ML20009E317

Transcript of Jc Elliott,Tc Cheng,Md Weingart,Pa Ranzau & Cg Robertson 810720 Testimony on Tx Public Research Interest Group (Prig) Addl Contention 53,ASLB Questions 4B & 4A & Pirg Contentions 28 & 52.W/prof Qualifications

ML20009E317
Person / Time
Site:	Allens Creek File:Houston Lighting and Power Company icon.png
Issue date:	07/20/1981
From:	Elliott J, Ranzau P GENERAL ELECTRIC CO., HOUSTON LIGHTING & POWER CO.
To:
References
NUDOCS 8107280040
Download: ML20009E317 (19)

Text

,

. i .

July 20, 1981 i } \/ E O .- $

1 1 'JUL 3, bd j ,

UNITED STATES OF AMERICA .

{

NUCLEAR REGULATORY COMMISSION 2 , 7 g g u, 3;.p n t g

L n ;ORI5 Ei1>5EYOIk THE E ATOMIC SAFETY AND LICENSING BOARD I 1 i

3- -

4 In the Matter of S S x , . . -

Docket No. 50-466 N J. W HOUSTON LIGHTING & POWER COMPANY S 5 S (Allens Creek Nuclear Generating S 6 Station, Unit 1) S I V 7 N .

v DIRECT TESTIMONY ON BEHALF v- t67 l Q

L 8 HOUSTON LIGHTING & POWER COMPA.4Y: 2 -

JUL ,", qp#1981 " _ -

9 (1) JAMES C. ELLIOTT - TEXPIRG ADDITIONAL o.s =%M /d CONTENTION 53/NON-CONDENSIBLE GASES ' ' (N 1 - a 10 (2) ROBERT C. CHENG - BOARD QUESTION 4B/ --

11 CONTAINMENT DESIGN BASIS 12 (3) MELVYN D. WEINGART AND CLOIN G. ROBERTSON

- BOARD QUESTION 4A/OOMBUSTIBLE GAS CONTROL 13 (4) PATRICIA A. RANZAU - TEXPIRG ADDITIONAL 14 CONTENTION 28/ POST ACCIDENT MONITORING 15 (5) CLOIN G. ROBERTSON - TEXPIRG ADDITIONAL CONTENTION 52/OUTSIDE CONTAINMENT SAMPLING 16 17 l

18 Q. Would each of you please state your name, your

(

l 19 current position and describe your educational and l

,,a professional experience?

21 A. My name is James C. Elliott and I am employed by l

1 22 the General Electric Company (GE) as Principal Engineer in the GE Nuclear Power System Engineering Department. My 23 l

g educational and professional background is described in 9

8107280040 810720 /f PDR ADOCK 05000466 T PDR

1 Attachment JCE-1, 2 My name is Robert C. Cheng. I have previously 3 explained my position and background in connection with my 4 testimony on Board Question 10.

5 My name is Melvyn D. Weingart and I have previously

, 6 described my position and background in connection with my 7 testimony on TexPirg Additional Contention 30.

8 My name is Cloin G. Robertson and I have previously 9

described my position and background in connection with my 10 testimony on Doherty Contention 8.

yy My name is Patricia A. Ranzau and I am employed by Houston Lighting & Power Company (HL&P) as Lead Engineer, 12 y3 Power Plant Engineering Department, Instrument & Controls.

A description of my educational and professional qualifica-4 tions is described in Attachment PAR-1.

Q. Mr. Elliott, what is the purpose of your testimony?

A. The purpose of my testimony is to address TexPirg i 17 l

AC 53 which states:

18

" Applicant should commit to a system to ascertain 19 accurately how much non-condensible gas is in the reactor vessel, to assist in estimating the possible 20 explosion hazard in the vessel during an ECCS. The need for this information was demonstrated at Three 21 Mile Island, Unit 2, during its recent incident.

Petitioner contends that inability to know accurately 22 the amount of non-condensible gas in the reactor increases the chance of an explosion and damage to 23 the fuel geometry and/or physical breaking of fuel rod clad."

24 1

l l

l

1 In the deposition of Mr. Clarence Johnson taken on 2 February 27, 1980, ho states that TexPirg's concern is

about ACNGS's ability to monitor hydrogen during a small 4 break LOCA. From this it is inferred that when the 5

contention says non-condensible gases, that the chief concern 6

is hydrogen and that "during an ECCS" really means "during a small break LOCA".

7 Q. Does the Applicant have a system which monitors the amount of non-condensible gas in the ACNGS reactor vessel?

A. The Applicant has the capability of monitoring the potential for formation of non-condensible gases in the ACNGS 11 reactor by using the reactor water level indicators. It is a 12 necessary condition that the reactor core become uncovered 13 before any large scale metal-water reaction can occur to 14 produce hydrogen. The reactor water level instrument at ACNGS 15 i

gives a direct reading of the reactor water level and thus 1

s l

would give the reactor operator the information needed in l 17 order to take appropriate actions. These actions are described l

' 18 in the operator emergency procedure guidelines, wherein the 19 operator is directed to depressurize the roactor using the 20 safety / relief valves if an automatic depressurization has not 21 occurred, and if the operator is unable to restore reactor 22 water level with the high pressure pumps (feedwater, RCIC, 23 HPCS, CRD). This action delays core heatup by inducing 24

1 steaming cooling, and allows low pressure pumps (4 condensate 2 pumps, 3 RHR pumps, 1 LPCS pump) to begin injecting water into the reactor. In the event water level is not restored, 3

4 any hydrogen generated would be vented immediately to the suppression pool. At TMI, neither reactor water level infor-5 6

mation nor the ability to rapidly depressurize and vent the reactor vessel, was available to the operator.

7 0 Is it important to know the amount of hydrogen in 8

• "**** " # # "" ""*E * " ***# "" "

9 reactor?

A. No. In general, all appropriate actions can be taken based on reactor water level. Additionally, knowing the exact amount of hydrogen is not necessary since there 13 will be insufficient oxygen in the reactor vessel to 14 support a hydrogen combustion. This conclusion is consistent 15 with the final analyses of the NRC and other experts relative 16 to the TMI accident. For a small break LOCA the primary 17 source of oxygen would bo due to radioly',is. Radiolysis is 18 the decomposition of water into hydrogen and oxygen under 19 the influence of high energy radiation. In the presence of 20 an excess of hydrogen, the recombination reaction is 21 faster than e decomposition reaction, so no net production 22 of hydrogen or oxygen occurs. Thus the free oxygen i

23 concentration would not increase due to radiolysis, nor 24

P 1 would it reach the flammability limit of 4% nor the 1

2 detonation limit of 9% . Also, it must be kept in mind 3 that hydrogen can be vented at any time during operation 4 of the plant.

5 0 Please describe how the Allens Creek reactor vessel 6 provides for venting of non-condensible gases?

A. A discussion of reactor vessel vent paths is pro-7 g

vided in Appendix 0 of the ACNGS PSAR in response to Item

• • • "' ' "U" *~0 '

"' "9 ***" ** '" * ** '*" "9 9

Applications for Construction Permits and Manufacturing License". Venting capability of the ACNGS reactor vessel is addressed in two parts:

1. Up to the main steam line nozzles: Venting 13 this region can be accomplished by opening any one of 14 the 19 safety relief valves on the main steam lines, 15 some or all of which may already be open depending on 16 These valves j the mode of core shutdown cooling in use.

17 and their operators are safety grade, seismically and 18 environmentally qualified for accident conditions, 19 have position indication in the control room (see 20 testimony on Doherty Contention 42), and are powered 21 from the onsite electrical system and operable from the 22 23

-1/

NSAC-80-1, March 1980 - Analysis of TMI-2 Accident.

24 r-1 control room. Eight of the valves,' which are part of 2 the ADS, have redundant safety related air and electrical 3 supplies. In addition, this region of the vessel can be 4 vented through the RCIC steam supply line Nhich connects to main steam line A, without opening the SRV's. This 5

6 path is through the RCIC steam turbine exhaust, which 7

discharges to the suppression pool.

8

2. Above the main steam nozzles: There are two means of venting this space:

9

a. Venting this area Can be accomplished via the normally open reactor head vent line and valve B21-F005, which discharges to main steam line A.

The non-condensible gas can then be vented to 13 the suppression pool / containment via any one of 14 three safety aJi.ef valves; two of which have safety 15 related air supplies.

l 16

b. Venting this area can also be accomplished via the l

17 l

i normally closed reactor head vent line and series l 18 valves B21-F001 and B21-F002, which discharge to l

19 This vent the drywell high purity drain tank.

20 path is normally only used during shutdown and 21 startup. These valves are safety grade and their 22 operators are Class IE, and are seismically and t

23 environmentally qualified. They are operable from 24 1

l

1 the Main Control Room.

All of the above venting paths lead to the containment 3 via the suppression pool. The control of hydrogen in 4 the containment is discussed in the response to Board j 5 Question #4A.

6 Q. Mr. Cheng what is the purpose of your testimony?

7 A. The purpose of my testimony is to. address Board 8 Question 4B wherein the ard requested HL&P to present 9 evidence that Allens Creek will meet current NRC requirements 10 with respect to GDC-50, Containment Design Basis and to 11 evaluate small pipe break consequences in the ECCS. Small 12 Pi Pe breaks in the ECCS were discussed in HL&P testimony by 13 Mr. Pappone on Board Question 4B, on June 1, 1981. Mr.

14 Weingart and Mr. Robertson will address Board Question 4A which deals with the question of whether ACNGS will meet the 15 NRC requirements for combustible gas control.

Q. With respect to Board Question 4B, what does General Design Criterion (GDC) 50 require?

A. In general, the current revision to GDC-50 (dated October 27, 1978), requires the following:

20 CRITERION 50-Containment Design Basis 21 The reactor containment structure and associated support 22 systems shall be designed so that the containment structure 23 can accomodate without exceeding the design leakage rate, the 24 l

l

1 calculated pressure and temperature conditions resulting from 2 any loss-of-coolant accident. The design margin shall reflect 3 consideration of (1) the effects of potential energy sources 4 that have not been included in the determination of the peak 5 conditions such as required by 50.44 (energy from metal-6 water) and other chemical reactions that may result from 7

degraded emergency core cooling functioning; (2) the limited experience and data available and (3) the conservatisms in 8

the calculations.

9 Q. Does the Allens Creek PSAR make a commitment to meet General Design Criterion 50?

A. Yes. The Allens Creek PSAR Section 3.1.2.5.1.1 12 commits to meet the requirements of General Design Criterion

50. The design temperatures and pressures set forth in PSAR 14 Section 3.8.2 exceed the pressures and temperatures associated 15 with accidents described and evaluated in Chapters 6 and 15 16 of the PSAR, including the requirements of 10 CFR 50.44.

17 l Q. What are your conclusions concerning Board Question 18 4B, compliance with GDC-50, Containment Design Basis?

19 The ACNGS design addresses and complies with the A.

20 requirements of GDC-50.

21 Q. Mr. Weingart, with respect to Board Question 4A, 22 what regulation governs the current requirements for 23 combustible gas control?

24 i

l 1 A. The requirements for combustible gas control are 2 found in 10 CFR 550.44.

3 Q. What are the requirements of 10 CFR 550.44 and how l 4 are these requirements met for Allens Creek?

l 5 A. In general, the requirements of 10 CFR 550.44 are the following:

6 i 1. A means shall be included in the plant design, for 7 the control of hydrogen gas generated after a LOCA due to the metal-water reaction of the fuel cladding 8 and coolant, the radiolytic decomposition of coolant, and the corrosion of metal.

2. The plant design shall include, following a LOCA, a

10 means to measure hydrogen concentration in the containment, a means to insure a mixed containment 11 atmosphere, and a method of controlling combustible gas concentration.

[ 3. Evaluations shall be performed to verify that

^ 13  : following a LOCA but prior to operation of the i combuse.ible gas control system that an uncontrolled

14 hydrogen-oxygen recombination would not take place i

or the plant could withstand the consequences of 15 uncontrolled hydrogen-oxygen recombination.

(

I 16 4. The amount of hydrogen contributed by the metal- .

water reaction resulting from a degraded emergency 17 core cooling function shall be assumed to be either five times the total amount calculated in 10 CFR S50.46 (b) (3) or the amount resulting from a 18 reaction of all the metal in the outside surfaces f the cladding cylinders surrounding the fuel 19 (excluding cladding around the plenum volume) to a depth of 0.00023 inches, whichever is greater.

20

5. A combustible gas control system, such as recom-21 biners, shall be required. A purging system shall n t.be the primary combustible gas control means.

2 Q. Does the ACNGS design comply with 10 CFR 550.44?

A. Yes. PSAR Section 6.2.5.1 commits Allens Creek to 24 1 meet the requirements of Regulatory Guide 1.7, Control of 2 Combustible Gas Concentrations in Containment Following a 3 Loss-of-Coolant-Accident. This regulatory guide provides 4 requirements which are to be followed in the calculations, t

5 analyses and design features pertaining to post-LOCA hydrogen .

l 6 generation.

7 As described in PSAR Sections 6.2.5.2.2 and 8 7.5.1.4.2.11 (d), ACNGS will have the capability to sample 9 and measure the hydrogen concentration throughout the drywell 10 and containr. int during an accident. This system consists of 11 redundant sample and return lines, isolation valves, hydrogen 12 analyzers, sample cylinders and pumps. The hydrogen volume 13 percent is recorded and alarmed in the control room.

14 A drywell mixing system is described in PSAR 15 Section 6.2.5.2.3. PSAR Sections 6.2.5.3.3.3 and 6.2.5.3.3.4 f

describe the turbulent convective mixing process of hydrogen 16 that occurs following a LOCA. These two methods of mixing 17 ensure there is an adequately mixed atmosphere in the 18 containment and drywell following a postulated LOCA. Combust-g ible gas concentrations in containment following a LOCA are l controlled with an electric hydrogen recombiner system.

This system as described in PSAR Sections 6.2.5.2.1, 6.2.5.2.4 and 6.2.5.2.5, operates on the principle of natural con-vection of warm air rising through the recombiner. As a 24 1 secondary or back-up system to the hydrogen recombiners, a 2 hydrogen purge system consisting of fans, ductwork and 3 filters, is available in the ACNGS design as an additional 4 containment hydrogen control means.

5 As required by Regulatory Guide 1.7, a 4 percent or 6

lower by volume hydrogen in air or steam-air atmosphere shall 7

be maintained post-LOCA in containment. The hydrogen recom-g biner system is designed to maintain the containment below this 4 percent lower flammability limit.

9 As described in PSAR Section 6.2.5.3, design O

evaluations have been performed by GE to determine the quantity of hydrogen released post-LOCA due to the metal-water reaction.

13 -

Q. Mr. Robertson, the question presented in Board 14 Question 4A was raised prior to the TMI incident. Are there 15 any requirements pertaining to combustible gas control that 16 have evolved as a result of the TMI incident?

17 A. Yes. As NRC guidance for compliance with 10 CFR 18 S50.34 (e), a proposed rulemaking, NUREG-0718, Item II.B.8, l 19 i parts (3), (4)b and (4) c require that a hydrogen control 1

20 system be provided to control the amount of hydrogen generated 21 by a 100 percent active fuel-clad metal-water reaction that 22 has been postulated to occur from a degraded core accident.

23 Q. Has HL&P taken steps to assure'that this guidance 24 l

l 1

s 1 has been incorporated into the ACNGS design?

2 A. HL&P has committed to provide a hydrogen control 3 system in the ACNGS design. Currently a number of different 4 methods are being considered throughout the industry und it 5 is expected that these efforts will produce valuable data 6 upon wi.~ch to select an optimum means of hydrogen control.

Further, it is expected that the pending rulemaking on de-7 8

graded cores will determine the necessity for such a system.

9 For the purposes of meeting the stated requirement, a post-accident inerting system using CO2 as an inerting agent 10 is being designed. However, for the reasons given above the need for this system will be under continuing review.

Q. Mr. Robertson, while we are on the subject of TMI, TexPirg Additional Contention 52 alleges that because of the TMI incident ACNGS should be designed Jo that reactor coolant 15 samples can be taken from outside the containment. Does the

16 .

Allens Creek design' include the capability to sample reactor 17 coolant from outside containment?

18 A. Yes. As indicated in Appendix 0, page 0-79 to 19 the PSAR, Allens Creek will have the capability to collect 20 liquid samples from the reactor coolant and suppression pool 21 from outside the containment building. This commitment was 22 made in response to Item II.B.3 of NUREG-0718, " Licensing 23 Requirement for Pendina Appli ,tiens for Construction Permits 24

r 1 and Manufacturing License," which requires us to be able to 2 collect reactor coolant samples assuming extensive fuel 3 damage and contamination of the containment building.

4 Q. Would you please briefly describe the sample 5 system to be used?

6 A. Final design for the post-accjdent sampling system 7

is not required until the OL stage but as it is now planned 8

the sampling system will be a manual system that can be 9

operated to collect a liquid sample. The sampling station will be located in the Reactor Auxiliary Building near the 10 Reacto,r Shield Building wall, in order to make the sample lines as short as possible. Shielding will be provided to ensure that doses to those personnel collecting the samples are maintained within GDC 19 as specified in NUREG-0737.

14 Q. Ms. Ranzau, what is the purpose of your testimony?

15 A. The purpose of my testimony is P.o address another .

16 TMI related contention. I will address TeaPirg AC 28 con-17 tention which alleges that:

163 "The control room design and the post-accident 19 display instrumentation for Allens Creek plant are not sufficient to ensure that the operators can 20 safely control the plant under all accident con-ditions. As at Three Mile Island, the operator may make one or more critical mistakes because of 21 defective instruments or their location in the control room."

22 Q. Has the layout of the control room for Allens Creek 23 been designed to consider the " location of instruments?

24

1 A. Yes. The location of instrumentation was an 2 explicit part of the design evaluation process for the main 3 control room boards. As described in the post-TMI 4 requirements, Appendix O of the ACNGS PSAR' Item I.D.1,

, the 5

control room utilizes human factors principles in the design.

ACNGS will use the General Electric Company NUCLENET/1000 6

7 Control Complex. NUCLENET is a generic design utilizing CRT based displays and is the most advanced state-of-the-art 8

design. The design was based on a methodology virtually 9

identical to that set out in Appendix B to NUREG-0659, which 0

reflects current NRC recommendations for control room design.

The design process included an analysis of all functions necessary to operate the plant safely, an allocation of functions between operator and machine, and a qualitative 14 verification of the functional allocation. GE assembled a 15 team of experts to design the NUCLENET/1000 Control Complex 16 which included experts in controls and control systems 17 design, computer technology, industrial design, operator 18 training, power plant test and operations, and behavioral 19 The premise upon which the design is based is that

! science.

0 optimum control is achieved when there is an allocation of 21 control functions between the operator and machine which 22 recognize that each performs certain functions better than 23 the other, and that, once the allocation is made, the design 24 permits efficient and effective manipulation of controls

,im- +-

l 1 by the operator. The balance of plant controls were also 2 integrated into the NUCLENET complex on the basis of guide-3 lines developed by GE.

4 Q. Has HL&P reviewed the ACNGS design to assure 5 compliance with current regulatory guideIines?

6 A. Yes. A design review was conducted of a full-scale 7 mockup of the control room with an interdisciplinary team g of instrumentation and control engineers from HL&P, Ebasco and Brown & Root. We also had a human factors consultant 9

fr m MIT and two plant operating personnel from HL&P.

10

.Q. Could you explain how this design review was performed?

A. As provided in Appendix O, Item I.D.1, of the ACNGS PSAR, the Allens Creek Control Room Evaluation Task Team 14 performed a preliminary assessment of the ACNGS Unit 1 control 15 room for the purpose of identifying additional human factors

. 16 -

engineering design conditions that could provide a basis for 17 further improvements. HL&P's instrumentation and controls 18 engineers assigned to Allens Creek built a full size mockup 19 of the front row panels of the NUCLENET/1000 Control Complex.

20 A human factors engineering evaluation was performed on the 21 mockup. The control room was evaluated as is, without 22 regard to planned changes in layout and changes required as 23 a result of Three Mile Island. The evaluation consisted of 24 e

1 the application of human factors engineering design check-2 lists. The checklists were developed and prepared by the 3 General Electric Boiling Water Reactor (BWR) Control Room 4 Owners Group following the TMI incident.

5 Q. What were the conclusions drawn from this prelimi-6 nary review?

7 A. The Allens Creek control room uses advanced tech-g nology which incorporates human factors engineering principles in design and operation. A control room design review has 9

been conducted based on a full-scale mockup, deficiencies 10 identified, and enhancement activities are being undertaken.

Ccapletion of the enhancements will result in a control room which meets the intent of NUREG/CR-1580 and NUREG-0659.

13 -

Results of these actions will be forwarded to NRC upon 14 completion.

15 Q. Have any specific design changes been made in the

• 16 control room to assist the operator in responding to accident 17 conditions?

18 A. We have added mimics, additional lines of demarca-l tion, uniformity of abbreviations and nameplate locations to 20 We have added an additional system the control room panels.

21 called the Safety Parameter Display System (SPDS). The SPDS was added in direct response to NUREG-0718, Item I.D.2.

23 Q. Could you briefly describe the SPDS?

24

r 1 A. The SPDS is described in detail in ACNGS PSAR 2 Section 7.5.1.6. The system will have the capability of 3 displaying the full range of important plant parameters and 4 data trends on demand. The system will also indicate when 5 plant parameters are approaching or exceeding process limits. .

6 The SPDS will be designed in conformance with the guidance 7 of NUREG-0696, February 1981, Section 5. The SPDS will be 8 a computer-based system of high quality and reliability. It 9 will be capable of functioning properly in the environments 10 that are present during transient and accident conditions.

11 Q. Now that we have discussed how the control room 12 and SPDS have been designed to -include not only post-TMI 13 requirements, but specifically human factors, how do these 14 two main safety parameter areas meet current NRC requirements f r instrumen:ation reliability?

15 A. The current NRC requirements with regard to safety-6 related instrumentation reliability are described in 7

g Regulatory Guide 1.97, Revision 2. This regulatory guide provides the current post-TMI instrumentation requirements.

Conformance with this regulatory guide is discussed in the ACNGS PSAR Appendix C and also in Appendix O, Item II.F.3.

As discussed in Appendix C of the ACNGS PSAR, safety-related instrumentation used in the ACNGS design will consist 23 of reliable, state-of-the-art devices.

24

e Attachment JCE-1 James C. Elliott I graduated from the University of Idaho in 196C with a Bachelor of Science Degree in Mechanical Engineering.

I received a Master of Science in Mechanical Engineering in 1966 from Stanford University. I am a professional engineer in the State of California.

I served in the U.S. Navy from June 1960 to September 1963. From November 1963 to September 1965 I was employed by the General Electric Company as a Turbine Generator Startup Engineer. In this capacity.I provided technical direction and supervision for installation, maintenance, and repair of steam turbine-generators, stationary gas turbines, and marine propulsion turbines. From September 1966 to June 1968 I was employed by GE as a Warranty Service Engineer.

In this capacity I administered nuclear power plant, nuclear fuel, and consulting contracts with utilities having operating GE nuclear reactors. From June 1968 to November 1972 I was employed by GE as a plant test engineer. During this period I first worked for five months as an instructor at the BWR Training Center in Morris, Illinois. Following that, I worked as the Simulator Specialist at the BWR Training Center. Next, I was Shift Supervisor for the startup of the KKM Nuclear Power Plant in Switzerland. From November 1972 to March 1976 I was employed by GE as a Senior Project Engineer. During this time I was the Resident Project Engineer in Birmingham, Alabama on the Barton Nuclear Power Plant project. Since March 1976, I have worked in various engineering capacities in the GE Nuclear Power Systems Engineering Department.

Initially I was responsible for developing and communicating recommendations to architect-engineers and utilities per-taining to BWR shutdown oxygen control and feedwater quality.

In this position I developed a system for onygen control in BWR's during shutdown, and co-authored several publications l

on BWR shutdown oxygen control and feedwater quality. In my current capacity, I am responsible for management of GE's " Post-TMI Hydrogen Control" program. I assisted in the development of a " Post Accident Inerting" hydrogen control system for Mark III containments.

e

Attachment PAR-1 Patricia A. Ranzau I received a BS in Math from Texas A&M University in 1972. I received a BS in Industrial Engineering from Texas A&M in 1973.

I have been employed by HL&P since September 1973.

I have worked entirely as an engineer in the Instrument and Controls group of the Power Plant Engineering Department..

I have had various responsibilities within that group and I am now Lead Engineer in the group. I also serve as I&C Team Leader for Allens Creek with responsibility for design of all nuclear and balance of plant instrumentation and control.

In addition to my normal duties I have participated in several workshops and courses in control room design with particular emphasis on human factors design. I also partici-pate in regular reviews of NRC and industry regulations and guidelines.

_ , _ . . . . - - , - -- , , . . . . , . - _ . _ . . _ , . ,. - --- -