ML20087B058

VEPCO Investigative Board Preliminary Rept. W/O Cover Sheet

ML20087B058
Person / Time
Site:	Surry
Issue date:	01/20/1984
From:	VIRGINIA POWER (VIRGINIA ELECTRIC & POWER CO.)
To:
Shared Package
ML20087B057	List: ML20087B058
References
FOIA-83-649 NUDOCS 8403080341
Download: ML20087B058 (29)

Text

r ,_ _ - - - - - - - - - - - - - - - - - - _

9 TABLE OF CONTENTS

1. Preface.............................................................. 1

- 2. Establishment of Investigation Committee ............. 2

3. Synopsis o f A e ci dent ............................................ 3 i

4. Baekgroun d In forma tion ....................................... 4

5. Su m m a ry of Events .. ............. ........ ....................... 6

6. St a t e m ent o f Fa eis ..... ..... ........... .......... ............... 9

7. Mait.tenance/Fa11ure History ............................... 15

8. C o n cl us i o ns . .. . . . . .. . .. ... . ... .. . ... . .. .. .. .. . . .. . .. . .'... . .. . . . . . . 17

?."  ;

9. R e co m m enda tions .............................................. 19

10. Summary of Investigative Aetions ........................ 21

11. E xh i b i t s . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 4

12. Co m mittee Approvals ......................................... 2 5 5

• l k

t 8403080341 840120 PDR FOIA GARDE 83-649 PDR 4

I- -

i - i O

. ESTABLISHMENT OF INVESTIGATION COMMTITEE o

On October 16,1983 at approximately 10:00 a.m. a formalinvestigation was initiated into the accidental death of Wayne C. Jones, a Control Room Operator at Surry Power Station. The f,ollowing individuals served as members of the investigation committee *

1) W.R. Cartwright, Manager Nuclear Operations Support (Chairman)

2) R.M.H. Terrier, Director - Power Operations Safety

3) H.W. Kibler, Euperintendent Maintenance

4) W.R. Rmer, Jr., Supervisor Administrative Services M.A. Reid, Director - Claims Investigations and W.E. Patterson, Shop Steward -IBEW Local 1064, were present during the investigation and participated ,

in interviewing witnesses.  ;. ',

d h

i e

2 L .

SYNOPSIS On October 15,1983 at approximately 9:00 p.m., Wayne Jones and Robert Matz were attempting to isolate Unit 1"A" High Pressure heater drain pump -

in preparation for repairs'to a leaking discharge flange. The procedure '

involved closure of the discharge, recirculation, and suction valves, and '

establishing a vaccum drag froin the pump discharge to the main steam condenser.

While completing this procedure a steam leak was detected. Mr. Jones entered the pit to open the suction valve. While doing so a section of the expansion joint ruptured resulting in the area being immediately engulfe .

steam.

When Mr. Jones was subsequently removed from the pit, emergency '

first afd was administered, but efforts to revive him were unsuccessful. He was pronounced dead at the scene by the local Medical Examiner. Mr. J was transported to the Office of the Chief Medica 1 Examiner in R[chmond. '

An autopsy was performed on October 17,1983 and the cause of death was  ;

determined to be burn injury to airways. .

The investigation comittee concluded that there were no violations of safety rules, negligence or operating errors. The expansion joint failed d overpressure caused by leakage back through the pump discharge and lation paths.

Recommendations have been made to enhance operations practices, personnel safety and to prevent a recurrence.

I 3

i o

" -- * = -- ,,.._ - _ ,_ __ _

1 i BACKGROUND INFORMATION E'quipment: Unit 1 'A' High Pressure Heater Drain Pump ,

(1-SD-P-1 A) (

9 The ,High Pressure Heater Drain Tank supplies condensate water thru a 24" pipe to the suction isolation valve, thru the suction expansion joint, to the inlet side of the High Pressure Heater Drain Pump at approximately 175 PSIG at 365*.

The condensate water leaves the pump at approximately 685 PSIG ct 365'F thru a 16" pipe to the discharge check valve thru the discharge isolation valve, thru the level control valve, to the inlet 1;ide of the feed water pumps.

4

' The recirculation line is located on the discharge line from the High Pressure Heater Drain Pump, the discharge line check valve. The before '

recirculation line provides a flow path for the H.P.

Heater Drain Pump, should the normal discharge line of the pump be restricted or closed when the 3

pump is running.

The recirculation line returns the condensate water to the High Pressure Heater Drain Tank .thru a 6" pipe, thru the check valve, thru the isolation valve, thru the recirculation control valve, on to the High j, Pressure Heater Drain Tank.

• A t

t 6

t 5

1

SUMMARY

OF EVENTS 10-14-83

- Shift order written to transfer from 1-SD-P-1B to 1-SD-P-1 A.

~

10-15-83 ,

t 1725

- Started 1-SD-P-1 A (5:25 PM) 1739

- Stoppehl-SD-P-1 A - flange leak on discharge.

(5:39 PM) 1938

- Started 1-SD-P-1 A - mechanics to check leak - identified leaks on pump discharge flange, inlet flange on discharge (7:38 PM) check valve and the pump suction flange.

1939

- Stopped 1-SD-P-1 A ,

(7;39 PM)

- Jim Simpson, Shift Supervisor, instructed Wayne Jones on the sequence of isolation for 1-SD-P-1A.

4 2030

- Jones began to isolate 1-SD-P-1 A (8:30 PM)

- Jones started closing suction valve.

- Matz arrived and assisted Jones in closing 9 ction valve 3/4 closed.

- Jones & Matz connected jumper hose from the discharge of the pump to the. main steam condenser to " vacuum drag" the pump.

- Closed the recirculation valve. l t

2100

- Jim Simpson, Shift Supervisor, arrived at work location

. and observed Jones & Matz moving a ladder to get (9:00 PM) to discharge valve.

Jones & Matz checked discharge valve isolation-took 2 turns and torqued isolation valve closed.

- Jones entered pit and closed suction valve.

- Simpson noticed a steam hiss by the suction of 'the pump that he hadn't heard prior to closing the suction valve.

- Simpson told Jones to quickly open the suction valve.

~ 6 L .

i

SUMMARY

OF EVENTS (CONT'D.)

10-15-83

- Jones attempted to open the suction valve.

Expansion joint ruptured discharging hot water and

- steam to pit. Water flashed to steam reducing visibility of pit to zero. .

- Simpson called Pete DeWitt, Assistant Shift Supervisor, on gaitronics and told him there's been an accident at the HP heater drain pumps.

I Simpson attempted to located Jones but couldn't '

because of the density of steam.

- Simpson instructed bystanders to get fire hoses and spray water on the steam. .:

- Simpson jumped into pit to rescue Jones.

- Extraordinary efforts by Simpson, Matz, Timothy Kunkle and others recovered Jones from the pit.

2111 - First Aid Alarm *

(9:11 PM)

Barry Burns (EMT) administered CPR and '

mouth-to-mouth resuscitation to Jones.

, 2115 - Simpson enlled Dave Christian (Supt. Operations)

(9:15 PM) at hame and reported the accident. .

-- Simpson reviewed Emergency Plan to determine appropriate u l '

' actions.

2130 - Simpson notified NRC of fatal ac,cident.

~

(9:30 PM) 2140 - Surry Rescue Squad arrived on site.

(9:40 PM) 2150 - Nightingale air ambulance arrived on site.

(9

50 PM) 2150 .

- Security requested Sheriff's Office contact local (9:50 PM) Medical Examiner.

2200 - W. L. Stewart, V.P. Nuclear Operations, notified (10:00 PM) of fatal accident.

2200 - Jim Simpson transported to Riverside Hospital by (10:00 PM) Surry Rescue Squad.

. 2210 - Isolated Turbine Building system 1 & 2 to stop fire (10:10 PM) water spray over HP drain pit.

7

-~- , ,,,,.,.

,,O -

r-

\

l t

l l

i STATEMENT OF PACTS -

l l

,'. Da Octohar 14, 1933 a shift order was written to transfer from  :

1-SD-P-1B (13 High Pressure Heater Drain Pump) to 1-SD-P-1A (1A High

. Pressure Heater Drain Pump).

Around 5:25 p.m. on October 15,1A HP Heater Drain Pump was started. Leakage was observed on the discharge flange on the pump and from under the insulatic,n en the check valve on th,e discharge side of the pump. The 1A pump was stopped at 5:39 p.m.

Maintenance personnel reported to work at 7:00 p.n. Lagging was removed from the check valve to aid in identifying the location of any

.1l i

leaks. Peter Grigonis, Mechanical Foreman; Carol Roberts, Mechanic; (.6 i

Jim Simpson, Shift Supervisor; and a Daniels' Mechanic were present to I

. observe at 7:38 p.m. when the 1A pump was started. Leaks were observed

• at the pump discharge flange, inlet flange on discharge check valve, and _'

..l l the pump suction flange. The 1A pump was stopped at 7:39 p.m. -

i Subsequently, Pete Grigonis contacted John Patrick, Supervisor ,

Mechanical Maintenance, discussed the situation an8 the decision was s

made to remove tha IA pump from service for maintenance.

Wayne Jones, CR0 (Unlicensed), and Robert Matz, CR0 (Unlicensed),

i were.not assigned specific watch stations for the shif t. Ji:n Simpson contacted Wayne Jones and told him to get with Robert Matz and start isolating the 1A HP Heater Drain Pump. Simpson explained to Jones the sequence of isolation. " Ensure that the discharge valve and that the i

recirculation valve were closed before the suction valve was ever closed and to have a vent path."

9

-,_,,_-.~,,,m.__ _ . _ , . _ . - , , _ , . - . . , _ . . -

4 l

Around 8:30 p.m. Wayne Jones contacted Robert Matz and told him about their assignment. Matz was working an assignment in the

• auxiliary building and he told Jones he would meet him at the pump as isoon as he finished what he was doing.

When Matz arrived ,at the 1A HP Heater Drain Pump Jones was in the pit closing the suction valve. Jones and Matz took turns closing the suction valve until it was approximately three-quarters closed.

They then connected a jumper hose (3/4" black rubber) from the discharge pressure indicator valve (3/4") to the main steam condenser to " vacuum drag" the pump. Jones pressurized the line. He then told Matz to contact No.1 Control Room Operator and advise him that they did have a vacuum drag on the pump so he could keep an eye on his main

! condenser vacuum. Matz did so using the gaitronics. The Control Room Operator acknowledged the message.

When Matz returned to the 1A pump, Jones was on top of the HP

. Heater Drain Tank closing the recirculation valve. Matz assisted f Jones and they completed closing the valve.

i Jim Simpson arrived at the work location and observed Jones and Matz trying to tighten down on the discharge valve. It appeared they couldn't close it any further. He observed they had already connected .

a hose to the discharge pump inside the isolation boundary. The vaive was open and the hose was hot and jumping. The other end of the hose 1

was connected to the main steam condenser.

When Simpson p.rrived at the 1A pump, besides Jones and Matz, two sechanics were standing-by af ter having removed the insulation from the check valve and were making initial preparations to remove the flange nuts.

10

--___. __ 9 . _ _ . , _ _ _ , . _ . - . _ . , , - . _ - . _ _ _ _ ~ ~ , . . - _ _ _ , _ _ . , - - -

j y

i t- ,

The final step in isolating the 1A pump was to fully close the y suction valve. Simpson told Jones "OK, I guess rhat's it'. Let's go ,

ahead and finish closing the suction valve." q

,: Around 9:00 p.m. , Jones entered the pit and began closing the suction valve. Simpson observed the valve stem closing (closed easily) until fully closed. Simpson motioned to Jones to leave the pit, which he did. Within seconds after Jones exited the pit Simpson

~

detected a steam hiss by the suction of the pump that he hadn't hoard i

prior to closing the suction valve. (This hissing noise meant to .

i Simpson that the pump casing or the pump was pressurized; it was g, pressurizing from a high pressure source; and that something was beginning to leak that hadn't been leaking before.) Simpson told . ,

l

' Jones to " hurry up, get back down there and open the valve back up." .}

-Jones jumped into the pit and made approximately one turn on the wheel f

- when there was an explosion. The pit was filled with steam and hot l- ..

water. , ,

1 (The sequence of events enumerated in the preceding paragraph is [,.

(

l as recounted by Robert P. Matz. In the judgement of the Investigation .

Committee this is the probable sequence of events.I Jim Simpson's . s.

explanation differed slightly and is provided for the record. ,

'Jcnes got down in the pit and started to close the i

valve. It was fairly easy to close, didn't look like it was anything he was fighting at the tima. It went all the way e

closed. Simpson noticed a hiss by the suction of the pump and realized something was leaking. He told Jones to quickly open the valve, get the valve back open. Jones made -

about one turn, two secone , and there was a tremendous '

i explosion. The pit was filled with steam and hot water. ,

The only discrepancy in testimony provided by Simpson and Matz is whether or not Jones exited the pit after closing the suction valve ,

and before being instructed by Simpson to re-open the valve.)

11 I

1 Simpson went to the gaitronics and notified the Control Ro?m that there had been an accident at the HP Heater Drain Pumps. He returned j

, o to the 1A pump and attempted to enter the pit from the South side, but ,

'l

,va6 unable to do so because of the dense steam and hot water. He

~

returned to the North side of the pit and instructed the by-standers ,

. (two mechanics) and Matz to get fire hoses and spray the steam to ,

quench it so he could get into the pit. While this was being done,

, Simpson tried several more times to enter the pit, but couldn't do so .

because of the dense steam and heat. ,

i ,.

Other personnel arrived at the scene and an atsted in manning the 4

fire hoses. Their efft,rts and actuation of the turbine building fire protection sprinkler system began quenching the steam.

As the steam began to dissipate Simpson located Jones at the 13 l

' side (Southside) of the pit; he jumped into the pit and grabbed Jones '

and tried to pull him up. Jones legs were wedged under the suction

.f piping to the IB pump. Simpson had to leave the pit because his feet h

~

I

had begun to burn. When he exited the pit he saw Pete DeWitt 2 (Assistant Shift Supervisor) and told him that Wayne Jones was in the 'i.

. '0 '

l

pit and that he thought Wayne was dead. He then dent back into the [.

f I

, pit. Steam was still pouring out of the pit and hot water pouring in. -

The pit was beginning to fill with water.

j ..

Through the efforts of several personnel (approximately 9:10 p.m)

Jones was removed from the pit and laid on the floor between the feed pumps and the bearing cooling pumps.

Barry Burns, an Emergency Medical Technician, was on the scene .

and he began CPR and mouth-to-mouth resuscitation in an effort to revive Jones. Shortly thereafter, Kenny Pier, also an Emergency ,

Medical Technician, joined in and assisted Burns in attempting to 12 I

revive Jones. He did not respond and rescue efforts to revive him were terminated.

Af ter Jones was removed from the pit and efforts 'to revive him were initiated, Simpson returned to the Control Room. At approximately 9:15 p.m. ha contacted Dave Christ 1an, Superintendent Operations, at home and' advised him that there had been an accident and he thought Wayne Jones was dead. Christian told Simpson to call Nightingale Air Ambulance and that he would be at the plant shortlye l

After Nightingale was called, Simpson checked the Emergency Plan .c to determine if activation was necessary. He determined it was t

l not, but that NRC notification was necessary. Around 9:30 p.m. he contacted the NRC Operations Center and informed them of the fatal  ;

i. accident. ,

( .

7;

! Shortly thereafter (approximately 10:00 p.m.), Simpson was I j -administered first aid by members of the Surry Rescue Squad for burns

+.

to both feet and then transported to Riverside Hospital in Newpo' t j News for examination. At the hospital, Salvideen creme was applied to j ;p

' both his feet and they were wrapped; he Fas given a mild sedative and .)

P-t . .

l released to return home.  :

, . q .,

At approximately 9:50 p.m. the Surry Sheriff's Office was

. requested to notify the local Medical Examiner of the fatal accident. . 4 Dr. June Tunstall, Medical Examiner for Surry County, arrived on site

-l around 10:25 p.m. She examined Wayne Jones and pronounced him dead.

She directed that he be transported to the Chief Medical Examiners l t

( ,' of fice at 9 North 14th Street, Richmond, Virginia. This was-done ,

using the station ambulance.

13 i

L ,

l.

MAINTENANCE - FAILURE HISTORY Operations and maintenance records were reviewed in an attempt to determine the maintenance history of the expansion joints or other related components. Flange leaks on the suction and discharge of the pumps have been common place so no effort was made to review those occurrences. Pump seal problems have also occurred rather frequently. These types of problems have required numerous tagouts of the drain pumps at both units over the operating life of the station.

The records search revealed that three expansion joint replacements j have taken place, all on Unit No. 2. No records have been uncovered that document any Unit No. I replacements, nor does anyone recall such. The

] circumstancac surrounding the past replacements are as follows:

l 2/2/76 Unit No. 2 came off at 1305 hours0.0151 days <br />0.363 hours <br />0.00216 weeks <br />4.965525e-4 months <br />. Found pin holes and I

cracks in 2-SD-P-1A expansion joint. The expansion joint

,Jh/Pl ,

was removed and the pump suction flange was lanked until an l'i'f expansion joint was available. The joint was replaced ,

during an outage of April 22,-1976'to June 2, 1976. '

I 12/29/76 At 0754 hours0.00873 days <br />0.209 hours <br />0.00125 weeks <br />2.86897e-4 months <br /> 2-SD-E-1B expansion joint blew three minutes j after starting 2-SD-P-1A (Unit No. 2) No record of problems j found with joint. The joint was replaced during the February 10, 1977 - April 10, 1977 outage.

l 5/14/78 At 0010 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> 2-SD-P-1B was taken out of service on Unit No.

i

2. On 5/29/78 the blown expansion joint was removed and blind flanged. A new expansion joint wss installed during the July 7, 1978 - August 1, 1978 outage.

15

'l ,

i

,. Il e .

r i

On October 7,1983 the expansion joint ongSD-P _13, was leaking t

slightly at the trunion to pipe veld. The leaks were repaired by welding.

Leaks were again discovered on October 20, 1983 on the same arpansion

. joint. The IB pump was not in service at the time. It was decided to t

barricade the area and not attempt isolation and maintenance until a unit 'l i

shutdown at which- time maintenance would be completed, and the suction valve opened before the unit;would be returned to power.

An interview with R. L. Johnson, Operations Coordinator (NUC),

I revealed that a similar failure to the Octobar 15, 1983 incident occurred on the 2-SD-P-1A expansion joint. Operations log entries on January 26, i'

1974 reference that the suction flange blev but no maintenance records could be retrieved. The interview revealed that the trunions on the joint I did pull away from the pipe just seconds after isolating the suction valve .

i

~! resulting in the explosive release of steam and hot water. No injuries resulted from this incident since Mr. Johnson, who had closed the suction <

valve had just exited the pit prior to the f ailure. /

i ,

i Ntuuse 7 i w

, a 4

o R

t 4

e D

16 t

-CONCLUSIONS 1

i Based on testimony of witnesses, review of records, examination t of accident scene and findings of Power Station Engineering, the '

o following conclusions h' ave been made:

!) Failure of the expansion joint was caused by overpressure

., conditions. The overpressure was created by leakage back p .

/;

i through the pump discharge and recirculation paths which ' *

'were exposed to 'the dischcrge pressure (approximately 600-675 i

j psig) of the operating pump (1-SD-P-1B).

,:J l

s ,c I

2) Safety rules and practices were not violated. All personnel ,

l Involved in isolating the pump were utilizing required safety t

i equipment.

5 t , .W

.l 3) There is no evidence of negligence on the part of employees @

or supervision.

,J:,

! ' F.

w':

l l

4) The work assignment was performed in accordance with standard (

j operations practices.

i ,

t

5) The tagout procedure was being performed in accordance with

, requirements of Section III of the Company Accident Prevention

. Manual.

• 3 9

17

RECOMMENDATIONS

.{ ,

i Although no operations errors have been identified curing i

t.his investigailon, the following recommendations should be implemented to enhance operations practices and personnel safety:

a

1) Continue the present requirement, estabUshed following the ,

accident, that no work is to be performed in H.P. I!s&ter Drain

, pits without station management evaluation and approval.

This requirement should remain in effect until system

.llj j mod'fications are completed.

i '

! 2) A formal approved procedure for removal and returning to service these pumps, and other similar systems should be p'. ,

l Instituted. (Management had already started on this program 'T I

t before the accident.) h.

Y

- 3)' The " vacuum drag" system should not be utilized to depres5urize y:.

-7 this system or other similar ones after isolation. Water or ;m l ,

f,N steain pressure should be vented or drained through properly '

s J:

_l piped drain systems. .The vacuum drag should only be utilized L' 4

after depressurization to carry away " nuisance leakage" to

~

facilitate maintenance.

i 1 4) ' Use of rubber hose on high temperature or high pressure water

-i -

or steam systems for " vacuum drag" or similar uses should

' -t-be discontinued. When temporary piping or hoses are required, .

- properly designed hoses must be used. Hoses of different l; 1

temperature and pressure ratings should be procured, properly ,

'-l

> identified as to its use, and administratively controlled to .

imure cerrect. usage. . ,n_

l 1) Isolation of this system or other similiar systems where there i

is a potential for overpressure should only be performed with .

constant pressure indication available to personnel performing 3

, the task. ,

4) System vent and drain valves should be placed to facilitate operations and there effluents piped in such a manner to safely ,

protect the operations and maintenance personnel involved as well as afford system protection. ,

, 'C ..

_ . 7) All power stations should be studied to identify similiar system ,

, conditions and appropriate action taken.

,.\.

f v'

.y 1 f. I l

~

s t 4::'

k O,(O

. ,. s l ' 'O r.

4 I

e ,

V 4

E f- d 1

, l , ,1 l

20 + ~i i

I

SUMMARY

OF INVESTIGATIVE ACTIONS 10-16-83 l.

Investigation Committee established IAW PPM.

Committee briefed by J. L. Wilson, Station Manager f-

• Committee briefed by D. A. Christian, Supt. of Operations Examined and photographed scene of accident Reviewed shift logs and preliminary statements of witnesses.

Interviewed and recorded statements of the following witnesses.

, , James J. Simpson - Shift Supervisor /Nuc - SRO Robert P. Matz - CRO (Unlicensed)

Michael C. French - CRO (Unlicensed)

Barry T. Burns - CRO (Unlicensed)- EMT Peter R. DeWitt - Asst. Shift Supv. - SRO

Danny.L. Modlin - Asst. Shift Supv. - SRO i Peter J. Grigonis - Mechanical Foreman l Carol F. Roberts - Mechanic l William G. Henry - Asst. Shift Supv. - SRO

Linwood L. Morris - Shift Supervisor /Nuc - SRO i

f 10-17-83 Interviewed and recorded statements of the following witnesses:

i Timothy L. Kunkle - CRO - RO David A. Christian - Superintendent Operations Initiated research of maitnenance records associated with HP

+

heater drain pumps.

Initiated action for PSE to determine mode of fa!! pre of expansion i joint.

Initiated design engineering study of HP heater drain system by PSE.

Initiated development of a schemctic drawing of HP heater drain

{ system. 5 Station Manager established interim policy of management

approval required before performing any work in the HP

,' heater drain pit.

10-18-83 l

Met with representatives of Virginia Occupational Safety Health

. Administration and briefed them on accident and progress of

.j investigation.

i I

21 i

i

SUMMARY

OF INVESTIGATIVE ACTIONS (CONT'D.)

i i 16-18-83

! Discussed with PSE Chief Engineer details of mode of -

failure study and design engineering study..

, 10-19-83 I

Evaluated physical evidence and testimony of witnesses and developed draft report.

10-20-83 Photographed failed expansion joint after removal from pit.

I Reviewed preliminary conclusions and preliminary proposed recommendations. -

j 16 21-83 i

- ~

Reviewed drain pumps. maintenance records associated with HP hea'.er i

l '

10-25-83 -

Reviewed testimony of witnesses >~

thru 1028-83 - '

Completed Maintenance - Failure history.

Developed and refined report conclusions. ,

Y!

o Developed recommendations. #

[.

. g

~

e o

~'

, .I s

22

,1 l

Exhibit 2D

~

f. - Il

'ik R.: ( '

'?

1n g',

s . 4,-Q.N D i.ph 7 c, A, t 4A

- g ig '.

tqlpf

'~

p

A

?llf,i

\ s j'h I ' ' .he.*

r u

J :n 9

p*- ,

, s . r: -

1. i-I l $ .dM. ., ..,..'

, i%.

;,. v I.- ..- .

j.

n.

f {%%: .-, #I" t t if:.s). '

.' a l dil

g j j; >J

'"t.15, yggl

.q . -

{;: d' d .' j: .?

9 .p P-f, . ; m.gsqf !"'

.

• i .

],g ?Tp ,.. I'

('y <; J', 4

h@v,

.1,9 h 7." L. 3 is l hi$

q f

fi d:s, .c .

~ , I- ,,

i p

. .- . .\ # ,.

s;

, ;i'

>.,, k g',t u -

& ~

_, i '$

(. .,

> 1 y' 8I -

r. . ++ '

( \

ff}j P

]v - .-

1 (PHOTOGRAPH TAKEN 10-20-83 ) -

t l The above photograph shows the Unit 11A (left) and IB (right) discharge lines. In the upper portion of the picture is the 'k

! 1A discharge Isolation manual operator valve.

l .] '

i  !

. , 11

(. . ..

/ / / / s'/ '//

tl//f////////r '

ExnIstr 4 EXPANSION JOlNT =-

p.

J

" N 8

9

/ ,

=. ,

m ~

g M, ,

r.-

f l s. _ + _4 N a C

IA-HP HEATER DRAIN PUMP

[ ,

.a Imd '

4-y .>

l PLAN '

HEATER DRA

[ - REC TAWK ,.

L. + s.

L

,. .g +-

(u) o 1 EiA\L'A' -

. Am y y

I TIE RODS (4)g ,; --

777 y[

.4 r..

m-3 ,.,s t, y..

y o-,

i + l ,.z, n

Jo, ,

. l  :;.h -

y , %s.. ,

e- '

Ai //// '

noy, TIE ROD LUG ATTACHMENT FAILE L -

r. ELEVAT)ON -

T-

~

j O O fl l

m

~

- THIS LUG Puu.ED LOOSE i '

AND WAS 1.YING ON THE

_ Y

FLOOR WITH TIE ROD o O

_'s I g ~.

f

-~

I - BOTH BOTTOM LUGS PultED //// p/

I noW FROM EXPANS10M JOthT

m ec v s.p. r es owATEw SDIOO I4TR , ORAtM RECYR. ,

• \

1 - s O -T W2 1 ttV 6 WHPD-6-3Ol7 j (

o \_

_ i.so ;7]

> / GD-lO6 0-so-n i2 .e so-e c.

IG,-WHPD-t-3Ol% g t.cv e ~

o e S0107

'l6" g o f l0 (I-S D-21]l goE 's 3 7 (1-60531 Q~-so-lG) ~ ~

j g isD 536 -5 16h D-3-301 ., 2 g-sD-7 e}

-Q-so-6 RO ?O b50-lD[' solo 2A (I-S D- S RO F E

solo 2B I*)

p 5 D '#~,\ 5

/-18" l-SD-53) (1-50-535 6 t'WH PD-7 -30 _y O

j pi () Pl Dio16

[(OMDENSE l olo7A 1 21 l l-5013][  % - S D -4') -- IN '8 c P1 d

>s0-531 T Est gp_

T E ST g-s o-3 ) .

]

l- S o-12 ) ##

t Pl I EST t-s o-7-) l Pt 1 lEJ-7A . _

% 24-W D RD 3O l l

l-s 0-10 \{", _3 o- g g }MENB h-SD-l ) _

A

\24" W DRD-1-301

~

1 - sD-P -1 A 1- S D -P -1B .

l 1-4.P. FEEDWATEI? I HEATER ORA)H PUM PS- -

lrAwsu FraoM f:M.-69AD l

i. . '[E* g. ' ~., ; '

~

-s. "

' t- 9,f,C[ f $7,e*:f' ,

^ . - ,

~~

s ..q: ~D. P Tf. - - ' ' ' - - - - - J

j -

PRELIMINARY REPORT HIGH PRESSURE HEATER DRAIN PUMP l EXPANSION JOINT FAILURE SURRY POWER STATION -

. 1 I. INTRODUCTION +

r A failure in a mechanical pressure balanced elbow expansion joint at th's Surry Power Station resulted in the death of a plant operator on October 15, 1983. The expansion joint is located in the suction piping of a ,

. pump utilized to transport high pressure high temper-  ;

ature drains in the secondary plant. The failure occurred while the operator was attempting to reposition . . .

the pump suction isolation valve adjacent to the J.

expansion joint. A sketch of the area is attached.

• A The purpose of this preliminary report is to provide an ,

if.j l'ndependent engineering evaluation of the cause~of pj

~

w failure in the expansion joint. Additional information [j-will be contained in future reports as. data becomes available.

. II. SYSTEM DESCRIPTION 1 A. REFERENCES

, 1. Flow Diagram 11448-FM-19A

2. Valve Operating Number Diagram ll448-FM-69A  ;

3. . Piping Drawings ll448-FP-36A & B ,,

11448-MSK-120A 4

s

. P,ma 1 of 7

4. Logic Diagrams HP Heater Drain Pump ll448-LSK-5-7A f

'HP Heater Drain Receiver 11448-LSK-5-7B -

~

- 5. Machine Location Drawing ll448-FM-6C & E ' !,

)

6. Equipment Specifications a.-- HP Heater Drain Receiver NUS-209

'l

.1 l[

b. HP Heater Drain Pump NUS-125

c. Pump Suctio'n Expansion Joint NUS-177 l i

B. - SYSTEM DESIGN .

>l The High Pressure (HP) Heater Drain Tank receives high '}- t.

pressure high temperature drains from the HP Feedwater , .f I

Heaters and the Main Steam Moisture Separator / Reheaters. ..

.[

.b -

e The tank is vented to the steam zone of the second 1 , .

The tank was constructed by  ?,.T '

point feedwater heaters.

u,.

RECO, reference Specification for Shop Fabricated pg j ,e s Tanks, NUS-209.  ; ..d 9.c+;

x: g:

~q .

The HP Heater Drain Pumps are eleven stage vertical "/j$N)

, y ;y centrifugal pumps with a capacity of 7i00 gpm each. /f[

Each pump develops 860 feet of head. A pump recircula- [ ' h; tion control valve is provided to insure minimum pump

• t flow. The pumps were manufactured by Bingham, refer- ' .' ,

ence equipment specification NUS-125. ,

Each pump is provided with'a pressure balanced elbow ,

~ ; f.

-expansion joint in its suction piping. The expansion joints are employed to absorb differential thermal .

^

expansion.. The mechanical expansion joints were i n.- , - - - , - - , . - - - , -,,r- - - . , , , - - - , , , + n - r-.., ,n-- --- , -- - -. r, , - -gy- g -, , - , - - , - - , ,

manufactured by Pathway Bellows', reference equipment .

specification NUS-177.

i System piping and valves were provided in accordance  ;

7 .

i with Specification NUS-20, pipe class 301. Pump isola- ,

j' s

tion valves are manually operated gate valves with gear assists. y s' ,

C. ' SYSTEM OPERATION )

'. q The two full size HP Heator Drain Pumps, 1-SD-P-1A and ed 1-SD-P-1B, take suction from the HP Heater Drain Receiver !h il 1-SD-TK-2. Tha pumps discharge into the steam generator ,?

h feedwater pump suction header (reference Flow Diagram ' .q j!

ll148-FM-19A). A level controller on the Drain Receiver N.: .

modulates a level control valve, LCV-SD106, in the 4f pump's common discharge line to maintain normal level "$e  !

i -

%i in the Drain Receiver. A high level controllerj diverts HP Heater Drains directly to the condenser on a high

• Yd T

level signal via LCV-SD-107. During normal operation, Mf

.r w one pump runs continuously with the other pump in ', l e

stand-by for automatic starting on a drain tank high i

' s.

level signal. Recirculating control valve FCV-SD100 y opens to provide minimum flow recirculation when the pump discharge control valve LCV-SD106 closes beyond the minimum pump flow requirement. Logic Diagrams J 4

ll448-SLK-5-7A and 7B contain logic and control functions.

,y.

.h

• l Page 3 of 7

f ... &. s e.a p D. CONTROL AND MONITORING DEVICE

SUMMARY

3 Control Board 5 1. Control selector switch for each pump, P.O. -

STOP - AUTO - START, each with RED (running),

GREEN (shutdown) and AMBER (Auto - Trip) indicating lights l

' 2. Annunciator display for HP Heater Drain Pump l AUTO - Trip operation

3. Annunciator display for HP Heater Drain tank HIGH and LOW 1evel

4. Pressure Indicator - HP Heater Drain Pump Discharge

5. Level Indicator - HP Heater Drain Tank

)

6. Annunciator Display - HP Heater Drain Tank Low Level TRIP ' i

• ~

Indicating Ammeter - Motor Current {

7.

i l

Computer ,

l

1. HP Heater Drain Pump Discharge Flow (FT-SD100)

2. HP Heater Drain Pump Discharge Pressure (PT-SD108) ',

3. HP Heater Drain Tank Level (LT-SDill)

III.

SUMMARY

AND CONCLUSIONS

, A. DESCRIPTION OF FAILURE i

' The Surry Unit 1 expansion joint failure is shown in Figure (s)' (later) . With respect to in-place orientation,

e t

t failure occurred on the two bottom lug attachments which hold the tie rods in place (see Figure (later)).

One lug attachment fractured completely from the pipe

~

while the other was approximately a 90% through frac-ture. The side gf the lug opposite the gusset plates was still partially intact while the other sides were completely fractured.'. Both lug attachments were tilted .

back towards the blind flanged end of the expansion joint. This resulted in an approximately 3" x 7" hole in the pipe where each lug had been attached.

f The mode of failure of both lug attachments were very b

similar. Fracture occurred through the gusset welds of each lug attachment and through the pipe base metal surrounding the lug. Visually the failure mode through j the gusset welds appeared to be brittle, although heavy- .)

corrosion deposits made visual observation difficult.

The gusset welds also appeared to be somewhat porous.

The failure mode of the pipe base metal appeared to be predominantly shear. Corrosion had also developed on this fracture surface but was not as heavy as the ',-

corrosion on the gusset weld fracture surface.

~Both tie rods were bent back at the lug attachment region, presumably from the force of the lug attachment failure. One tie rod was sheared completely through near the apex of the bend. It appears that this failure

" men 4 nf 7

r7- 1 i

i i u occurred following the lug attachment failure and is [1 -

i not felt to have been the cause of the expansion joint U

- failure. ,

D. DESCRIPTION OF WORK BY BATTELLE The metallurgical fa(lure analysis of the expansion joint is being conducted by Battelle Columbus Laboratories under the direction of Vepco Power Station Engineering personnel. Dr. Robert D. Stout was hired as a consultant .f

, - l-to Vepco during the failure analysis in order to fully <!

satisfy the work contract between Vepco and Battelle. fi' 1

4:

• After receipt of the expansion joint from Vepco, the c

a joint was photographed from several different perspec- ..

tives. The bellows sheaths were removed to determine 9I'

,',q if any visual damage had been sustained to the pellows. gg

'y)

' Measurements were taken of the tie rod lengths and ,

p..

-flange to flange ~ distances to determine to what degree ]!^

the expansion joint had deformed and to determine 3 whether any overall yielding of the tie rods had taken I 'I 1 -

place. ,

Visual examination of all fracture surfaces did not ,

,e reveal any obvious. initiation point for the fracture.

As previously discussed, the fracture appearance of ,

both lug attachments appeared identical. In light of this, it was decided to have the initial attention of .

' the analysis focus on the completely fractured lug .

'o' nr4M5FE half from the pipe

, 'b' 4,

section.

These fracture portions were sectioned for J

Based on fractographic and metallographic analysis. '

the results obtained following the completion of the

- in-depth ~ analysis of this lug attachment, a decision will be made whether or not an in-depth fractographic d and metallographic analysis is required of the partially ,

fractured lug attachment.

In addition to this work, chemical and physical proper- '

ties of the pipe section were obtained and stress 1[

analysis of the failure was performed. .

M C. PRELIMINARY CONCLUSIONS AS TO FAILURE I

To date (10-28-83, prior to the completion of the , ,[

rv Battelle Laboratory work), analysis of the failure has ,;

The i

been based solely on visual unaided eye evidence.

i.}

fr'acture mode and the location of the fractured. portions r j

l positions with respect to each other and their origina .

}

strongly suggest that the cause of failure was due to a ,

Additional sudden overpressure of the expansion joint.

testing and evaluation is required to confirm this ,;

preliminary conclusion.

l l

l l

l Page 7 of 7

( _

l

( - 24*-WDRD 2-301

)

l .  ;

, u.

MEJ-78 -

~ \ _)  :-

l-SD-TK- 2 . 1-50-1

~

(ABOVE) f l-S D-P - 18 ,

O(,i conc. TANK SUPPORT II l

I l 1Q1 y i

i

/ 7'- 10" J l "

/ \

/ ~

\

's N -

'/-

HAHDWHEEL3 1- S.D- P- 1 A O j

\~ ' _ _ . -' N -'\ _T i l

- l-S0-10 ,

.A i,

(

' 'i o

• m oy MEJ A ,.4 i l

24"-WDRD-1-301 ,

5 '- 8" s i 3" ENT IESTAGATION _QF E'APANStON JOINTSHEET FAtt_URE 1 OF IME i~E COMPOSITE d

ccAt_E : 3/6": l'- O