ML20215B734
| ML20215B734 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 01/24/1986 |
| From: | Beck J TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
| To: | |
| Shared Package | |
| ML20197E251 | List:
|
| References | |
| FOIA-86-272, FOIA-86-454 ISAP-II.B, PROC-860124-02, NUDOCS 8610080459 | |
| Download: ML20215B734 (12) | |
Text
'
W v .
i COMANCHE PEAK RESPONSE TEAM ACTION PLAN ,
ISAP II.b
Title:
Concrete Compression Strength Revision No. 0 1 2 3 Revised to Reflect Incorporates Reflects Coments i Description Original Issue NRC Coments SSER on Plan __,
1 L _ _ 2=^
& (4 Prepared and '
/ ,/ /I g / ,4/f %
I
. Recommended by: t ,
Review Team Leader J4MW# /-
no j Date lo-S-84 1kiW &lv2zl8[
,1.1. I t L.
R '
\.
I l }/ ,
Ws/eh r av v Team #N -
Os c ).h> l Date (C'f( }d l ll $ $I I Y SL
/ /
B610090459 861003 FDR FOIA GARDE 86-272 PDR
V R;visicut 3 Pcan 1 of 11 ISAP II.b Concrete Compression Strength
1.0 DESCRIPTION
OF ISSUE IDENTIFIED BY NRC The TRT investigated allegations that concrete strength tests were falsified. The TRT reviewed an NRC Region IV investigation (IE Report No. 50-445/79-09; 50-446/79-09) of this matter that included interviews with fifteen individuals. Of these, only the alleger and one other individual stated they thought that falsification occurred, but they did not know when or by whos. The TRT also reviewed slump and air entrainment test results of' concrete placed during the period the allager was employed (January 1976 to February 1977) and did not find any apparent variation in the J uniformity of the paramatars for concrete pieced during this period. Although the uniformity of the concrete placed appears to minimize the likelihood that low concrete strengths were obtained other allegations were raised concerning the falsification of records associated with slump and air content tests. The Region IV staff addressed these allegations by assuming that concrete strength test results were adequate. Furthermore, a number of ,
other allegations dealing with concrete placement probless (such as l deficient aggregate grading and concrete in the mixer too long) were also resolved by assuming that concrete strength test results were adequate. The TRT agrees with Region IV that, while the proporderance of evidence suggests that falsification of results did not take place, the matter cannot be resolved completely on the basis of concrete strength test results, especially if there is any doubt about whether they may have been falsified. Due to the importance of the concrete strength test results, the TRT believes that additional action by TUEC is necessary to provide confirmatory evidence that the reported concrete strength test results are indeed representative of the strength of the concrete installed in the Category I concrete structures.
1 2.0 ACTION IDENTIFIED BY NRC Accordingly, TUEC shall determine areas where safety-related ,
concrete was placed between January 1976 and February 1977, and l provide a program.t,o assure acceptable concrete strength. The program shall include tests such as the use of Schmidt Hammer Tests on a random sample of the concrete in areas where safety is critical. The program shall include a comparison of the results with the results of tests performed on concrete of the same design strength in areas wh'ere the strength of the concrete is not questioned, to determine if any significant variance in strength occurs.. TUEC shall submit the program for performing these tests to the NRC for review and approval prior to performing the tests.
Rsvisica: 3-Peg 2 2 of 11 ISAP II.b (Cont'd)
3.0 BACKGROUND
l Ii Falsification of concrete strength tests is alleged to have l occurred between January 1976 and February 1977. Air i entrainment and slump tests have been reviewed, and no apparent variations were found in the the uniformity of the I
parameters for concrete placed during the allegation time frame. Because these parameters were in accordance with the l laboratory approved concrete mix designs this reduces the chances that low concrete strengths were obtained. Concrete compressive strength tests have been used to resolve allegations of falsifications of slump and air entrainment tests and allegations dealing with concrete placement problems 1 (such as deficient aggregate grading and concrete in the mixer 1 too long). Due to the importance of concreta compressive l strength tests, the TRT requested that additional testing be performed by TUEC to confirm that concrete strength tests performed on the concrete in question are representative of the actual concrete strength. Therefore, TUEC has decided to implement a program to test the concrete in question for verification of acceptable strength, s
1 i
l 4.0 CPRT ACTION PLAN l 4.1 Scope and Methodology The objective of this action plan is to verify the quality of the concrete in question. The CPRT tasks to be Laplemented to achieve this objective.are:
- Define the population of concrete poured during the time period in question and a second population outside of this time period for comparison.
- Perform Schmidt Hammer Tests on a random sample of concrete from each population.
i
Rsvicion: 3 Pegn 3 of 11 i 1
ISAP II.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd)
- Compare the results of the Schmidt Hammer Tests of the
- two populations and assess the significance of any differences observed relative to the design value of the concrete compressive strength.
These tasks are described in more detail in the following paragraphs.
The relative strengths of concrete poured during the period in ll '
question (concrete at issue, or CAI) and the concrete poured outside this period (control concrete, or CC) will be compared using the Schmidt Hammer Test as a measure of strength. The Schmidt (Rebound) Hammer Test, a non-destructive test, will be conducted in accordance with ASTM-C805-79 " Standard Test Method For Rebound Number of Hardened Concrete". The Schmidt Hasmar is essentially a concrete hardness tenter which measures the rebound of a spring loaded plunger after it has struck a smooth concrete surface.
Using this indirect test of strength, the two populations of concrete will be compared empirically and statistically. In addition to reporting the raw rebound number data, statistical summaries such as means and variances will be computed. Both parametric and non-persmetric distributions may be considered for the two populations. For parametric distributions, goodness-of-fit tests will be performed on the samples.
Concrete cylinder data for the two populations will also be obtained, reviewed, and used for reference.
The two populations of hansaer rebound values will be compared at the. tenth percentile level. The tenth percentile level is selected as a point of comparison based on ACI Standard 214-65, " Recommended Practice for Evaluation of Compression Test Results of Field Concrete", which gives the general guideline that no more than one in ten cylinder compression tests fall below the design strength. The population of rebound numbers from the control concrete will be used to establish a target tenth percentile rebound number. The tenth percentile rebound value of the concrete at issue will then be compared with this target value and other target values which are fractions of the tenth percentile of the control concrete.
Hypotheses that the tenth percentile rebound number for the CAI is greater than or equal to various target values will be tested at a minimum level of significance of 5 percent.
i i The reason for considering more than one target value is based on a preliminary review of data (including cylinder data) which indicates that at the population tenth percentile level l
R:vicica 3 Pega 4 of 11 ISAP II.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) the CAI may be lower than the CC. How much lower may be quantified by testing hypotheses using different target values which are fractions of the CC tenth percentile rebound value.
In addition, the significance level at which a hypothesis is just accepted will be determined. A- higher significance level passed indicates a greater confidence that the hypothesis is true.
If it is determined that the tenth percentile of the CAI rebound values is not sufficiently close to that of the CC, either further testing will be performed to verify that the tenth percentile concrete strength for the concrete at issus ,
is greater than the minimum design strength, or an evaluation will consider the significance of a lower strength concrete.
4.1.1 Test Program 4.1.1.1 Engineering shall determine the areas where concrete was placed in Category I structures between January 1976 and February 1977.
4.1.1.2 From these areas, engineering shall determine l the number of concrete truck loads for which part of the concrete of that truck load is exposed and testable.
4.1.1.3 Each truck load identified as exposed and testable will be assigned a unique numbec.
4.1.1.4 Grid areas corresponding to these truel. loads will be selected at random to be tested.
4.1.1.5 The concrete surface in a selected area shall be prepared for testing per ASTM C805-79.
4.1.1.6 The prepared areas shall be tested in accordance with' ASTM C805-79.
4.1.1.7 Engineering shall determine the areas where concrete was placed in Category I structures between March 1977 and August 1977.
l 4.1.1.8 From these areas, engineering shall determine the number of concrete truck loads for which
'part of the concrete of that truck load is exposed and testable.
4.1.1.9 Each truck load identified as exposed and testable will be assigned a unique number.
l R vicion: 3 Paga 5 of 11 ISAP II.b (Cont'd) 4.0 CPRT ACTION PI.AN (Cont'd) 4.1.1.10 Grid areas corresponding to these truck loads
- will be selected at random for testing.
4.1.1.11 The concrete surface in a selected grid area shall be prepared for testing per ASTM C805-79.
4.1.1.12 The prepared areas shall be tested in accordance with ASTM C805-79.
4.1.1.13 If necessary, concrete of known strength (determined by compression testing of core samples) will be Schmidt Hammer tested to calibrate hammer rebound numbers with concrete strength. This contingency is
~ described below.
4.1.2 Sampling Plan At Comanche Peak concrete placement quality procedures were based on required air entrainment and slump tests to be performed on each truck load. Test cylinders from the first truck load and every tenth truck load thereafter were required to verify quality. These procedures were based on ASME-ACI 359 and ACI 318 which reference appropriate ASTM standards. Since the original quality control program was based on the unit of a truck load, TUEC will use a truck load as the unit to be tested in the present quality evaluation. This is consistent with the implicit assumption that a truck load represents the smallest unit of concrete with uniform material properties. The area for testing will be limited to the exposed surface area where the Schmidt Hammer Test can be performed. The determination of the number of truck loads which were placed as exposed testable concrete will be determined as follows:
- Number of truck loads (slabs on ground) = ,
(l'0" depth X surface area) / 10 yd8 per j truck
- Number of truck loads (suspended slabs, columns, walls) = Total cubic. yards / 10 yd 8 per truck For deep slabs placed against the ground, a depth of one foot is used because during placement vibrators were used causing the concrete to flow and level out.
Thus only the truck loads placed the last foot of the
R;vicien: 3 Pega 6 of 11 1,
' ISAP II.b (Cont'd) .
4.0 CPRT ACTION PLAN (Cont'd) slab ~would be exposed. For suspended slabs between 18 and 24 inches, the total number of truck loads is distributed equally to the top and bottom halves of the sisb. For suspended slabs between 24 and 46 inches, the volume of concrete is split into three equal quantities, with one third at the top, one third on the bottom, and one third in the middle of the slab. Only i
the top and bottom layers are assumed to be testable.
Engineering has determined that there were 326 concrete placements in Category I structures between January 1976 and February 1977. Of these, 198 vere identified as being accessible for surface testing. It has been determined that the total number of truck loads corresponding to exposed surface concrete is approximately 1305. A similar number of truck loads will be used to define the population of concrete not in question. Both concrete populations will le randomly sampled to a sample size of at least 100. The number 100 was chosen as an upper bound on a practical sample size for the statistical comparison of the two populations and is_ not based on Appendix D.
4.2 - Participants Roles and Responsibilities The organizations and personnel that will participate in this effort are described below with their respective scopes of work.
4.2.1 TUGC0 Nuclear Engineering (TNE) - Civil / Structural Discipline 4.2.1.1 Scope
- Ccacrate population determination
- Sample selection
- Locate test areas and prepara operational
- Assistance in evaluation of test data and preparation of Ra'sults Report 4.2.1.2 Personnel Mr. R. Hooton THE Civil / Structural Discipline Supervisor l
R;visicus 3 Pcg3 7 cf 11 ISAP II.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd)
Mr. C. Corbin Civil Engineer 4.2.2 Brown & Root 4.2.2.1 Scope
- Prepara concrete test surfaces 4.2.2.2 Personnel Craft personnel as required 4.2.3 Third-Party Activities 4.2.3.1 Scope
- Review of sample selection
- I
- Perform hammer tests
- Document tests
- Review test data i
- Review and statistical evaluation of I test results l
- Preparation of Results Report 4.2.3.2 Personnel j Mr. H. A. Levin TERA, CPRT Civil /
Structural Review Team Leader Dr. J. R. Honekamp TERA, TRT Issues Manager Dr. F. A. Webster JBA Associate (CPRT Statistics Advisor)
Dr. D. Veneziano MIT Professor of Civil Engineering (Engineering Statistical Consultant) i Mr. G. Lagleder SWRI Manager (Testing and Inspection) l
_ _ . . _ _ . _ _ _ _ . . . _ . _ _ _ _ . . ~ . . _ _ _ _ _ _ . . _ . -
l -
Ravicient 3 8 of 11
~
l P::ga ISAP II.h 1 (Cont'd) i 4.0 CPRT ACTION PLAN (Cont'd) 4.3 Personnel Qualification Requirements Where tests or inspections require the use of certified inspectors, qualifications at the appropriate level will be to the requirements of ANSI N45.2.6, " Qualification of Inspection, Examination, and Testing Personnel at Nuclear Power Plants". CPRT third-party inspectors will be certified to the requirements of the third-party employer's Quality Assurance Program, and trained to the applicable inspection procedures.
Third-party participants in the implementation of this Action Plan will meet the personnel qualification and objectivity ,
i requirements of the CPRT Program Plan and its implementing procedures.
Other participants will be qualified to the requirements of -
i the CPSES Quality Assurance Program or to the specific requirements of the CPRT Program Plan. Activities performed .
j by other than third-party personnel will be governed by the t
- applicable principles of Section III.K. " Assurance of CPRT Program Quality", of the.CPRT Program Plan.
4.4 Acceptance Criteria i >
Three possible hypothesis tests will be considered for the evaluation of hammer rebound data, and one (or ones) selected with the most power. The three test methods include:
4.4.1 Method A tests whether the tenth percentile rebound value of the CAI is greater or equal to the target value of the CC, where both populations are assumed to be normally distributed.
NOTE: The target value is defined as the tenth percentile or a fraction of the tenth percentile value.
4.4.2 Method B tests whether the percentage of rebound values in the CAI population above the target value of the CC is greater or equal to 90 percent. In this test the CC :'
population is assumed to be normally distributed for purposes of establishing'the target value (which may be ;
i defined as the tenth percentile or a fraction thereof), '
l but the distribution of CAI rebound values is l
unspecified (non-parsnetric).
i 4.4.3 Method C costs whether individual CAI rebound data
- values belong to the same distribution as the control 1 l
l 6
. . - . , . . _ - _ _ _ . . - , , . . , , . - . - - .._,_r.,_,,._....._.y-- .,,._,_,n,.,_..% - - .. ,,,...,..,w y-, ..,w,._e.,. --,...,.,,,..,m .-_,.,y, _ - , - , _
. R;visign: 3 Pega . 9 of 11 i .
ISAP II.b (Cont'd) f 4.0. CPRT ACTION PLAN (Cont'd) concrete rebound values. No assumptions are made regarding either. population distribution (non-parametric).
Although the power functions for these three methods are not directly comparable, preliminary indications are that both Methods A and B should be utilized, since they are of similar power when the CAI rebound data is assumed normally distributed in Method B. Method C appears to have very little power, and will therefore not be used to evaluate the data.
Based on the sample outcomes for the two concrete populations, a test statistic will be computed and the hypotheses regarding the concrete at issue will either be accepted or rejected at the 5 percent level of significance. In addition, the level of significance at which the hypotheses would just be accepted will be determined.
If after reviewing the results of these hypothesis tests and the cylinder data, the CAI population is found to be significantly lower than the CC population at the tenth percentile level, then a decision will be made to either 1 calibrate the Schmidt Hammer Test to concrete of known strength, or to test cores from the CAI.
If it is decided to calibrate the Schmid; Hammer Test, concrete samples of known strengths and of similar age as the
- concrete in question will be used to calibrate the hammer
' rebound values.
The rebound values for the concrete in question, will then be r converted to cylinder strengths taking into account the uncertainty in the conversion. If.the tenth percentile ;
strength is greater than the minimum specified the design ~
strength, f', it will be concluded that the strength of the concrete in* question is acceptable. If not, additional
' alternatives (e.g. analysis or further testing) will be '
considered.
Any such decisions will be documented as a revision to this action plan.
l i
1 l
..------,---,m,-~.we,,ww-,c-wy- ----,-,,.r, - , , ----e,, r+ y- ww,,-e- -
Revision: 3 Page 10 of 11 ISAP II.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) 4.5 Decision Criteria The action identified by the NRC (Section 2.0) will be considered complete after all Schmidt Hammer Tests have been completed, the results statistically analyzed, and the two concrete populations compared.
If the comparison indicates a significant difference in the two populations, then a decision will be made to further evaluate the CAI in accordance with Section 4.4.
e j
\ . ,
kvicient 3 Prga 11 cf 11 ISAP II.b (Cont'd)
ATTACHMENT FLOW CHART 11
=:1 3
w= .
9r -
4 ;
. . p.
g 3 .nw;l E8I c
g g
- E 'ill slot I l'7e 7
i g
!sIL., 6 -
m ._
! "3
'W l- !! ,
s -
i= Ffl
- e
- lji _-
a u ' '
it
- . .= !
jjl(a~ll
~
U i h i!!!! h!bl ij l-
.9 L If 3'
3 l L E 2*JNl ,,!$i3 c:n wa
- conni- m o iG !
-1g el! 3=
E en s = ;.
5 i h l
- . . - . . , . . . .--- - - - - . - - - - , - . - - - . _ , - , , , - - . , , , . , . - . _ _ - . - - - .