ML19344D632

Final Deficiency Rept Re Improper Consolidation of Concrete in Lower Portion of Wall in Reactor Containment Bldg Dome. Caused by Improper,Random Vibration During Concrete Placement

ML19344D632
Person / Time
Site:	Waterford
Issue date:	04/08/1980
From:	Conich J, Hartnett R, Milhiser R ARKANSAS POWER & LIGHT CO.
To:
Shared Package
ML19344D630	List: ML19344D629 ML19344D632
References
NUDOCS 8004250352
Download: ML19344D632 (6)

Text

i O  ?

LOUISIANA POWER & LIGHT COMPANY WATERFORD SES UNIT No 3 t

FINAL REPORT OF SIGNIFICANT CONSTRUCTION DEFICIENCY NUMBER 13 I

REACTOR CONTAINMENT BUILDING DOME CONCRETE I

I ~

s, J ich, site Manager .

A V/ ff a L a R J Miihiset, Project Superintendent

/

R A Hartnett, QA Site Supervisor i l J Hart, Project Licensing-Engineer-l April 8, 1980 8 0 0 4 25 035A

l l SECTION I -

SUMMARY

i During the placement of concrete for Reactor Containment Building dome parapet wcll placement 521-1B, improper consolidation of concrete in the t lower portion of the wall occurred. Proper consolidation of the concrete l as required per Fegles Power Services' Procedure CP-303-3 and Ebasco Con-crete Masonry Specification LOU 1564.472 was not achieved. Upon removal of the exterior wall forms, numerous honeycomb and rock' pockets randomly i distributed in the lower portion.of the wall were evident. Fegles Power l

Services initiated Nonconformance Report W3-1576 describing these defects. l l The concrete defects were primarily due to improper, random vibration .

i l during the concrete placement. ,

SECTION II - DESCRIPTION OF DEFICIENCY l  !

The Reactor Containment Building dome placement 521-1B is a curved re-inforced concrete structure 118'-7" long, 3'-0" thick, and 11'-6" in height. The placement ir between elevation +179.00' to +190.50' with a  !

haunch for the isAtial section of the dome concrete. The concrete place-  !

ment commencer' at 6:15 a.m. on August 8,1979. At approximately 10:00 a.m., the ccacrete pump line plugged and the placement continued by plac-ing the concrete with the use of crane and buckets. The concrete vibra-tors in use were inadequate in both size and number to provide proper vib- .

ration of the concrete. In some situations, concrete was allowed to fall  !

without the use of the tremies in excess of the specified limit of five (5) feet. Concrete was also allowed to dry on the reinforcing steel and j

~

measures taken to remove it were inadequate.

At approximately 12:00 (noon), additional vibrators, men, and wire brushes l were added to the existing placement crews to insure proper vibration of the concrete and cleaning of the reinforcing steel of dry concrete. The l placement proceeded without further incident and was completed by 6:00 p.m.  !

On August 13, 1979, the outside wall forms were removed. Numerous surface honeycomb and voids were encountered in the lower portion of the placement  !

with some of these areas extending past the embedded reinforcing steel. L This indicated that consolidation of the concrete as required by Fegles Procedure CP-303-3 and Ebasco Specification 10U 1564.472 was not achieved.

SECTION III - ANALYSIS OF THE SAFETY IMPLICATIONS Had this deficiency been left uncorrected, this noncompliance with concrete i placement practices would compromise the ability of the Reactor Containment Building to satisfy some of the design bases loads and loading combinations as set forth in Sub-section 3.8.4 of the Final Safety Analysis Report.

)

1

, Pega 2 l

l SECTION IV - CORRECTIVE ACTION AND INVESTIGATION / EVALUATION / REPAIR  ;

l A) Corrective Action:

The immediate corrective action taken during the placement of the dome i i south parapet wall after the discovery of the inadequate concrete con- i solidation was to add an adequate number of vibrators, laborers, and inspectors to ensure proper placement of the concrete. Tremies were replaced back onto the placement to ensure that the concrete drop was within the specified limit of five (5) feet. Additional laborers with l wire brushes were assigned to remove dried concrete from the resteel. ,,

Af ter the removal of the wall forms, Fegles initiated Nonconformance Report W3-1576 on August 13, 1979.

Corrective action implemented to prevent the recurrence of the pro-blems encountered was the issuance of a Stop Work Order on any addi- l tionn1 dome concrete placement by Fegles Power Services until the fol-lowing corrective actions were completed:

l. Feglee shall provide a qualified job Superintendent. i

2. Fegles shall assign an experienced placement supervisor who has ultimate charge at the placement to direct the overall placement.  !

Additionally, one general foreman shall be assigned for each separate placement (each side).

3. Fegles personnel shall be indoctrinated and trained to the require- t ments of Ebasco Specification LOU 1564.472 for concrete placement ,

with particular attention to lif t height restrictions, vibration requirements, cold joint determination and subsequent action re-  ;

quirements. This training shall take place at the Waterford Site.

LP&L and Ebasco shall review the plans for this training and shall be notified when the training classes will take place so that the  ;

training may be witnessed by LP&L and Ebasce. Fegles personnel  !

who shall he trained shall include the following:

a. Project Superintendent l

b. Concrete Supervisors j

c. Foreman

d. QC Personnel Vibrator operators shall be trained in the use af vibrators and the proper method of vibration.

4. A third concrete pump shall be supplied as backup.

l S. Gas driven generators shall be provided as backup for vibrators.

l 6. Larger vibrators shall be provided, i

'Pega 3

7. A " bull" hose and air manifold shall be set up to supply the dome area and the pneumatic vibrators which shall be available.

8. More lights shall be used to light the dome.

9. A sufficient number of trained craft personnel shall be available -

to adequately place concrete, vibrate concrete, and to clean ex-posed resteel and forms of concrete spillage.  :

10. The concrete lines shall be modified to eliminate severe bends.

11. Adequate concrete buckets shall be supplied for backup.

12. Four concrete inspectors shall be available. At least two of the inspectors shall be certified Level II concrete inspectors. The ,

remaining inspectors shall at least be certified as Level I con-crete inspectors.

13. A pre-placement meeting shall be conducted at least 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> prior to placement with Fegles , Ebasco, and LP&L in attendance, to dis '~  ;

cuss in detail the plans for concrete placement.

r

14. LP&L QA will be notified at least 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> prior to scheduled place-ment commencement so that LP&L can conduct a pre-placement inspec - >

tion and to verify that all of the above (Items 1 through 13) have  ;

been accomplished. i All corrective action items listed above were accomplished, the Stop Work .

Order was rescinded and concrete placement for the Reactor Building Dome [

resumed.

B) Investigation / Evaluation / Repair: i Investigations / Evaluations ,

Investigation of the defects in dome concrete placement 521-1B was in I a four phase prog'am r of: (1) surface chipping to sound concrete, (2) i microseismic testing (Pulse Echo Method), (3) core drilling and (4) I chipping of exploratory openings.

Findings from the investigation are:

a) All surface defects were located in the lower half of the wall l and mostly in the eastern section. The overall defective area  !

was estimated to be less than 10% of the wall surface.  ;

The inside surface defects were in general superficial, i.e. not deeper than two (2) inches. In small local areas approximately two foot square the defect went beyond the surface reinforcement.

l I

-N

• Pcg2 4  ;

The outside surface defects were relatively deeper as compared with those of inside surface. Nearly half of the defects went beyond i the surface reinforcement. The depth of defective concrete was ap-proximately nine (9) inches at the maximum, and less than three (3) .

inches in average. The difference in the depth of defective con-  !

crete is possibly due to the facts that there are three layers of  ;

rebars near the outside surface and only one layer of rebars near i the inside surface, and this has made it more difficult for the con-  !

crate to properly reach to the outside surface. 1 i

It is determined according to the above that the surface defects are f superficial in their penetration, and the defects from both inside and outside surfaces are not interconnected through the three (3)  :

foot wall.  !

e i'

b) A total of approximately 600 pulse echo tests had been performed according to a preplanned grid system laid on 2 foot centers.  !

Where tests could not be performed due to surface roughness, the l grid system was locally modified for proper coverage. Where clear  !

internal reflects were found, additional test points were selected. l l

A majority of internal reflectors which could be considered as indi-  ;

cations of minor entrapped air voids, cold joints and cracking were j obtained from tests on the outside surface. Less than 12% of the '

total echo teets, 71 out of 600 found internal reflectors. Only  !

four (4) (less than 2%) of those with internal reflectors were found  !

from tests on the inside surface, and only four (4) from tests on  !

the top surface. These results indicated that the internal reflee-tors found in the horizontal tests were practically non-existent in j the vertical test orientation.

i According to the results of pulse echo tests, the possibility of  !

cold joints,' which was the original major concern, exists only in l a few small (lateral extent less than 3 to 4 inches) areas. This  ;

finding was confirmed by the results obtained from vertical and  ;

horizontal corings and exploratory chippings. The minimum insitu l concrete compressive strength is 4900 psi and well above the de- l sign strength 4000 psi. Therefore, it is concluded that the in- i ternal defects are structurally insignificant, and are only re-  !

quired to be fixed in the areas where concrete had been removed [

for the investigation. i i

c) Debris was found on the exterior faces near the construction joint  !

at EL 179.0. The debris did act extend to the interior of the  ;

joint and was traced from the outside. The condition was caused  !

by the surface contour of the joint, which formed a pocket between {

the concrete and the form. Cores showed the interior of the joint  :

to be roughened and well consolidated.

l Repairs Af ter completion of the above investigations / evaluations the following corrective action repairs to concrete placement 521-13 wete effected.  :

i k

. P, cgs. 5 i

i

1) All surface defects were removed by chipping to sound concrete. l The surface areas of the wall were then repaired using a shot-crete process in accordance with Fegles Procedure CP-303-11,

" Concrete Repair by Shoterete Process".  :

2) The regions of unconsolidated concrete found in the exploratory chippings were repaired with dry pack bonded with epoxy in accord-ance with Fegles Procedure CP-303-05, " Concrete Repair and Inspec-e tion".

3) As further outlined in Procedure CP-303-05, all debris as well as ,

l any i= properly consolidated concrete was removed from the con-struction joint by chipping and repaired with dry pack grout se-cured with epoxy.

4) All core holes and exploratory windows have been repaired by coating the surfaces with epoxy end replacing the concrete in these areas.

Reoair Verification ,

In accordance with Fegles shotcrete procedure, cores were taken of the  ;

shotcrete repair for insitu strength and bond verification.

The shoterete exhibited good bond as evidenced by cracking through the aggregate at the joint in one core and a nearly invisible seam in the  !

cther. Also, in general, the shoterete appeared sound with a uniform distributf on of air and fine aggregate. The compressive strength test re sults indicate values well over the design requirement.

To develop a higher level of confidence in the bond or repair addi-tional cores were taken in the shocrete repaired area. The bonding exhibited in these cores was also found to be acceptable.

I a

i i

L