT for the heats.
This data is listed in Table 2.
It shows that the expected range of NDTT is 20 to 50'F for the A-36 and 40 to 60*F for the A-572 in the structure.
Using this information it is possible to estimate the worst A-36 in the structure by shifting the test data for the 1 or 2 heats tested.
Figure 2 shows the results of this extension of the data.
While some might object to this method of extrapolating the data, as will be seen, the worst A-36 is not as bad as the worst A-572 for which there is actual data.
- - - - - - - ~ - - - -
- ~ - -
~
9 Vepco and Stone & Webster, in preparing their brittle fracture analysis DC-81, had a sample of A-36 and a sample of A-572 tested for impact properties.
The A-36 was purchased from Bethlehem, but the A-572 was a left-over sample from the structure -
" cut from the material used in fabrication" (Sprung - S&W).
The data is shown in Table 3 It was not provided voluntarily in any Vepco or Stone & Webster presen-tation or in answer to direct NRC request (April 26, 1976).
f This data is added in Figure 3.
It is interesting to note that the 15% shear temperature, generally correlated to NDTT, falls at about 74'F.
This is actually 4'F higher than the Sun Ship estimate for A-572 in the structure by chemical composition.
Also notice that the A-572 data falls below the Sun Ship estimate l'or the lower bound of A-36 in the stru cture.
Thus the structure is reasonably characterized 1
with the lower bound set by actual vepco and Stone & Webster data.
This figure now presents a reasonable engineering judgment of the material characteristics of the structures.
Before going on to further analysis of Figure 3, note the very poor slope on the A-572.
The PVRC Ad Hoc Group (August 1972) warns about this type of material when it specifies testing for mils lateral expansion and Charpy at NDTT + 60' "to weed out nontypical materials such as those which might have low transition temperature but abnormally low energy absorption on the upper shelf. "
Also, it is instructive to note that the A-572 was used in extremely critical locations, Figure 4.
Thus the importance of this poor quality A-572 exceeds its representation by weight in the structure.
Finally it would be desirable to develop as statistically accurate estimate as possible of the impact properties of the structure at 80*F -- the minimum service temperature.
Table 4 lists the weight distribution of the original 13 heats as well as the estimated or actual figures for percent shear and standard deviation at 80*F.
These have been combined statistically to yield the following values for the structure:
Mean % Shear @ 80'F 44.5
=
Standard Deviation - %
16.2
=
These are also plotted in Figure 5.
In order to evaluate the significance of the impact properties of these structures, it is necessary to examine good engineering practice and codes.
O I
i 10 Codes & Practice The establishment of codes always follows good engineering practice.
In addition, codes are often approximate ways to quantify the intention of good practice.
We believe the intent of the engineering profession through the 1960's and 1970's has been to operate critical engineering st ructu res in the upper elastic-plastic region.
We would estimate something over 30%-40% shear, as measured by Charpy.
We believe that this was the intent of the NDTT plus system of the Naval Research Lab, the PVRC recommendations of August 1972, the G-2000 subsection of the nuclear code and the Subsection NF of the nuclear code.
In addition, the standards set by three foreign countries for supports in critical service have been obtained.
All of this data is summarized in Table 5.
Clearly, it illustrates that operation in the lower elastic-plastic region or plane strain re not condoned as good practice nor permitted by codes.(gion is 1)
If we now repeat the graph which represents our best engineering guess of the impact properties of these structures and add the various code criteria we see that about 30% of these structures fail to meet good engineering practice or codes.(Fig.6)
This practice was available when these structures were designed and was required by a number of codes by the time the rebuilding was undertaken.
In fact, an internal Vepco memo (A.P.Stakutis -
Project Engineer /S. Ragone, Sr. V.P. - 6/24/74) is instructive.
In this letter the repair of the pressurizer supports is dis-cussed.
We would deem the pressurizer supports and the steam generator supports to be equally important in guarding the integrity of the primary pressure barrier.
Vepco says this about these supports:
" l)
The operating average temperature of the l
pressurizer supports ranges from approximately 100*F to 125*F.
2)
The minimum impact properties of the ASTM A-516 Grade 70 material from which the support is fabricated is 15 ft-lb at -30 F, with an NDTT of -20*F for the 3" thick material.
3)
Employment of the NDTT +60'F desian criterion
- reveals that the support can be stressed up to yield strength without danger of propagation of any inclusions or flaws of any size.
Therefore, a s suming -3 0' + 60
- F = +3 0 'F < 100 'F (minimum operating temperature), fracture mechanics analysis is not required."
(1)
Reader is also referred to WRC #186 - Aug.1973, p.5
- Emphasis added l
l
- - - - - - - - - - - - - - - - - - - ~
~---
I I
1 11 l
1 Further support for both Sun Ship's contention that operation in the upper elastic-plastic region be required for reasonable j
safety assurance and that each heat be tested is provided-in l
NRC Regulatory Guide 1.104 for overhead crane handling systems.
{
The guide suggests NDTT +60' minimum and that tests be made "on each heat of steel used in structural members essential 1
to structural integrity."
Incidentally, the use of this I
standard on crane systems shows that the NRC recognizes that the potential results of failure is the Dnportant criteria, not whether the structure in question looks like a bridge, i
pressure vessel or ship.
Vepco's rebuttal included the ludicrous argument that "the North Anna structures are sbnilar to bridge designs and do not in any way resemble vessel designs."
Apparently they did not grasp the significance of John Barsom's well-reasoned fracture control plan which states:
"The objective in developing fracture-toughness specifications for a structu re should be to establish the necessary and sufficient l
toughness values that ensure the adequacy of the material for the intended application."
We and our consultants pointed out (May 20) that the failure of these structures could be more catastrophic than a pressure vessel or ship and infinitely worse than a bridge.
Therefore, only the highest standards should apply.
In addition it should be noted that there are at least three technical differences between these structures and bridges that further strain the Vepco analogy:
Through thickness stresses are considerably less e
in bridges because of very little welding.
There are much leas residual stresses in bridges, again because of lack of welding.
There are seldom such massive and highly restrained joints in bridges.
Finally, returning to Vepco's and Stone & Webster's attempt to use Subsection NF to certify these structures.
Figure 7 is the data presented by Vepco and Stone & Webster which met the criteria -- except for two points -- although the Sub-section clearly states all specimens shall meet 25 mils.
To this we have added the A-572 data which they failed to produce.
This does not meet the code.
Fu rther, Figure 8 relates mils lateral expansion to percent shear.
Since no test can be below 25 mils by Subsection NF, it appears that i
I for these structures a minimum of 35% shear is dictated.
{
Figure 5 shows that about 30% of the structure would noc pass the current NRC standards if a complete sampling of the structure was undertaken.
1
_---_---___-__J
12 ADDITIONAL ANALYSI_S As pointed out in previous discussions the " Martha Ingram" barge failure provides useful information in analyzing the suitability of the steels for these structures.
Figure 9 presents the impact properties for the barge.
Note that at the failure temperature the steels exhibited good charpy energy and lateral expansion but poor shear (about 30%).
Thus in this case, the shear was more indicative of the steel property than the other properties.
In fact, the 25 mils lateral expansion was attained at 17'F below failure temperature.
This illustrates again the wisdom of Dr. St ou t 's observation "that there is no single notch-toughness test which can predict the transition temperature of a specific s tructu re. "
To further illustrate the point, Figure 10 shows one of the ways that the Ship Structure Committee correlated brittle fracture data.
Note that their Zone I Bounded by a Maximum Shear of 30% and a Charpy of 35 ft-lb encompassed most of the failures.
However, note also that to completely eliminate failures requires going to minimum of 35% shear, and to eliminate borderline performance requires going to 60% shear.
These results are completely in agreement with the code requirement on Figure 6.
Now let's add the actual data available an the steam generator supports.
These consist of three independent pieces of data.
% Shear Cy 1 Sample (Avg. 3 tests) 30 32 out of 8 Samples removed from A-36 Heat 182C174 or 171C866 by Vepco @ 80*F 1 Sample A-572 by S&W (DC81) 14 15
@ 70'F (Heat Unknown) 1 Sample of A-572 - Heat 123C349
@ 80*F (tested 6/29/76)
Flange 20 13 Web 30 29 The last sample of A-572 is one discovered in Sun Ship's storage which was tested on 6/29/76.
It probably is not of the same heat as the Stone & Webster A-572.
As can be seen on Figu re 10, these samples all fall in the area where failures have occurred.
These are not statistical inferences but actual test data.
Also note by comparison of Figures 6 & 10 that the French Code, the Japanese Code and the U.S. Nuclear Code Section III are designed to keep above 60% shear, thus l
eliminating borderline performance.
13 A-572 PROBLEMS l
In preparing for this presentation Sun Ship had a retain sample of A-572 tested.
The results are shown in Figure 11.
This steel is a typical low-energy performance steel --
- Figurel2, We have determined that there were two 30-ft.
beams of this heat purchased originally (Heat 123C349).
We had hbout 30 feet left from which we made the tests reported.
We also sent Vepco and Stone & Webster about 10 feet of A-572 in 1973 (which resulted in the data used in DC-81).
We have no idea if these two samples of A-572 came from the same heat.
This would mean that 20 or 30 feet of this heat was used in the fabrication.
Unfortunately, the A-572 is very strategically located (Figure 4 ).
The stress analysis for the A-572 beams is given in Figure 4.
Note that under design accident conditions A-572 is expected to withstand stresses up to 38 KSI.
Obviously the A-572 beams are critical to structural integrity.
We do not know where the 20-30 ft. of Heat 123C349 is located.
In addition, we know nothing about the remaining heats of A-572.
But, as Vepco and Stone & Webster point out, "all were made by the smme steel producers in the same time frame -- this fact should narrow the number of variables in the steel making and rolling process for these critical beams."
Fig.ll suggests that at least two heats of A-572 are bad.
Thus we believe that some program to determine the propertias of the A-572 in the structures must be undertaken.
Replacement of these beams would seem to be indicated even before con-sideration is given to heating the structures.
i 0
1
14 _..
DISCUSSION If we examine the vepco and Stone & Webster assurances of safety, we find they boil down to taking their word for the safety.
No standards are applied, no complete data supplied, just " trust us."
"It is my engineering judgment that the North Anna j
l supports are adequate."
"We find that there is ample conservatism in material properties and acceptable flaw sizes to dispel concern in this area."
"It is the opinion of Vepco that the structures as repaired do not contain lamellar tears of significance to service performance."
" Ou r view (is) that the supports are adequate to perform their intended function."
"We feel that the supports are safe."
As we appraise Vepco's position, it asks the NRC to accept on faith a probability of safety.
Based on the history of non-disclosures, linited disclosures, selective disclosures, engineering mistakes, management mistakes, and outright mis-statements that have been documented in this investigation and the history of this plant, the faith which Vepco asks be placed in its conclusions is not warranted.
Vepco has twice been cited for gross negligence in failing to report safety information to the NRC -- one for the geological fault for which they were fined, and one for the service water pump house settling.
They have failed once more to report all the relevant data on safety as recently as May 14, when in answer to a direct request for mechanical data on all heats they failed to reveal the A-572 data.
In addition to Sun Ship's presentation of May 20, Sun Ship provided documents on March 29 which showed how Vepco and Stone & Webster used data as they saw fit, withheld data detrimental to their case and generally tried to pull the wool over the NRC's eyes.
Their latest writings do not indicate any change in this pattern.
l 15 t
If the ACTS is interested in delving deeper into these structures, we suggest these questions:
What happened to the S&W fracture analysis report DC81 and its recommendations?
How were the U.T.
specs permitting flaws up to 3/4" developed?
What.is the test data for the remaining 17 heats?
What is the significance of the extremely low through thickness toughness for this structure?
Why did vepco and S&W wait until December 1975 to do any toughness testing?
What can be said about stress levels in the un-stress relieved structures?
Why didn't vepco and S&W explain how they handled the prismatic representation of gussets that lead to high apparent through thickness stresses?
Did they do any finite element analysis?
How many defects were found by M.T.
inspection and how many toe cracks?
How many defects were found by U.T.
inspection?
I What was size, character and location?
s Hokever, we firmly believ.e that the public interest and public confidence warrant an end to this controversy.
Vepco and Stone :& Webster have been given ample time to provide assurances of safety.
They have failed to do so in a sound, technical manner.
We believe that more must be learned about the~ highly stressed beams and a suitable fix made.
Then heating.of.these structures is certainly aheasonable approach.
4 e
i w
a j
a i
i I
CONSULTANTS l
l i
l 1
1 1
i e
l
16
~
8EVIEd 0F VEPCC COMLidNTS Cli SUN SHIP PRESENTATIONS by WILLIAL1 S. FEILINI June 26,1976 (I); INTRODUCTION
}
The VEPCO comments on the Sun Ship presentations involve major issues regarding the basis for certification of structural directed to these issues.
f review is reliability. The i
The first issue is that the characterization of fracture is not properties, with respect to section size effects, In discussing the Pellini represented properly by VEPCO.
presentation to NRC, VErCO cites a sharp transition (60 F above the NDT) as providing elastic-plastic arrest properties. This section size. It is D21 the is the case for steels of 1.0 in.sec tion size. Section size case for cteels of 2 and 3 in.
expanding the r
increases above 1.0 in have the effect o plane strain region in the order of 3d and50'F for ~ J n. ':md s 1.0 in section size.
3 in. nizes respectively, as compared to (L) for A conservative estimate for the plane strain limit 1.0 in. is NDT + 10 to 20*F and for 3 in, it is NOT + 40 to 60 F.
curve (and the EIR The effects are represented by the K Id is curve) which define the temperature rance above NDT that analyses that apply to the plane strain appropriate for Ky In brief, desirable arrest protection is not developed state.
in the 6d F rar.ge above NDT for the thick sec* ion sizes of the It should be noted also that arrest protection structures.
temperature range is not defined by an intermediate loading rate Ky curve.
(1)
~
,~
17 l
l The second issue, is'that unique deci'gn features-(redundant) l are used as an argument by VEPCO against the use of a kid ( r.Kyg)
' curve analysis. Differences between pressure vessels, ships and trussed structures may decide if* total fracture occurs t or not, following initiation. These differences have nothing to dn witii analyses'of fracture initiation,.because it is a purely localised event.
The third issue is the VEPCO assertions that meeting of a f
specific C value, that corresponds to plane stra,*n properties, y
provides certification of structural reliability. No such guarantee is provided; it is a matter of analysis based on an appropriate Ky curve and other proper engineering considerations that decide if fracture initiation is possible. Thus, meeting of 23, 24 or 25 mil L.E. for the C specimen, at 80'F does not define per se y
that the structure is fracture safe. Tha ASME Code (NF) does not define what ' type of fracture reliability evolves from meeting minimum cited values.of L.E. by the C test-- it is left to y
appropriate analyses methods to do this.
The fourth issue is that t6e VEPCO case for structural reliability'is primarily based on the argument of redundancy.
l However, the discussions of fracture properties do not provide l
an adequate definition as to the real importance of emphasizing redundancy. In our view, this is because the fracture properties are not adequate to preclude fracture initiation with the assurance i
required for a critical structure.
1 (2) 1
(II) DISCUSSION 18 There are three rotttes by which the structural reliability with respect to failu"e by frac,ture may be examined for any structure. These are (1) arrest criteria (2) initiation criteria and (3) structural redundancy.
l Arrest criteria are the most positiva hecause they provide assurance that fracture cannot develop. Initiation criteria must include many factors other than the fracture properties of the base steel, and thus are generally complex and difficult to validate in practice for structures that are not stress relieved.
Structural redundancy is generally considered as a factor in relation to the applicability of either arrest or initiation criteria. For example-- if arrest criteria are assured then there is no dependence on protection by redundancy. If initiation criteria are involved (by necessity due to fracture properties) then the relative structural redundancy is an important factor.
If structural redundance is positively assured, there is no need to dirruss fracture properties; except as to the ponsible develop-ment of partial fractures, which do not cause immediate catastrophic failure of the structure as a whole.
The first point that should be resolved in assessing the fracture reliability of any stucture is if the steel used features arrest properties or not.
All available data indicates that the A36 and A 572 steel of the section size used, does not have fracture arrest properties.
Even if it is assumed (not proven at this point) that the NDT is 40 to 60 F maximum for the steel population in the structure, I
a service temperature of 60*F higher level (that is 100to 120'F) does not result in desirable fracture arrest properties for 3 in.
thickness. In fact the service temperature may be as low as 60*to 80 F, as cited by VEPCO. Accordingly, the properties are close to
. plane strain for which E could be measured, at the lower end Id of the service range; or at best of low level elastic-plastic properties at the high end. If the maximum 11DT temperature exceeds 40 to 60 F, plane strnin properties (XId) should be expected over the entire service temperature range. In brief, a dynamically loaded KIC specimen could be made to fracture in a typically flat, brittle fracture mode at service temperatures.
(3) i
i.
19 i
The difference in fracture appearance for steels of 40', 60'or 80 F 2
NDT, and 2 to 3 in. thickness, would primarily involve the presence or absence of small shear lips in the service temperature range. It i
woul,d be informative to conduct such tests, at least for steel of 40*F NDT.
VEPCO has avoided any direct claim that the steel is of arrest l
properties at service temperatures of 80*to 120 F because this is obviously not the case. Since the steel can fracture in a brittle mode, withflittle or no shear lips, it is then necessary to examine l
Very closely the conditions that may cause fracture initiation.
Fracture initiation is a very localized event-- it can develop in wc1d or HAZ regions and in backing bars, clips, etc. In attempting to establish initiation prevention criteria, the critical issue is no longer the base metal alone. It includes all metal sites that can serve as points of initiation.
No credible case for fracture prevention can be made unless it is proven that initiation is not possible for normal and other conditions of service. For structures that are not stress relieved, there is no practical way to prove prevention of fracture initiation.
This is particularly true for large welded structures containing a large number of complex weld connections. If the base steel is not of fracture arrest properties, the only credible alternative is to invoke protection of the structure as a whole by virtue of redundancy, if justified.
However, redundant design does not preclude the development of local frac +"res in regions that do not cause catastrophic failure of the enth e ;tructure. VEPCO appears to accept this fact by citing (page 2 of enclosure 1, letter to NRC, June 11 1976) quote:
"the wide margin of safety of the bridge (trussed) structure, where, under rare circumstances, critical members have failed long before any further problem with the structures."
These events have not been entirely rare and when they developed, the continued use of the bridge structures was due to not knowing that a local fracture' event had occurred. When detected for a critical member, it has been general practice to repair as a reasonable precautionary measure. Moreover, these events have led to case inquiry,' removal of fracture test specimens and other measures deemed appropriate by responsible authority.
(4)
_ _ _ _ - _ - _ - - - - - - - - - - - - - - - - - - - - - - - - - - - ' ~ ~ ~ ~ ~ ~ ~ ~ ' ~ ~ ~ ~ ~
20 1s it inferred by VEPCO that acceptance of localised frecture should be the case for the support structures */ Is VEPC0 prepared to L.uarantee that no rare" events of localised frac ture can I
develop in the structures during their service life and considering normal and other conditions of service'!
Juch a guarantee does not evolve by the use of a modified (45 F shift) AASHTO analysis. Similarly, such a e;uarantec does not evolve by barely meeting the ASME Subsection NF fracture r
test at service temperatures.
properties of 25 mil L.E. by the Cy In fact, consideration of such low fracture properties should lead to reasonable inquiry that calls for the prudent application of ASME Appendix G analysis methods-- including weld, HAZ and any other potential source region for fracture initiation.
VEPCO objects to the use of Section G analysis while the subsection NF endorses its use as valid for support structures.
The VEPCO objections to Section G analysis appear to center on the fact that the structure should not be analyzed as for the case of a pressure vessel. The primary objection is that the support structure is protected by redundant design and the pressure vessel is not. However, relative redundant features have no direct bearing on the results of the analysis with respect to initiation of fractures.
curve analysis Apparently the VEPC0 objections include hId as well as KIR curve analysis. Prof. Corten has used an intermediate loading rate Ky curve type of analysis. Pellini's presentation was misunderstood by VEPCO. The real point that he made was for the use of K curve for ane. lysis of initiation from base metal, and Id a considerably lower curve for the case of HAZ initiation for welds that are not stress relieved. The real issue in this. respect is the credibility of the specific E curve that is used for analysis 7
purposec. Objections to the Kyg curve raised by VEPCO, do not resolve this question and do not in any way answer why a proper K
curve analysis should not be made.
Id The VEPCO objections to considering failure experience for ships, all types of pressure vessels etc., is purely based on the argument of structural redundancy for the support structures.
Carried to logical conclusions, this point of view says that fracture properties are not of primary consequence. The same claim for safety would evolve if no discussion was made of fracture properties.
(5)
l The VEPC0 claim for safety, if accepted, would hold for any level 21 of planc strain fracture properties as well as for low levels of elastic-plastic fracture properties.
There is no confusion on this point by the Sun Chip consultants.
They collectively recognize that VEPCO has not advanced a fracture control case based on either initiation or arrest criteria. It has l
simply invoked redundancy in defense of the structures. The appear-ance= nc fracture prevention case based on initiation criteria is provided by Prof. Corten's calculations. If these calculations are l
accepted, then there is no need to consider the differences in structural redundancy between various types of structures as empha-sized by VEPCO. His case is that fracture initiation is not expected.
It is repeated that the VEFC0 calculations are equivalent to proving, by mathematics, that notable failures of many engineering i
structures should not have occurred under static load conditions.
Accordingly, Prof. Corte.n's case is not acceptabla tc Sun Ship consultants, particularly for structures that are not stress relieved.
Redundancy has nothing to do with Prof. Corten's calculations.
1 In brief, Prof. Corten's calculations must be credible for any condition of relative structural redundancy. Since they are concerned with local regions of the struc ture, they must utand the test of examination in terms of general service experience involving fracture.
The argument of unique redundant features cannot be used for these calculations.
Collectively, the Sun Ship consultants have addrected the question of requirements for establishing structural reliability based on frac ture properties. Their opinions clearly are thr.t it is not possible to certify the structure based on what was, or is now known of the fracture properties of the steels involved.
By basing its case so strongly on redundancy, VEPCC has apparently arrived at the same conclusion.
Closing Note This review is addressed only to the major issues, so as not to complicate this important aspect by details.
f
[
ad
.lilliam S. Fellini j
a
e 4-i 22 STATEMENT OF RICHARD ROBERTS
.j July 1, 1976
.I would like to offer the following comments relative to the VEPCO June lI, 1976 response to NRC.
f COVER LETTER 11EM i:
No code existed for the initial design and repair.
However, it is the designers responsibility to provide a structural design which can be reasonably fabricated and safely utilized.
Even in the presence of a code and a design which meets all code requirements, when an engineer is aware of situations or facts which render his design inadequate it is his responsibility to provide a design which will function safely. One can not hide behind a code and state, "well we meet alI code requirements". Codes were not and are not in-tended to provide refuge for a design when situations exist which require consid' ration of things not covered by code e
rules.
COVER LETTER ITEM 2:
The AASHTO requirements are based on many items not present in the North Anne Supports.
The design loading rates are well documented for bridges.
The primary tension.T. embers of bridges are normally loaded in a very simple manner pro-ducing a one dimensional tensile field.
The prin.ary tension member in bridges are basically fatigua limited.
The AASHTO
. - - - _ - - - - _ - _ - _ - _. _ _ _ _., _ _. - _ _ _ _. _. _ _ - _ _ _ ~
23 l
specification as conceived is based also on tests of large l
welded bridge details simulating real' bridge behavior. The North Anna Supports are made of large heavy shapes joined through the use of large welds. The state of stress at criti-gal locations are exceptionally complex containing axial loads, torques and bending moments. No experience is avail-able on the performance of such configurations.
In particular, no fracture experience or fracture mechanics based methods are available for the estirr.ation of the f racture response of such complex load combinations.
Although there are many similarities between the sup-ports and bridges, there are enough differences and unknowns to warrant a greater degree of conservatism than used for bridge design.
It is very reasonable to design the supports to criteria similar to those employed for the vessels they support.
COVER LETTER ITEM 3:
At this point VEPCO is invoking the fact that for cer-tain chemistries, certain generic responses are predictable.
If they so desire to use generic results, then one should design for the worst generic case in the absence of complete data.
PAGE 4, ENCLOSURE I-VEPCO comment that f racture toughness requirements for structures such as the North Anna supports or bridges are
. not af fected by a f atigue argument shows a lack of understanding 2.
1 l
l
p.
24 of how the AASHTO specifications were derived.
Fatigue does play.a major role as it is the primary source of cracks in bridges.
If the cyclic life is !smited to a number of cycles which produces very smalI crack then low toughness requirements c,an be employed. This is one of the key premises of the AASHTO specifications.
PARAGRAPH 3, PAGE 2, ENCLOSURE 1:
Paragraph 3 on page 2 of enclosure I shows either a lack of understanding of the application of the AASHTO specifications-or an attempt to confuse the issue. The use of a 45*F tempera-ture shift is not conservative.
In fact for the maximum measured yleid strength of the A36 material, 46.5 KSI, the temperature shift for static to dynamic loading is 145'F not 160*F as initia!!y proposed by Corten. This produces a tempera-ture shift between the postulated loading rate for the North Anna supports and dynamic loading of 40*F not 45"F.
This value when combined with the AASHTO requirements of non plane-strain behavior at the design loading rate and the lowest expected service temperature give a CVN requirement of 15 f t.-ib.
at 70'F.
This comes about from the fact that the point of non plane-strain behavior for A36 material was estimated for the AASHTO purposes as occurring 50'F above the 15 ft.-lb. CVN temperature. Thus taking the service temperature, 80'F, sub-l tracting the 50*F estimate for non plane-strain behavior and I
, then adding back the 40 F temperature shif t for loading rate l
_ _ _ _ _ ________________ a
l 1
l!.
25 effects produces a requirement of 15 ft.-lb. at +70*F for the
~
supports if their CVN requirements are set up consistently l
with the AASHTO requirements.
The above comments set out the support requirements on a basis consistent with the AASHTO specifications. However, as i
already indicated in my comments on. item 2 of the VEPCO response, it is my opinion that the AASHTO specifications should not be applied to the sup' port structures as there are enough differences in general de-sign, f abrication and expected service as compared to bridges.
V 7 i/76
(
l i
1 6
l 4
t
---_-__.__-_-.-----_-l__
____,_.___--_____--m 4
e i
f a
o k
l 6
i e
P 9'
b e
O TABLES I
i TABLE 1
SUMMARY
STATISTICAL STUDY VEPCO CHARPY DATA REFERENCE TABLE 5.65.3-1 STATISTICAL PARAMETERS DATA
_ RANGE MEAN STD. DEVIATION y
~
Longitudinal Specimens (24 discrete data points) at 80*F Heat W96-Ft-lbs 19-81 47 17.8 Lat.Exp.
23-66 43.6 12.8
% Shear-20-80 43.8 16.4 Longitudinal Specimens (27 discrete data points) at 125'F Heat W96 Ft-lbs32-104 72 18.2 Lat.Exp.
35-81 61 10.8
%, Shear 30-90 70 14.4 Short Transverse Specimens (27 discrete data points) at'80'F Heat W96 Ft-lbs 8-22 15 3.8 Lat.Exp.
11-26 18.3 4.0
% Shear 10-30 14 8.0 4
Short Transverse Specimens (24 discrete data points) at 125 'F Heat W96 Ft-lbs 15-35 23.8 4.7
~
Lat.Exp.
20-39 28.6 4.4 j
% shear 10-50 28.8 13.3 l
O
~
TABLE 2 HEATS IN STEAM GENERATOR SUPPOR_T STRUCTURES VEPCO SUN SHIP (l)
TEST EST.
NDTT NDTT HEATS MATERIAL
'F
- F 171C560 A-572 40 182C087 A-572 50 172C497 A-572 60 123C349 A-572 40 171C871 A-36 30 171C866 A-36 40 30 182C174 A-36 30 182C172 A-36 40 182C159 A-36 40 182C178 A-36 40 182C535 A-36 40 30 181C686 A-36 20 182C150 A-36 50 172C586 A-36 517J1106 A-572 517J1058 A-572 182C090 A-36 172C963 A-36 182C156 A-36 182C154 A-36 (1)
J.
Durant Emperical Estimate
' - ~ - -
)
TABLE 3 CHARPY RESULTS - A-5 7-2 STEEL (S&W TESTS)
TEST TEST LAT EXP, PCT.
1 ORIENTATION TEMP.,P FT-LB MILS SHEAR Longitudinal 0
7 4
1 0
3.5 3
1 32 6.5 8
3 32 8.5 6
6 10 120 30 28 34 120 25 27 34 160 39 38 43 160 33.5 36 52 212 79 70 90 212 82 68 90 Thru-Thickness 0
2.5 3
1 0
2.5 3
1 32 5
5 3
32 3.5 2
3 RT) 5.5 6
10 RT)70 5.0 6
10 120 8
10 30 120 8
10 33 160 12 14 42 160 15 18 55 212 21 26 89 212 16.5 20 100 O
- - - - - - - - - - - - - - - - - - - - - ~ - - - ' - - - - - - - - - ---- - - - - - - - ~ ~ ~ ~ - - - - ~ ~ ~ ~ '
t TABLE 4 MATERIAL CHARACTERISTICS OF STEAM GENERATOR SUPPORT STRUCTURES
% SHEAR @ 80*F MEAN APPROX EST.(2)
MEASURED ( }(4)
(3)
STD.
HEAT NO.
WT.% USED II)
TYPE NDTT
- F NDTT
- F EST. MEASURED DEVIATION 717C560 9.2 A-572 40 43 182C087 2.1 50 36 172C497 1.5 29 17(5) 60 123C349 3.0 40 43 171C871 23.1 A-36 30 50 171C866 17.9 30 40 50 44 16(6) 182C174 6.7 30 50 182C172 11.6 40 43 182C159 3.0 40 43 182C178 9.4 40 43 181C535 3.5 30 40 50 57 10 I) 181C686 1.0 20 57 182C150 8.0 50 36 100.0 (1)
Based on original fabrication (2)
J. Durant - Sun Ship - Empirical Method Based on Chemical composition (3)
Vepco (4)
See Appendix (5)
Assumes A-572 Sample of DC-81 is from worst heat (6) 24 Samples - Table 1 (7) 6 Samples - Table 1 O
N
____ _ ---- - - - - - - - - - - - ' - - - - - - - - - - - - - - - - - - - ~ ~ - - - - - ~ ~ ~ - ~
y*
TABLE 5 l
COMPARISONS OF VARIOUS CODES OR PRACTICE l
USED TO CONTROL IMPACT PROPERTIES IN CRITICAL SERVICES I
i CODE REQUIREMENT EQUIVALENT % SHEAR @ 80*F III German 28 ft-lb @ 68'F 36% Sheaf 1
II)
France 20-29 ft-lb @ 32*F 55-68% Shear III Japan 20-35 ft-lb @ 3 2*F 55-76%' Shear Ship Structure Practice I2) l Primary' Members NDTT + 32*F Min 37% Shear Secondary Members NDTT + 12 *F Min 23% Shear Lloyds Ship Structure Min 30% Shear ASME Section III - G-2000 NDTT + 60*F Min 60% Shear Subsection NF II 25 Mil Min.
Min 36% shear (1) John Harrison - BWI (2) Fracture-Control Guidelines for Welded Steel Ship Hulls - 1974 (3) ASME Section III l
e e
e
~
t e
a t
e m'
b e
FIGURES 0
4
ecne d
l if e nv e
o e c
CL ne n
d a
5 if e
nl M
oe Cv e
%L 59 N
S O E
I T
IT C R
E E
R P
DI U
PM T
I I
S G 1 U N O O E
I R R L U A S G V TR F N O I
E P
E P WU T S E R B
O P
T HI A
S RE N N O
E G
I TAL M 6
T E A 3
T R E 6D T
A 3 N[y S
s At ta s
e e
H T
C 2
I T
r S
o A
1 IP N
I T
R C
C I
A I
T S
T S
N I
A S
A A
L L
1 E
P E
N A
L P
0 0
0 0
0 0
0 0
0 0
0 9
8 7
6 5
4 3
2 1
1
- u65,9 y E '30
[S1U nO ui'ZO_y4,d g53n E
)
i
~
e c
n
,m e
d m
l i
f e e
r v oe cn CL e
A d
i fnl oe A
Cv e
0
%L 4
a 5
1 9
N d
S O s
g E
I 5
I T
.3 T C 1
R 8
E E P R n
I O D RP LA C N J
M s
T i
I A D MT 5
P U
S G
]
I I
a 2 U N F
OO E
4 I
R R L e
U A S I
4 G V TR
- F N O I
j d
E 0 R E P U
s r
8 E P T
WU A
T S R
E R E
O P
B P
M T
H A f
I E
e T
S R NE OE N
a I
G TA M
0 L
{
E A
,T T
5 R E T
f T
T-6D 3
S TD N
N At a
se 6
e T
G g
C n
A T
a 0
I t
S R
7 3
n A
o L
ts W
E P
N g,
C C
I I
A t
I I
s T
T R
E S
S T
A A
i f
S g'
L L
]
P E
E i 0 4,
N 1
A
~
~
~,
~
P t
~
L 0
0 0
0 0
0 0
0 0
0 i0 0
9 8
7 0
5 4
3 2
1 1
c h
- gE*.g y E '>. 0
~ 'IUf v 5 'ZO,y4$" "b~ $ a X c
]
f 9
6.-
.+. s
. p
.yn6."*O;y,%m.
y m
)
s a.
7
. T f.
' ;:.Q.,. p U
'e.
r.
-., ~.
., c..c f e ygg e-c j, m' Oy;gh+1:
n
, 1:"
f 7
- n. s - 4.dg
. S A:a.
- c. :
gh,. w.
..N;.* Q '
._.6
- d x
4 i
mt s
- ' - u r
. i*
e.-
.,{
Cg n'
. _ wt a
a t
c m
.r e
n
- 1,( F '
o.w,*p [] w a
~
%2 i
I -
d
..~
ji
%;. v -
i f
n'
- r. w*
[.._
! ~ 2~J o*
.n
- i
- .
C **
'vp g. e: }.
d s gr.we.
9 5
4.-
_.~r
,y;
~
3
~
~
N f h; a.'*.~
y,i s
g y: h S
O<
i
,f E
I I
T C
T C wy nE 4
a kg R
E E
s ya P R 4
J I
O D o P L w s
R
- ^
z f.
e
/
A t
T o
i C N r
e I
A D p X
P U m M
g T e i
T I
I
'S G e U Nc
$,.A'.. i 3
i O Orv E
,.{N.,
I L e R
t
_U A S S- '
j l
g G V T R
g iI, C
NO j
I F
7s E P f
E P 4
WUS u^
g T
. gc '
E R - -
i B
O-F 5
P T
H A S R g-OE
,n E
N N
- s
,p t
lT ;
G i
'T 2
I
+
T D y' A M A
3 J
.NP,'a -
jT5 L
6 C'E R E 7
' 3'f.
T T
A s
T 5-E, s
_N Yid.
S s
,'N A,
A t
H a
a e
t o
r 4
ic '
0 e
n 2
H o
0 s
A 1
'a T3 r
1 s
e o
u
/
t'~
s r
1 o
/., d 4
P W - % G-
~
,i
~
~c-t l
a A
u.*
h R
t I
e s
T se-T s y' - -
E S
0 A
A*'
S
- f. y ~..
p
%P
.A L:
t E
i
~
N 7 *, rJ ~
1 f:
g wp.-
- P I
7.,
0 0
0 0
o 0
0 o
0 o
a 0
9 ~
8 7
s 5
4 3
t 1
g-7 g
G.g e
9 y
6.,
i.y g f:-5eX
~
yf
\\)I
s e
a O
~
FIGURE 4 STEAM GENERATOR-LOWER SUPPORT O
A -572 i
MAX. STRESS
[*
~'
38,000 PSI y
~
,k?,N, a-m
/
-y et s
i,-
~
\\
NM,
~.
/
Nh
/
,. /
/
bh\\
kh u!
s.
\\
sr
/
-5 72 A
/
q i
u/
m J/(y
\\
/,//-
I s
e s
\\N.
o!
~
O 9
D
f.
3 l
~
a 85 s
's 4
~
l
\\
Y t
gm n
Z
?
N
\\
85 u
\\
s
'o s
6 b
R' i
\\
~
Y C6
\\
C*
\\
Q "E
8o l
r r
i e
C
\\
i d
w
{w c$
tg 5
83 m
l9
<B ss o
i s
s' s
1 e
-5 gs 5
i i
a l
.s S-s I
-5 g
Z g
l a
m e-
\\
z 5
g-Ry
\\
\\s a
j v
vu E
EE
's
- y. o w 0
I 5
55
~
i' T
?"
,\\,
P, d l
8 8
R S
a e
a a
e I
+
TTD N
N 00 I
0 M
2 E
G S
E N
l i
l A
N P
0 f
O A
0 V
/
I J
1 T
)
C E
H S
C 0
9 o
N ER M
p F
A 0
E 8
TS a,
FO.
0
'7 F
SE0 I
8 E
S "y
T N
N C
0 O RT h
EA I
' 6 g
MR G
I T
PO 5, A
g 3
R R l
N N A
R T
6 P
R E
O g
Y A N P
0 N
'5 E
T P R
CP MO E
E R
G R
I U
AU Sr E T PS GC T
M MR C
0 I
S 0,
4 A
E d's F
O I
E k
B R
T DT t
U T
S EA p
S S
T R AE I
/
0 N P
3 I
MN f
H L
I I E T
S A
/
N I
R E
0 I
/
AI E
2 R
T
/
T A
C CC S
M
/
I I !
E T
TT 0
F N
S SS O
1 A
AA A
L LL
/
L P
t EP
= -
=
' [
P 0
0 0
0 0
0 0
0 0
0 9
B 7
6 5-3 1
1 O
1 i-
ns S
8Ai d
Ch e E
S
'~
D a s Q'
t U
E
^-
e u L
n n
iNl o w
A e
d mtet D
A e
aI s os v
c r
uf mi
/a
+-
r A
s t ob
]
e or a pN F f l
f 2
.r m67 a 3-5 S
a
~
/
- N E
A
/
M E
E E 2
I 7
T H
/5 B
p A E S
E G
A C
_,6 T I
A
/
YR E V 3R R
O E
C E I
A V
/
fA S
P A
iM2 P
u N E
/
dM I
U E
d 6
n S
/3 I
P A
pO N&
/
j uB Ol yRO F
j I
AA TN p
A N D
A
,T
/
8 ja M
H S T
/
E R T O
/
N
/
T I
A
^
7 C
l P
f yI M
/
7
/
I E
y RU 7
G I
- I F
p h
4 7
0 0
0 0
0 0
8 7
6
}
4 3
1
- gz ra dxsa 35 s
4.
s 4.. 4.-
9 4
e t
g
.E
~
@n }-
T o
a X4 E
8 zg g
oGh h = "ss z
O zm x
m g
x
=
Z 4
xU a
- E f
x s
=g g
w m%I
"=
J j
I X
g w
g.
g3,0 N
i X
=
. o E
em2 s.
x-cg5 N
4 n
M N
I w-o c
y
)
- c Nh
\\ 4p,&
I k
5+Q.
i=
"k.
?
Em e
s
_=
w m
- s
'E N
8 5
z.g___________s,_
^
g.
i
=2 g
p g h~
N 1
N.
- 5 m
,1 P NY s
e i
i
-s e
a e
e a
e a
a e
- 4 E
(
m x
o e
em N
o V
O 3
s go R
to A
I
=
m
- =
m b
o x
t R
G 1
= 2
=
8 ox o
3
)
m N
O_ g o
a a
"o E
8 3
Z w
r o x e
t
=
2
(
B
,o w
r r
I;2
~5 w
s g
n R
2
= a <
o u
e a
e i
e t
i e
i e
i g
3 R
8 R
S 8
?
8 2
2 01-1d 'ACH3N3 AdEYHD = 0 511W 'NOISNYdX31VH31V1 =JC
s.. '
la ~
~
FIGURE 10
~
FRACTURE CONTROL GUIDELINES FOR WELDED STEEL SHIP HULLS 1974 I
LEGEND o Plates from hulls that failed in service x Plates from hulls with boderline peefoemonce e Plates from hulls with successful performance E
R100-S 90 w 80 SAMPLES FROM STEAM 73 GENERATOR SUPPORTS U
AA-36 6 60 a
i 6 A-572 I
x 6 50
[] A -572-flonge k
40 g" =
.cI g A-572-Web 11:
- hroo : :! x I
20-
= m m.e:
i j
1 0. CT..,if o 4
. _i _.
e U
0 10 20 30 40 50 60 70 80 90 CHARPY FT-LB AT CASUALTYTEMPERATURE l
q
)
1 8
C 9
D 4
(
3C ta 3
e 2
H 1
t n
a 0
w e
4 R
o H
2 n
knU 0
j j
a 2 2
+
0 I
0 2
tae 0
a 8 H
1 n
wo 50*
0 n
n e
o I
6 S 2 k
w 1
7 n
ot T
a R 5 U
n e O-nH k
A 0
P U
I 4
PF 9
1 UO 4
l1 S
3 1
S C
Ot
+%
R E 1
OI 3
0 I
=+*
E T
2 2
R TR 1
1 U
AE t
RP a
G FI EO e
I 0
NR H
0 EP 1
G N 1
MW 8
0 AO Cxo*
+
~
I 8
ETN D
SK e
y 0
I g
6 r
pe t.
x n EE y%
0 i
I pr 4
r r a e a e t
h h LCS
(
0 t
{)
2 0
0 0
0 0
0 0
0 0
0' 0
0 9
6 7
6 5
3 2
1 1
I'
WH OS L
L AM R
O N
~
L E
7 E
TS E
p C
N AM
/
R O
F R
2 E
1 P
F Y
E E
G L E R
R U
R E C U
G V N T
T E
I N
E.
T A
A I
F E
t D
R W
D M E
N P
E R O
I O
M R
ee S F E
L F
E R T
t E
t D
O i
P m
E m
F L
o O
P C
M e
A r4
/
X u7 t
E c9 u1 r
t S
hip S
EC RU O
S L
^
_ Uzh8EE u.O $Da J
)
N)
E C
I I
CCON L
)
E T T ITTTO I
I E NL S
C A
SS I
GI NI T U
AA AT S
LPD LLNE A
I R
L R
(
P S(
R T
(
1(
2 67 3.
AA g&
0 I
3 1
N O
0 1
2 T
1 C
R ER A
0 E
DI 1
1 H
1 S
L A
D N
>=
0 N
0 1
I A
D 1
U Y
i T
P I
x x
R G
0 i
A N
z 9
H r
3 C O
1 L
==
=
N x
s S
a
==
0 B
i E
T E
8 L
E R
r RU W O
T a
F T
P a
G
=
=
E P
0 I
I B
U 7 H
=
i F
P S
C I
=
T H
R O
S O
r x
0 N
T 6 N A
O V
R
=
I E
=
Y T
N x
=
=
0 A.
P i
E x
5 R
El G
=
A R
H M
C A
0
=
I E
a 4
=
T
=
S a
i0 x8 3
s n
=
0 l
u 2
a 6 1
a L
O 0
0 0
0 0
0 0
0 0
0 9
8 7
6 4
3 2
1 1
R
%A EH S
8 e
i i
9 APPENDIX 9
RETYPED FOR CLARITY Copy to:
APStakutis WFNeilon g_1 SRagone-3-AM RBBradbury-AM GLockyer (enc. )
(Att:
WCSpencer)
RCoupland-AM RJSpahl SVLowry-AM GJBurroughs MScheibner (enc. )
JLPerkins-2-AM (enc. )
MASalvi HWSorensen-AM JVHarrison-AM EEErlandson-4-AM JRCavallo/ Job Book (ene.)
General Files
{
JRCavallo (enc. )
Mr. Stanley Ragone June 24, 1974 Senior Vice President - Power Attention Mr. W.
C. Spencer J.O.Mos. Il7) V Supervisor - Nuclear Engineering 1201; Virginia Electric and Power Company NAS-6712 P.O.
Box 26666 Richmond, Virginia 23261
Dear Sir:
i l
PURCHASE ORDERS NOS. NA-177 AND NA-ll77
{
REPAIR OF PRESSURIZER SUPPORTS l
ULTRASONIC EXAMINATION NORTH ANNA POWER STATION 1975 EXTENSION - NORTH ANNA POWER STATION Enclosed, for your review and comment, is one copy of P.X. Engineering Company, Inc., report of ultrasonic examination of the circumferential flange-to-web welds on the Units 1 and 2 pressurizer supports, with supplemental summary charts prepared by S&W.
S&W's QA-NDT Division specialists have reviewed the enclosed report and determined that the weld portions removed and replaced by P.X. contain no reportable indications.
Ultrasonic examination of ti.e original weld metal portion of the flange-l to-web welds revealed reportable indications above DAC, specifically in the Units 1 and 2 bottom flange welds and the Unit 2 top flange weld.
l These indications can be characterized as intermittent in length, with i
no appreciable cross-sectional area (as evidenced by ultrasonic detect-ability primarily from only one of three search angles).
A typical 9 in. length of weld on the Unit 2 pressurizer support was excavated by P.X. at S&W's direction, and it was determined that the ultrasonic reflectors present consisted of intermittent, extremely small deposits l
of trapped welding flux.
The results of this investigation were l
reviewed with your consultants, SwRI.
To determine the acceptability of these flux inclusions, the Engineers have eviewed the stress design calculations applicable to the press-l i
urizer supports and have determined that the absence of appreciable l
defect cross-section area, coupled with the low design load (approx-imately 9 ksi), has essentially no effect on the structural integrity of the pressurizer support under critical design (faulted) conditions.
~
(
A-2 Additionally, to ensure that such flux inclusions would not propagate under load, the Engineers have performed a fracture mechanics review and have noted the following:
1.
The operating average temperature of the pressurizer support ranges from approximately 100 F to 125 F.
2.
The minimum impact properties of the ASTM A516 Grade 70 material from which the support is fabricated is 15 ft-lb at -30 F, with an NDT7 of -20 F for the 3 in. thick material.
3.
Employment of the NDTT +60 F design criterion reveals that the support can be stressed up to yield strength without danger of brittle fracture and without danger of propagation of any inclusions or flaws of any size.
Therefore, assuming
-30 F + 60 F = +30 F 4 100 F (minimum operating temperature),
fracture mechanics analysis is not required.
Based on the above, we believe that no further examination of the pressurizer support repair is indicated, with the exception of final visual and dimensional inspections required prior to shipment.
Mr. C.
E.
Bingham of VEPCO has indicated that SwRI will be required to make an additional trip to P.X.
for review of ultrasonic examinations performed, prior to S&W instructing P.X. to sandblast, paint, and prepare the pressurizer supports for shipment to the jobsite.
Please advise us as soon as possible when and if this visit will be scheduled.
Please advise us by July 1, 1974 concerning your concurrence or comments on the above.
Should you require additional information, please advise us accordingly.
Very truly yours, j
A.
P.
Stakutis l
Project Engineer l
Enclosure i
JRC:rfd
t 3,
\\'
f g
,JSpahl (At.t : 1:CSpenect)
RCoupland-A,l r.
SVI.oit ry-N !
/
GJDurroughs Micheibner(ene.) a-3 Jl.h rkir.:; Al'(ene. )/ '
- -!ASalvi IfISorennon-AM JVI'a rrison-A!4 li!:Eri nad: ton- '-N I JITav.tllo/ Job Book (enc.)
General Files JRCavallo(ene.)
1 June 2 4.1974 4r. SLacicy.ci. pac 5
Scaior Vice 1re' 1.ient Powcr At t ention.'ir.
'.!. C. Spencer J.O. fos.11715 Superviser - :htelear F.r.gineering 12050 Virginia 1:!cetric and Pauer Coac..tny NAS-6712 P. O.
!'o.v. 26J56 Iticinond, Vira, inia 25261 I
Donr Sir.
Ping!AS: ORn2P.S :;0~. NA-177 MlD !!A-1177 REPAIn OF PP'.SS!.MI..b:t SUP?C1TS ULTRAS 0:!IC :::W:I::A'iIT!
t!ORTii K :!A P,"5R S'i ATIRI 1975 P.'T::'?iTO.! - :: ):'.T.: XNA P't!".R STATIO:!
Er.clo:::.', fer ye::: :: tic:.' an! c::.m7 cat, is one cepy of P.X. En:;i eering Co..pany, Inc., ;Nort of ultrasonic c:'.a :iration of the cirewJ:wntial flana.u-to-i.'ub ucl.!s on the Units 1 and 2 pressuri:vr supports, witit su;.pleuental su c.ary charts prepared by Efdl.
Srdi's QA-::DT !'ivision specialists havn revie.ted the enclosed report and deterniac.1 that the wold y.ortions rctiovel anil replaced by P.X. contain no relortabic in lications.
Ultrasoni: examination of the ori' tin 21 veld netal portion of the flan 2e-to-vcb valds revea:cil reportabic indications above DAC, specifically in the Units 1 and 2 bottom fl.ax.:o ucids a.td tho.Atit 2 top flant,o wold.
Ther,e indic; tion:t can be c'isracteri:cd as intermittent in length, with no 7 appreci't,1c crocr-sectional area (as cvidenec.! by ultrar.onic dctcciability priciarily freu only one of tiirce scarch aanle::).
A typical 9 in. length of wel: ca the ti..it 2 pressuri:cr s.:pport tas e:cavateel by P.X. at S.'/.;'s direc-tion, and it was 2.? terr.iaed taat the ultranonic retie: tors prescat cr$:tsisted of intr raitt e it, c:.trnely sr.all deposits of trapne.1 e:1Jian flux.
The resnits of this iriver.tination were rcviewed with your co tsultants, Su!:I.
To deternine the acce,t.eility of these il.r: inciasions, the Enc.it.cers havo reric't:d th c:.rt r.:;.rti 'n calculation. annlic.Sle to the pic:curi.er :n:pperts and i:nve sh.t cr ti':r'il that t.:e ai:';cac? of appreciah10 deicct cross-sectio:: are.1, ervi,il. d e dia :...e len
- i
- t load (arr...r::nat ely 9 !. :1), has e'ss 'nially r.c ca cet on ra.. st rue: ural i..t e 'rity o..' the pres.ariner su. pert under critical o r. i s. : (i.i ! t e.i) c. a 'it ions.
Y'sl -.*A./
d ?c.'.!A L.,7'. ?j. *g 'f,,,g Z l jq 4250uh u
y c5DN s
..iditionalry, to casure t!.
such flux inc1't.tions we.... not propagate under 1
, ' loc 1, the !!reti.ucers hav
- performed a frn:ture mech:,nies r:vic'.i und have r.oted A-4 th0 follwing:
1.
'The operatit:9. avers;;c tr.r.tperature of the pressurizer support ranys fro.i spiercxiaately 130 F to 125 F.
2 l'au niniinn inpac:: nropertich of the NJi:! A516 Grade 70 natorial fro 1 t/.tici tne suppert is fabricated is 15 ft-lb at
-30 F, stith na !$';T of -20 F for the 3 in. thich natorial, t
3.
Puploy.umt of the fiDIT +60 F d :sig. criterion revc115 t! st the 1,upport cit.: be stressed up to yicist stronnch trithout danger of brittlo '~ractura nad withoat den; r of proca?,wien of any ir.clu-sions or 11aus of any si:c. 'ihereferu, assumin;; -33 F + 60 F =
+30 F < 103 F (:ainicou operating te.aperatura), fracture r:cchnaics.
analysis is not required.
i)f* ' 4)
P Based on the above, ue believe that no further ex..uination of the pressurizer support repair is ir.dicated, uith the o.:ceptic.n cf final visuni and dimensional 4
inspectio.w ree.uired prior to shiperst.
?!r. C. E. Binghan of VEPC0 h.1s indicated,/
that Suui irill be re-t ired to nnko en c:!ditional._tr.i.'_ t.o.P.X. f. or.revie'.t. oi. ult.ra_-
i sonic exa.ainations perior.:.:d, prior. to SS.' inst actiIg P.X. to ssudolnst, plaint, and pre;nre the pressurizer supports for shirnent to the je.bsite.
Pleaso advise us as soon as pos.;iblo when and if this visit vill be scheduled.
Picase adviso ur by July 1,197/.
concernin; your coricurrence or concents on the above. Should you rcruire additional in'fornation, pienso advise us accordingly.
i Vory truly yours, A. P. Ste.kutis Project Enginocr Enclosuro J!tC:rfd i
l.
l i
(
^~
APPENDIX METHOD OF CONVERTING IMPACT CODES TO EQUIVALENT SHEAR @ 80*P For estimation purposes an energy-fracture appearance transit.lon curve could be defined with the following parameters:
A.
At NDTT = 15 ft-lbs = 15% Shear B.
At NDTT + 5 0
- F = 5 0 f t-lbs = 5 0% shea r C.
At NDTT + 100*F = 100 ft-lbs - 100% Shear 1 ( ! 15 error of estimate 7
/
100 =
/
,, /
/
/
y 80
/
,/
=
/
/
y~
m
[
60 -
g f
/
~
50
/
Vepco Data w
/
/
X Energy 0
/
0 % Shear 40 =
p j
3
's'
/
}>
20
/
=
f.
s V
l l
+ 50
+ 100
% Shear or Energy given by For NDTT 4. T
,Z. NDTT + 50*F T-NDTT 35
+
15
=
50 ForNDTT+50*F1.T.7_NDTT+100*F=T-[NDTT+50*F50)+50 50
/
S)
J 4
A-6 For various criterion one can estimate, given the energy requirement, the equivalent shear requirement.
GERMAN CODE Given 28 ft-lbs. at 68'F 28 = 68 - NDTT (35) + 15 50 fl3 = p8 (35) -NDTT (35) 50 50 13 = 47.6
.7(NDTT)
-34.6 =
.7(NDTT)
NDTT T 50 13 (35) + 15 = 27.6 50 then 90 = NDTT + 30 i
So at 80
% Shear or Energy = 20 (35) + 15 50
= 36% Shear or 36 ft-lbs at 80*F j
i i
e 1
l l
I l
l
(
A-7 s
FRENCH CODE 20-29 f t-lbs. - 3 2 *F 20 ft-lbs.
20 = 32 - NDTT'(35) + 15 50 5 = 22.4 -.7 NDTT I-17. 4 =. 7 NDTT NDTT T 25'F Therefore 80 *F = NDTT + 55'F
@ 80
% Shear or Energy = _5 (50) + 50 = 55% Shear or 55 ft-lbs.
50 29 ft-lbs.
29 = 32 - NDTT (35) + 15 50 14 = 22.4
.7 NDTT 8.4 =.7-NDTT NDTT if+12 Therefore 80'F = NDTT + 68'F
@ 80
% Shear or Energy = J8 (50) + 50 = 68% Shear or 68 ft-lbs.
50 i
9
t A-8 JAPANESE CODE 20 - 35 ft-lbs at 32*F 20*F same is done before equivalent to 55% Shear or 55 ft-lbs.
T 35 ft-lbs. at 32'F 35 = 32 - NDT (35) + 15 50 20 = 22.4 -.7NDTT
-2.4 = -. 7 (NDTT) 4 T NDTT Therefore 80*F = NDTT + 76*F l
Then @ 80
% Shear or Energy = 25 (50) + 50 50
= 26 + 50 = 76% Shear or 76 ft-lbs.
i e
4 0
l
(
A-9 G-2000
{
NDTT + 60 =
% Shear =
10 (50)
+
50
= 60% Shear 30 SUBSECTION NF 25 mil lateral expansion if one assumes 15 mils at NDT and 35 at NDTT plus 60*F Then 25 mils = NDTT + 30 T 30 (35)
+ 15 56 T 36% Shear l
l
1
{
l APPENDIX A-10 REBUTTAL TO VEPCO COMMENTS - JUNE 11, 1976 A few of the comments made by Vepco in their letter of June 11, 1976 to NRC will be briefly addressed where they haven't already been covered.
Vepco mentions that their consultant SWRI did not ifind toe cracks in the welds.
We refer the reader to the photographs in Appended Section 1 of Sun Ship's May 20 presentation.
Actual photographs of numerous toe indications are shown.
In addition, several of Sun Ship's employees have observed toe cracks on the job site.
Vepco says "The literature does not substantiate the inference of the Sun consultants that either the weld metal or heat affected zone (HAZ) of the material in the supports would represent a safety concern."
We suggest they read Subsection NF of the ASME Section III which requires testing of l
these areas if fracture toughness is of concern.
Also, they are referred to the article, "The Practical Application of Fracture Tests to Prevent Service I ailure," F. M. Burdekin, which appears in the Appendix to Sun Ship's May 20 presentation, and to the " Fracture-Control Guidelines for Welded Steel Ship Hulls", 1974.
Vepco and Stone & Webster submit an affidavit from Charles B. Miczek concerning hearing the structures.
To properly evaluate what reliability should be placed on Mr. Miczek's expertise, we suggest anyone interested read the deposition Sun Ship took from him under oath.
Because of its volume we have not reproduced it here but will gladly produce it in its entirety if anyone requests it.
A few selected quotes, however, are:
Q.
Did you have any understanding as to what it was that you were supposed to do as management over-viewer with respect to the supports?
A.
Well, perhaps I assumed that --- I assumed my own responsibilities.
O.
What did you assume them to be?
A.
Very minimal; really a window-dressing kind of a guy.
g l
l-A-11 Q.
Are you familiar with the phrase, " critical flaw size" in connection with fracture mechanics analysis?
A.
I'm not a fracture mechanics expert.
These are just words to me.
I don't know what they mean.
I e
i
t
(
, em.. a v.
3 12 SU N ' SHIP BUILDING & ORY DO CK CO.
INTER. OFFICE CORRESPONDENCE SHEET CHARGH N0'n. 50104 A S0109 STONE & WEBSTER (VEPCO) SUPPORTS suaJtcT RECORD OF MATERIAL FROM BETHLEHEM STEEL CO.
onre
. Trine 29, JPN VROM D. Rhodes To P. S. Hepp Material Number Heat No.
Ship Date Size of Pieces Weicht Subtotals 171c560 4/9/71 WFl4-605 4
29.746 4/9/71 5
45,627 3/18/71 2
19,095 3/13/71 1
18,427 112,895#
182c087 4/23/71 WF14-605 1
7,436 4/20/71
. ~
1 18,427 25,863#
172c497 4/23/71 WF14-605 1
18,427 18,427#
123c349 5/16/71 WF14-605 1
18,427 5/8/71 1
18,427 36.854 171c871
?
WF14-426 3
27,158 4/23/71 1
9,053 3/26/71 2
35,784 3/24/71 2
40,044 4/3/71 1
20,022 3/26/71 2
32,802 3/29/71 2
30,672 4/9/71 5
26,311
~4/10/71 1
8,005
t A-13 CHARGE N0's. 50104 & S0109 STONE & WEBSTER (VEPCO) SUPPORTS Fage :
RECORD OF MATERIAL FROM BETHLEHEM STEEE CC.
Material Number Heat No_
Ship Date Size of Pieces Weicht Subtotals 171c871 4/9/71 WF14-426 3
18,6.28
)
4/3/71 1
12,780 4 16/71 2
21,335 282,594 171c866 4/3/71 WF14-426 1
21,726 3/26/71 2
43,452 3/26/71 2
40,044 3/26/71 1
20,002 3/26/71 1
12,780 3/29/71 1
12,780 4/9/71 3
13,898 4/10/71 1
8,005 4/9/71 3
25,054 3/27/71 2
21,335 219,076 182c174 4/3/71 WF14-426 1
21,726 4/17/71 1
14,430 4/17/71 1
5,236 5/1/71 1
14,431 5/8/71 1
4,791 5/8/71 1
8,005 4/3/71 1
12,780 81,399 182c172 4/23/71 WF14-426 1
14,431 5/29/71 1
5,866 4/30/71 1
5,866 4/17/71 1
14,431
__ ___ _ _ _ - - - - - - - - - - - - - - - - - ~ - - - - - -
A-14 CHARGE N0'er 60104 & 50109 Page 3 STONE & WEBSTER (VEPCO) SUPPORTS RECORD OF MATERIAL FROM BETHLEHEM STEEL CO.
Material Nwnber Heat No.
Ship Date Size of Pieces Weight Stbtotalo 182cl72 4/17/71 WF14-426 2
28,860 4/21/71 2
28,860 5/5/71 1
13,330 4/17/71 4
30,672 142,316 182c159 4/17/71 WF14-426 1
17,892 4/24/71 2
9,266 5/8/71 2
9,266 36,424 l
182c178 4/23/71 WF14-426 2
27,761 4/20/71 2
26,661 5/1/71 1
5,866 5/5/71 1
5,236 4/21/71 1
14,430 4/24/71 4
20,945 5/8/71 i
14,431 115,330 l
182c535 4/22/71 WF14-426 2
43,452 43,452 181c686 3/23/71 WF14-142 2
11,644 11,644 182c150 4/6/71 WF14-142 1
5,822 4/3/71 8
46,576 4/3/71 7
44,162 96,560 TOTAL POUNDS ---+ I 222,834 l
l DONALD E. RHODES DER /ch
A-15 IDENTIFICATION OF TEST STABS SENT TO LUKENS SLAB NO.
lA Flange Section 14W605 Beam A-572 Heat No. 123C349 1B Web Section 14W605 Beam A-572 Heat No. 123C349 2A Flange Section 14W426 Beam A-36 Heat No. 171C781 2B Web Section 14W426 Beam A-36 Heat No. 171C781 3A Flange Section 14W426 Beam A-36 Unknown Heat 3B Web Section 14W426 Beam A-36 Unknown Heat 4A Flange Section 14W605 Beam Sun Ship Weld Mock Up A-572 Unknown Heat 4B Web Section 14W605 Beam Sun Ship Weld Mock Up A-572 Unknown Heat SA Flange Section 14W426 Seam Sun Ship Weld Mock Up A-36 Unknown Heat SB Web Section 14W426 Beam Sun Ship Weld Mock Up A-36 Unknown Heat
4
(
conn wo. ao aos 570 (R 1/72)
A-16 LUKEN S STE E L C OM PANY COATE SVILLE. PA.
/
Lap 0RATORY PuntHAsE ORDER IfD.
DATE July 1. 1976 ATERI AL.aVRcHASE ORDER NO.
377971 REPORT NO.
VEN00P Sun Shipbuilding and Drydock Co.
Page:
1 of 10 LABORATORY TEST REPORT Following are chemistry, impact, and dropweight data obtained from material submitted to Lukens by Masrs. J. Durant and R. Bicicchi of Sun Shipbuilding and Drydock Co.:
1.
Slab 1A:
a.
Gauge - 4" b.
Chemical Analysis -
C Mn P
S
_C u Ni Cr Mo Si Al V
Ti
.257.
1.187.
.0077.
.0177.
.247.
.127.
.047.
.037.
.227.
.062%.07%
.0027.
Dropweight Tes t Results - NDT is + 30*F.
c.
d.
Impact Test Results -
Lat. Exp.
Test Lat. Exp.
Dir.
ft.-lbs.
(Mils)
- 7. Shear Temp.
Dir.
_Ft.-lbs.
_(Mils)
% Shear Long.
10-12-12 7-12-10 10-10-10 0
Thru-ga.
3-3-2 5-3-2 1-1-1 16-18-18 17-20-19 20-20-20
+50*F.
10-5-B 10-4-9 10-10-10 4-18-18 3-16-16 10-20-20
+80*F.
10-12-5 8-10-3 10-10-1 10-18-22 9-12-25 10-20-20
+125'F.
15-16-15 17-18-18 20-20-20 33-32-34 31-30-32 40-40-40
+212*F.
22-22-28 22-23-26 20-30-30
$UPERVISOR - LA00RATORY LUKENS STEEL CouPANY
4
(
acau no.4e ais $70 ( R 1/72 )
i A-17 LUKEN S STE E L C OM PANY COATE SVILLE, PA.
J July 1, 1976 LAgCRAf0RY PURCHASE ORDER Ng, DATE (TERIAL PURCHASE ORDER NO.
J77971 REPORT N0.
- VENDC, Sun Shipbuilding and Drydock Co.
Page:
2 of 10 LABORATORY TEST REPORT 2.
Slab IB:
a.
Gauge 1/2" b.
Chemical Analysis -
C Mn P
G Cu Ni Cr Mo Si Al V
Ti
.24% 1.20%.011%
.0197.
.24%.12%.04%.03%.22%.058%.071%.002%
c.
Dropweight Test Results - NDT is +20*F.
d.
Impact Test Results -
Lat. Exp.
Test Lat. Exp.
Dir.
Ft.-lbs.
(Mils)
% Shear Temp.
Dir.
Ft.-lbs.
(Mils)
% Shear Long.
2-2-2 3-2-2 1-1-1 0*F.
Thru-ga. 2-2-2 2-3-2 1-1-1 5-5-5 8-7-8 10-10-10
+50*F.
3-4-2 5-5-3 10-10-1 28-28-30 26-28-29 30-30-30
+80*F.
2-2-4 2-3-4 1-10-10 15-15-18 17-18-22 20-30-30
+125'F.
5-5-6 7-8-8 10-10-10 30-28-26 27-26-24 40-40-30
+125'F 70-66-70 67-58-63 70-60-70
+212*F.
12-12-10 16-14-12 20-20-20 SUPERVISOR - LABORATORY LUKENS STEEL COMPANY
(
con to. ao ais:570 (R 1/72)
A-18 LUKEN S STE E L C OM PANY COATESVILLE. PA.
J LASDR ATORY PURCH ASE ORDER KC.
DATE July 1. 1976 ATERIAL PURCHASE ORDER NO.
J77971 REPORT NO.
VENDO. Sun Shipbuilding and Drydock Co.
Page:
3 of 10 l
LABORATORY TEST REPORT 3.
Slab 2A :
a.
Gauge - 3" b.
Chemical Analysis -
C Mn P
S Cu Ni Cr Mo Si Al V
Ti
.267.
1.25%.010%.029%.02%.02%.03%.01%.07%.002%
.0317.
.003%
c.
Dropweight Test Results - NDT is + 20*F.
d.
Impact Test Results -
Lat. Exp.
Test Lat. Exp.
Dir.
Ft.-lbs.
(Mils)
% Shear Temp.
Dir.
Ft.-lbs.
(Mils)
% Shear Long.
12-10-10 13-11-10 10-10-10
+30*F.
Thru-ga.
5-5-6 3-4-4 1-1-1 12-14-15 14-16-17 20-20-20
+50'F.
12-12-10 14-12-11 10-10-10 42-43-38 37-40-33 50-50-40
+65'F.
14-16-12 18-20-13 20-20-10 72-68-66 70-68-64 70-70-70
+80*F.
18-18-17 20-19-18 20-20-20 98-94-96 86-84-86 90-90-90
+125'F.
23-22-20 24-23-20 30-30-20 100-98-102 94-93-96 99-99-99
+212*F.
28-30-30 30-33-34 30-40-40 tw SUPERytSOR - LABORATORY LUKENS STEEL COMPANY
i comm ua. ao ais 570 (R 1/72)
A-19 LUKEN S STE EL C OM PANY COATESVILLE, PA.
J LAOORATORY PURCHASE GRDER N#.
DATE July 1. 1976 LTERI AL PURCH ASE ORDER N0.
377971 I
REPORT ND.
l l
VEND 0s Sun Shipbuilding and Drydock Co.
Page: 4 of 10 LABORATORY TEST BEPORT 4.
Slab 2B:
a.
Gauge 7/8" b.
Chemical Analysis -
C Mn P
S Cu
__ Ni
_Cr
_Mo
__ S i Al V
Ti
.24% 1.25%
.0137.
.028%_
.02%.02%.03%.01%.08%.002%
.0317.
.003%
c.
Dropweight Test Results - NDT is O'F.
d.
Impact Test Results -
Lat. Exp.
Test Dir.
Ft.-lbs.
(Mils)
% Shear Temp.
Dir.
Long.
5-5-4 6-7-5 10-10-10
-20*F.
Thru-ga.: Not tested because 18-20-20 19-21-22 20-30-30
+20*F.
gauge is only 1-7/8".
28-32-36 31-34-38 30-40-40
+50*F.
50-55-45 51-52-49 50-50-50
+80*F.
68-74-70 68-75-69 70-80-80
+125'F.
74-78-78 73-74-75 80-90-90
+212'F.
m SUPERvl50R - LABORATORY LUKENS ETEEL COMPANY t_.______________..___.___.__
$ cane.soais570 (R 1/72)
A-20 LUKEN S STE E L C OM PANY COATESVILL5. PA.
LAIORATORY PURCHASE ORDER NO.
DATE July 1, 1976 ATERIAL PURCHASE ORDER NO.
377971 R(FORT NO.
yg,gg, Sun Shipbuilding and Drydock Co.
Page:
5 of 10 LABORATORY TEST REPORT 5.
Slab 3A:
a.
Gauge 7/8" b.
Chemical Analysis -
C Mn P
S Cu Ni Cr Mo Si Al V
Ti
.30% 1.14%.020%
.0277.
.06%.10%.06%
.037.
.07%
.000%.
.035%.002%
c.
Dropweight Test Results - NDT is +20*F.
d.
Impact Test Results -
Lat. Exp.
Test Lat. Exp.
Dir.
Ft.-lbs.
(Mils)
% Shear Temp.
Dir.
Ft.-lbs.
(Mils)
% Shear Long.
2-2-6 3-4-10 1-10-10 0*F.
Thru-ga.
2-2-4 4-3-5 1-1-1 18-22-20 20-24-23 20-30-30 +50*F.
8-10-5 9-12-6 10-10-10 58-55-51 57-53-47 60-60-50 +80*F.
8-9-6 10-12-8 10-10-10 46-28-35 42-26-37 40-30-40 +125*F.
18-15-12 21-16-14 20-20-20 47-52-37 41-45-35 50-50-40 +125' 78-78-77 76-79-73 80-80-80 +212*F.
28-30-28 26-31-28 30-40-40 l
n p*
6
$UPERVISOR - LABORATORY LUKENS $1 EEL COMPANY
g aou co. ao aio 170 (R 1/72)
A,
LUKEN S STE EL C OM PANY COATESVILLE, PA.
1 LABORATORY PURCHASE ORDER N,0.
DATE July 1. 1976 ATERIAL PURCHASE ORDER NO.
377971 REPORY NO.
VEN00m Sun Shipbuilding and nrydnete r'n,
Page:
6 of 10 LABORATOBY TEST HEPORT 6.
Slab 3B:
a.
Gauge 3/4"
- b.. Chemical Analysis -
C Mn P
S Cu Ni Cr Mo Si Al V
Ti
.30% 1.14%.021%.029%.06%.10%.06%
.027.
.07%.001%
.0357.
.002%
Dropweight Test Results - NDT is + 20*F.
c.
d.
Impact Test Results -
i Lat. Exp.
Test Dir.
Ft.-lbs.
(Mils)
% Shear Temp.
Dir.
s Long.
8-4-10 10-8-12 10-10-10
-20*F.
Thru ga. = Not tested 14-18-16 17-19-18 20-20-20
+20*F.
because gauge 30-25-24 26-25-22 30-30-20
+50*F.
is only 1-3/4".
30-30-32 31-32-34 30-40-40
+80*F.
50-58-50 50-60-53 50-60-60
+125*F.
i 60-58-63 64-60-65 70-60-70
+212*F.
SUPERVISOR LABORATORY LUKENS STEEL COMPANY
I roas no. ao ais 570 (R 1/72)
A-22 LUKEN S STE E L COM PANY COATESVILLE. PA.
LABOR ATORY PURCH ASE ORDER N,0.
DATE Y"Iv 1 1976 ATERIAL PURCHASE ORDER N0.
J77971 REPORT NO.
,ggg,,
Sun Shipbuilding and Drydock Co.
Page:
7 of 10 LABORATORY TEST REPORT 7,
Slab 4A:
a.
Gauge - 4" b.
Chemical Analysis C
Mn P
S Cu Ni Cr Mo Si Al V
Ti
.257.
1.18%.013%.032%.20%.02%.05%.01%
.217.
.027%.075%.003%
c.
Dropweight Test Results - NDT is O'F.
d.
Impact Test Results -
Lat. Exp.
Test Lat. Exp.
Pir.
Ft.-lbs.
(Mils)
% Shear
_ Temp.
Dir.
Ft.-lbs.
(Mils) % Shear Long.
5-6-8 7-8-10 10-10-10
-20'F.
Thru-ga.
2-2-2 3-2-4 1-1-1 32-28-30 34.27-33 40-30-40
+50*F.
10-5-7 13-7-10 10-10-10 55-51-50 49-47-46 50-50-50
+80*F.
10-10-10 14-12-14 20-20-20 48-32-70 27-66 60-30-60
+125*F.
12-14-14 13-19-18 20-30-30 38-48-62 39-66-58 50-60-60
+125'F.
78-82-84 80-83-85 80-90-90
+212*F.
20-20-18 23-23-20 40-40-30 1
O
&V
.o',
SUPERVl10R - LABORATORY LUEENS STEEL COMPANY u_,_________._____---
coeto. ao sie 570 (R 1/72)
A-23' LUKEN S STE EL C OM PANY COATESVILLE. PA.
J l
l LABOR ATORY PURCH ASE ORDER 70' DAfg July 1. 1976 dATERI AL PURCHASE ORDER NO.
377971 REPORT NO.
VENDO, Sun Shipbuilding and Drydock Co.
Page:
8 of 10 LABORATORY TEST REPORT
- 8. gg:
a.
Gauge 2-5/8" b.
Chemical Analysis -
C Mn _
P S
Cu Ni Cr Mo Si Al V
Ti
.23%
1.157.
.015%.032%.20%.02%.05%.01%.21%.081%.073%.003%
Dropweight Test Results - NDT is + 30*F.
c.
SUPERVISOR LA80RATORY LUKENS STEEL CONPANY
hia...<sio a v7n A-24 LUKENS STE EL C OM PANY COATESVILLE, PA.
s s
LABDFATORY PURCHASE ORDER NO.
DAfg July 1. 1976 IATERIAL PURCHASE ORDER N0.
377971 REPORT NO.
vtWDD*
Sun Shipbuilding and Drydock Co.
Page:
9 of 10 LABORATORY TEST REPORT 9.
Slab 5A:
a.
Gauge - 3" b.
Chemical Analysis -
C Mn
_ p S
Cu Ni Cr Mo Si Al V
Ti
.27%
1.297.
.0137.
.0377.
.027.
.027..037. M
.057.
.0007.
.032%
.0027.
Dropweight Test Results - NDT is + 20*F.
c.
d.
Impact Test Results -
Lat. Exp.
Test Lat. Exp.
Dir.
_Ft.-lbs.
(Mils)
% Shear Temp.
Dir.
Ft.-lbs.
(Mils) % Shear Long.
2-2-4 3-3-6 1-1-1
-20*F.
Thru-ga.
2-2-5 3-2-3 1-1-1 18-16-16 20-18-17 20-20-20
+20*F.
5-4-4 6-5-6 10-10-10 30-10-22 27-14-20 30-20-30
+50*F.
8-8-6 9-10-6 10-10-10 32-51-60 27 50-58 40-60-60
+80*F.
10-8-18 12-9-20 20-10-20 40-90-72 37-84-70 50-80-80
+125'F.
22-24-20 23-25-20 30-30-30 92-88-90 88-84-86 99-90-99
+212*F.
30-28-32 35-26-33 40-30-40 ruY SUPERVISOR - LABORATORY LUKENS STEEL COMPANY I
F.
ll com's wo. ao atoc570 (F 1/72)
'i
,A-25 LUKENS STE EL COMPANY COATE5VILLE. FA.
1
.h J
I l
LA00R ATORY PURCH AIE ORDER f D.
DATE July 1, 1976 NATERIAL PURCHASE ORDER NO.
J77971 REPORT NO.
Sun Shipbuilding and Drydock Co.
,g..,,,
Page:
10 of 10 l
LABORATORY TEST REPORT
'l 10.
Slab SB:
a.
Gauge b.
Chemical Analysis -
C Mn P
S Cu Ni Cr
_ Mo Si
_Al V
Ti
.307.
1.327.
.0167.
.0407.
.027.
.027.
.037.
.017.
.057.
.0007.
.0327.
.0027.
c.
Dropweight Test Results - NDT is O'F.
\\
SUPERVISOR LADDRATORY LUKENS ITEEL CCMPANY l
l
July 7, 1976 ADDENDUM TO SUN SHIP REPORT l.
Since completing the Sun Ship report, additional data has become available warranting these comments:
1)
Test data from retained samples by both Sun Ship and Vepco became available on July 1 & 2.
The Sun Ship data is in the Appendix; the Vepco data is attached to this report.
These data have been plotted (nct statistically) and are shown on the attached graph.
We estimate that approximately 35-40 wt.% of the steel in the structures is represented in the graph.
It appears that the structure is actually poorer in impact properties than Sun Ship's report estimated.
We would estimate that it is approximately equal to Steel 3 in the Summary exhibit.
l 2)
Upon learning of Dr. Tetelman's association with the l
ACRS, we reviewed his book, " Fracture of Structural Materials."
We discovered that with regard to using Charpy V appearance (% Shear) as a criteria, we were "re-inventing the wheel."
Dr. Tetelman in 1967 (p. 117 & 118) stated the following:
"..., the best correlation between Charpy impact tests and service failures seems to be that less than 70% crystalline (more than 30% shear) 14 appearance on the Charov bar, when broken at a particular temperature, indicates a high_probabili_tv that cleavace will not occur in service at or above the particular temperature, if the working stress is about 1/2 ey.
...thus an arbitrary criterion based on impact energy has no general relation to the NDT temperature and consequently has no physical meaning; criteria based on fracture appearance are more realistic and should be used whenever possible in the absence of established NDT criteria." (emphasis, Dr. Tetelman's)
"In mild steel the NDT and Charpy V notch i
appearance transition temperatures may be used for safe design criteria."
l (1) Parenthetical expression not in original text.
9
l lj ll)li s
)
1 8
C 9
4 D
3
(
C t
3 a
2 e
7 1
rH t
n a
0 Aw e
4 o
H 2
~
n
/
k 7
n 0
U F
2 2
C
+
/
'r jr 0
I 0
i i
7 s s
2
'e s '0 e7 j
p 0
.:j3 8
1 tq
/
n
~
4 W
g 1
A5#
eT s
0 6
o' pn 1
wt S 2
'o 7
a T
n e U
R 5 O-r H
0 A
pj 4
E P
1 R
PF 9
U UO 4
T S
3 2
-+
A S
C 4
t D(
0 R
1 R E 3
2 E
1 OI 2
+
1 P
E T
=
1 TR AE M
R t
U E
RP a
~
G EO e
0 T
NR H
0 I
1 F
EP G N 1
0 MW 8
/ei
+
. 8 xo*
AO C
ETN D
SK 0
y 6
g r
x n pe EE 0
y%
4 l o pr r
r a e o e e
t ah h hLCS
/
0 2
VL yw 0
0 0
0 0
0 0
0 0
o g
0 8
7 6
5 4
3 2
j 1
64 1i hg 4
l A
~
O tl\\lj l
lIl l
l1l 1,ltl lll l
e i
VEPCO DATA - REPLACEMENT STEEL ON REBUILDING
\\
TEMP.
A-572, HEAT Sl7J1106 A-572, HEAT 517J1058 SAMPLE
'F NDTT = 80*F NOTT BTW 100-120*F Ctr FT--LB MILS CV FT-LB MILS lL 80 8
6 2L 80 5
5 3L T80 5
5 4L 125 36 25
(
SL 125 32 8
6L 125 12 8
7L 140 18 19' 8L 140 13 16 9L 140 35 33 10L 165 43 38 llL 165 52 35 12L 165 50 45 1TT 80 5
3 2TT 80 5
4 l
3TT 80 8
4 4TT 125 22 14 STT 125 32 18 6TT 125 29
~
18 7TT 140 14 18 BTT 140 18 18 9TT 140 18 19 10TT 165 28 31 llTT 165 30 35 12TT 165 31 33 l
W
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _. _ _ _ _ - _ - - - _ - - -