ML20151A582
| ML20151A582 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 07/13/1988 |
| From: | Curran D, Tousley D HARMON & WEISS, NEW ENGLAND COALITION ON NUCLEAR POLLUTION |
| To: | Atomic Safety and Licensing Board Panel |
| Shared Package | |
| ML20151A561 | List: |
| References | |
| RTR-NUREG-CR-4728 ALAB-886, OL-1, NUDOCS 8807200051 | |
| Download: ML20151A582 (79) | |
Text
. .
, July 13, 1988 UNITED STATES NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD
)
In the Matter of )
)
Public Service Company of )
New Hampshire, et al. ) Docket Nos. 50-443 OL-1
) 50-444 OL-1 (Seabrook Station, Units 1 & 2) ) ONSITE EMERGENCY
) PLANNING & TECHNICAL
) ISSUES NEW ENGLAND COALITION ON NUCLEAR POLLUTION'S MOTION FOR CLARIFICATION OR RECONSIDERATION OF THE BOARD'S ORDER OF JUNE 23. 1988 Introduction In a telephone conference dated June 23, 1988, the Board issued an order addressir.g, inter alia, the scope of issues to be litigated with respect to RG-58 coaxial cable. The Board stated that Applicants must prove that RG-59 coaxial cable is a "techni-cally acceptable replacement" for RG-58 coaxial cable. Tr. at 1179. However, the Board also barred NECNP from litigating the environmental qualification of RG-59 coaxial cable. Id. NECNP seeks clarification from the Board that while the Coalition is barred from raising the question of whether RG-59 cable is qualified for its original applications, the term "technically acceptable" embraces the concept that the RG-59 cable must be demonstrably qualified to meet the specific performance criteria for the twelve applications in which it is to be substituted for RG-58 cable. If the Board did intend to prevent NECNP from litigating the question of whether RG-59 coaxial cable is environmentally qualified for the 12 applications in which it is l$k $DO k h3 G
-r N l 1
l l
to be substituted for RG-58 coaxial cable, NECNP asks the Board to reconsider its position.
Araument The Licensing Board has barred NECNP from litigating the issue of whether qualification tests applied to RG-59 coaxial cable were sufficient to qualify that cable. Tr. of June 23 con-ference at 1179-80. This decision is based in part on the Licensing Board's reading of ALAB-886, in which the Appeal Board rejected NECNP's late-filed contention on the environmental' qualification of RG-59 cable.
Although we continue to believe that the RG-59 coaxial cable has not been demonstrated to be qualified for any application or use, we do not seek to resurrect here the contention that was rejected by the Appeal Board. Rather, NECNP seeks to clarify and establish that the litigation now underway embraces the question of whether the RG-59 cable is qualified for purposes of meeting the performance specifications for the twelve applications in which Applicants intend to substitute RG-59 coaxial cable for RG-58 coaxial cable. Egg Applicants' Suggestion of Mootness, filed May 27, 1988, at 3.
Under NRC regulations, the concept of environmental qualification includes both an understanding of the environment to which the equipment will be subjected and the performance specifications which must be met in order for the equipment to function satisfactorily in its specific applications. The Appeal
Board recognized the requirement in ALAB-891, when it stated that "before a nuclear facility uses for a particular purpose a com-ponent subject to the environmental qualification requirements, it must be demonstrated that that component meets those require-ments when an emoloved." ALAB-891 (April 25, 1988), slip op. at 25 (emphasis added).
Thus, in addition to describing the environmental conditions to which a component may be subjected during an accident, NRC regulations at 10 CFR 5 50.49(d) require operating license l applicants to specify:
1 (1) The performance specifications under conditions existing during and following design basis accidents.
(2) The voltage, frequency, load, and other electrical characteristics for which the performance specified in -
accordance with paragraph (d) (1) of this section can be ensured.
These requirements are further explicated in IEEE STD 323-1974, as supplemented by the provisions of Reg. Guide 1.89, which sets forth acceptable procedures for the satisfaction of the commis-sion's environmental qualification rule. See Reg. Guide 1.89 at 1.89-2, IEEE STD 323-1974, Section 6.2.
l The equipment qualification file for RG-58 coaxial 1
cable and RG-59 coaxial cable contains no performance criteria l
I 1
e f
I
I l
1 for either type of cable.1 Without knowing the performance spec-ifications required of the RG-59 coaxial cable in the particular substitute' circuits in which it will be used, and without knowing what minimum insulation resistance and other electrical charac-teristics of the RG-59 coaxial cable are necessary in order to ensure that those performance specifications will be met, there is no valid basis for determining whether the RG-59 coaxial cable is environmentally qualified for use in the circuits that presently use RG-58 coaxial cable, or that the RG-59 coaxial cable is technically acceptable as a substitute for the RG-58 coaxial cable.
This matter was previously brought to the Seabrook Licensing Board's attention by the Staff in Board Notification 84-032, dated February 13, 1984 (Memorandum for Addressees on Attached List from Darrell G. Eisenhut, Director, Division of Licensing, Office of Nuclear Reactor Regulation, re: "Additional Informa-tion on Environmental Qualification." The Board Notification raised a number of concerns about environmental qualification programs that were brought to the Commission's attention by 4 1 In environmental qualification testing of the RG-59 coaxial cable, the cable was tested against certain performance require-ments for insulation resistance. NECNP took these to be environ-mental qualification performance requirements, and cited them in its motion for leave to file a late-filed contention on the RG-59 coa :ial cable. However, Applicants later asserted that these insulation resistance requirements were procurement specifica-tions rather than acceptance criteria. See ALAB-886, slip op. at 8.
Sundia National Laboratories ("5NL"). Attached to the Board Notification as Enclosure ? was SECY-83-457C, dated January 18, 1984, which included an aftachment entitled "NRC Response to Issues Raised by Sandia National Laboratories During NRC Commis-sion Meeting of January 6, 1984." (The cover memorandum of the Board Notification and SECY-83-457C and its enclosures are attached to this motion as Attachment A.)
Among SNL's concerns was the following:
Qualification test procedures and/or requirements do not always reflect application conditions, such as Acceptance criteria for coaxial and triaxial cables are not documented as related to use conditions as identified in Vendor Inspection Program Docket 99900277.
SECY-83-457C, Attachment 2, "NRC Response to Issues Raised by Sandia National Laboratories During NRC Commission Meeting of January 6, 1984," at 6 (Item 2A(b)). The NRC responded that:
Applicants and Licensees in their review must ascertain that the test acceptance criteria is applicable to the end use of the equipment. Specifically:
Item 2A(b) - If the acceptance criteria are not docu-mented nor reviewed then the equipment (coaxial and triaxial cables) is not orocerly cualified.
Id. at 7 (emphasis added). Thus, the Commission emphasized that in addition to the accident environment, operating license applicants must also describe what performance specifications must be met for each specific application of the cable and
1 demonstrate that the cable meets those specifications when sub-jected to the accident environment.
This point was recently illustrated in environmental qualification research conducted by SNL and reported in NUREG-CR-4728, "Equipment Qualification Research Test of a High-Range Radiation Monitor," dated February 1988 (Attachment B) . Based on a test in which a cable and radiation detector which had been individually tested and deemed to be environmentally qualified, l but which failed when combined together in a system, SNL warned l
l that "utilities should be reminded that installations of com-ponents and systems qualified by parts may not constitute a qualified system." Id. at 47.
In conclusion, it is impossible to determine whether a com-l ponent is environmentally qualified without an understanding of each of the specific applications for that component. While NECNP may be barred from litigating the qualification of RG-59 coaxial cable in its original applications, the cable's qualification for its replacement applications is a central issue in this case. We ask the Board to clarify this point; or in the alternative, to reconsider its position. To hold that NECNP may not challenge the environmental qualification of RG-59 coaxial cable with respect to its specific substitutions for RG-58 coaxial cable would be inconsistent with NRC regulations and would prejudge the acceptability of the RG-59 cocxial cable as a substitute for RG-58 coaxial cable.
l
Respectfull submitted, OcAm o.
2)e% 754r/cr /Oc Diane Curran Dean R. Tous ey HARMON & WEISS 2001 "S" Street, N.W.
Suite #430 Washington, D.C. 20009 (202) 328-3500 July 13, 1988 CERTIFICATE OF SERVICE I certify that on July 13, 1988, copies of the foregoing pleading were served by hand, overnight mail, or first-class mail on all parties to this proceeding, as designated on the attached service list.
7)d~a C-Diane Curran I
l l
l l
l-l
- I .4 Attachment A h
I i
l l
t l
l l
l l
l
9 per#*,, UNITED STATES
.1 e y ,. . . .
[ g NUCLEAR REGULATORY COMMISSTON i'l ~ . . _
t l wAsHiNoroN. o. c. zosas
\ ***** / -
February 13, 1984
.g MEMORANDUM FOR: Addressees on Attached List FROM:
- Darrell G. Eisenhut, Director l Division of Licensing i Office of Nuclear Reactor Regulation l
SUBJECT:
ADDITIONAL INFORMATION ON ENVIRONMENTAL QUALIFICATION l (BOARD NOTIFICATION 84-032)
In accordance with present NRC procedures regarding Board Notifications, the enclosed is provided for your information. This infonnation is applicable to all nuclear power plants. By copy of this memorandum, we are notifying appropriate boards and parties and the Comission.
The enclosed letters provide additional information regarding equipment quali-fication. The enclosed letters augment the information found in Board Notifi-cations84-004 and 84-007 and discussed in the January 6,1984 Comission meeting on equipment qualification. -
The staff will keep you informed regarding the resolution of this issue.
DN ML Darrell G. E s.enhut, Director :
Division ofL icensing Office of Nuclear Reactor Regulation
Enclosures:
- 1. Memo WJDircks to Comission dtd 2/2/84
- 2. Comission Paper SECY 83-457B dtd 1/10/84 8401190059
- 3. Comission Paper SECY 83-457C
- dtd 1/18/84 cc
- See next page I
- e. .-
r
~
Addressees: The Atomic Safety and Licensing Boards for:
Byron 1 & 2 (Miller, Callihan, Cole)
Catawba 1 & 2 (Kelley, Foster, Durdom)
- Comanche Peak 1 & 2 (Bloch, Joroan, McCollum)
. Limerick 1 & 2 (Brenner, Cole, Morris) .
tiidland 1-& 2 (Beckhoefer, Cowan, Harbour)
. Perry 1 & 2 (Bloch, Bright, Kline)
Seabrook 1 & 2 (Hoyt. Harbour, luebke)
Shoreham (Brenner, Caroenter, Morris)
Uolf Creek (Wolfe, Anderson, Paxton)
,Zimer (Frye, Hoooer, Livingston)
The Atomic Safety and Licensing Acceal Soards fer:
Diaolo Canyon l a 2 (ftoore, Johnson, Buck)
aterford 3 (Kohl, Johnson, Hilber) 4 k
Enclosure 3 g@ *"v e
~~
January 18, 1984 y lgg SECY-83-457C (Information)
M: The Commissioners From: William J. Dircks, Executive Director for Operations
Subject:
DISCUSSION /POSSIBLE VOTE ON EQUIPMENT QUALIFICATION POLICY AND PROPOSED RULE; NRC RESPONSE TO COURT OF APPEALS DECISION Purcose: To provide the Commissioners with information on the status of resolution of Sandia concerns about the NRC's EQ program and other matters.
Discussion: In the Commission Meeting of January 6, on the subject agenda item, A. W. Snyder and D. A. Dahlgren of Sandia discussed Sandia concerns regarding the NRC's EQ program, fire protection, and pressures Sandia perceives it has experienced in conducting research programs. Sandia's written explanation of their con-cerns was transmitted to the Commission by my memorandum on this subject of January 10, 1984.
Per the staff requirements memorandum to me from John C. Hoyle dated January 10, 1984, the staff and Sandia have jointly addressed Sandia's concerns with the exception of the issue of timing of the release of research results in implementation of foreign information exchange arrangements. This issue will be the subject of a separate Commission background paper.
The staff prepared responses to er.ch of Sandia's concerns and discussions were held with appropriate Sandia staff to assure that these concerns have been correctly interpreted and are being addressed. In some cases the staff's responses were modified to reflect feedback from Sandia. The resolution basis includes a commitment on the part of the staff for additional discussions on the subject of NRC pre-approval of Sandia travel. Dr. Dahlgren has stated in Enclosure 1 Sandia's position regarding the~ staff responses (Enclosure 2) to Sandia concerns.
CONTACT:
B. Morris, RES 44-37946
The Comissioners 2 Dr. Dahlgren has specifically stated his belief that all Sandia ~
issues relatcd to environmental qualification of electrical eouipment and the E0 rule have been satisfactorily addressed in the staff responses. It should be noted that although there was agreement between Sandia and staff that the fire protection issues raised were being addressed as documented in Enclosure 2, these issues are not within the scupe of the subject policy issue or the EQ rule.
During our January 12, 1984 meeting with Sandia, they reviewed the Duke Power 0.G. O'Brien Test Report and have since notified us of concerns regarding the way the test was conducted. A copy of this report has been sent to Sandia and in addition, this report is scheduled to be reviewed by our staff consultants for the Catawba licensing review. The staff will keep the Comission informed on this item.
b J William J. Dircks Executive Director for Operations o Enclosuras: -
- 1. Ltr to W. J. Dircks frm D. A. Dahlgren, SNL, 1/12/84
- 2. NRC Response to Issues Raised by SNL During NRC Comission l Meeting of January 6, 1984 l
l l
l
i<
- jl '*
\,
JAN 121984 Mr. William J. Dircks Executive Director for Operations U.S. Nuclear Regulatory Commission Washington, D.C. 20555
Dear Mr. Oircks:
The NRC staff and Sandia representatives have met and discussed the issues and concerns. raised by Sandia with Chairman Pal-ladino and at the NRC meeting of January 6, 1984. The staff appears to have understood the concerns of Sandia, and all parties have come to a consensus understanding of the issues raised.
Sandia agrees that the staff has addressed or is addressing the issues (concerns) i raised by Sandia. We believe'that all the issues we raised which directly or indirectly relate to environmental qualification of electrical equipment in nuclear plants, as defined by the EQ Rule 10 CFR 50.49, have been addressed.
This is based on the Sandia review of the NRC staff responses to our concerns which are Enclosure 1.
Sincerely, N
l 0. A. Dahlgren
Enclosure:
- 1. NRC Response to SNL Concerns l
l l
A
(
A NRC Resconse to Issues Raised bv Sandia National Laboratories During NRC Commission Meeting of Januarv 6, 1984 l
l l
1 i
l l
s
s .
2
- 1. Qua'ifi, cation Methodologies Have Shortcominos Work has been performed under NRC research on many of the items indicated in Attachment A. The results indicate shortcomings are present in some of the current criteria and qualification methodologies as currently practiced. Where issue determination is complete and a good data base has been established, the NRC is acting to revise the relevant rules, regulations or other guidance. ,
i a
.o 3-Attachment A o- Should LOCA be simulated by sequential or simultaneous exposure to steam and radiation?
o Can gamma sources adequately simulate effects _of. beta radiation? -
o Is it necessary to include oxygen in LOCA simulation chambers?
o What is an acceptable acceleration method for radiation doses and rates in pre-aging and accident simulations?
o Under what circumstances is the Arrhenius methodology for accelerated thermal aging valid?
o Are mechanical stresses significant in aging of electrical equipment
_ (cables, seals)?
o Are the procedures of IEEE standards for qualifying specific types of electric equipment adequate?
o Can electrical cabinets cope with the envircaments produced during fires?-
o Will adverse' fire environments (e.g., suppression agents, smoke, corrosive
- gases, humidity) damage equipment such that sufficient equipment does not remain free of fire damage?
o Do tne spatial separation options of Appendix R and associated exemption requests truly ensure the operability of sufficient safety systems during
.. fire?
o Should barriers, penetration seals, and other barrier elements be tested at positive pressures?
o Should cable tests assess cable functionality, as well as burnability, l requirements?
o Should ventilation systems be qualified to handle smoke and other i
combustible products without jeopardizing cooling functions?
4 4
m-( avsoonse (Environmental Qualification of Electrical Equipment)
The first seven issues listed in Attachment A to "Qualification Methodologies Have Shortcomings" are major elements of the scope of work taken from the SNL ongoing Electrical Equipment Qualification Research Program Plan being conducted for the NRC. They represent areas of equipment qualification where an additional understanding and verification of the test procedures is believed to be needed.
The SNL equipment qualification research has made significant contributions to our understanding of these issues and has identified the need for improvements to the qualification procedures. Those improvements which have been thoroughly researched are being implemented by NRC by revisions to regulatory guides and I. used in the licensing review of equipment qualification. However, it has not been demonstrated in the SNL research tests that nuclear plant safety equipment, properly qualified to existing IEEE standards and NRC regulatory requirements, would not perform its safety function.
The qualification methodologies as represented by the national consensus standards must be properly implemented. Tha NRR review of qualification test programs and IE reviews of test performance and test quality assurance and control programs are being carried out to assure that this is accomplishad.
l 5
NRC Response _(Fire Protection)
The NRC does not require tests to qualify electrical cabinets nee +d'for safe shutdown to the environments in the same sense as safety relate < aquipment qualification is done for LOCA environments. We do, however, perform-reviews to assure that safe shutdown can be achieved when electrical cabinets or other equipment might be exposed to fire environments. The NRC research program will develop test data to assess limitations of equipment and effects on equipment operability and responses under fire related environments to verify these evaluations. The fire related environments to be considered will include suppression agents, smoke, corrosive gases, and humidity. In addition, data to better characterize fire sources and the resulting environments in fire areas will be obtained to give insights into the safety margin provided by spatial separation. Functionality of cables will also be considered.
With regard to the question of whether fire barriers, penetration seals, and l other basic elements should be tested at positive pressures, Sandia has per-
! formed an extensive research program to evaluate this issue for penetration l seals. These tests showed that if the penetration seals contain highly -
l combustible material (in this case urethane foam), or permit communication through cracks or other openings, positive pressure during the test makes a difference in the performance of the seal. The staff requires that approved penetration seals be constructed of non-combustible materials and that they do not permit communication through the seal. Therefore, positive test pressures are not required. Technical specifications require licensees to r regularty inspect fire penetrations for cracks which could degrade their l performance. When seals are disturbed or removed for other reasons other compensating measures are instituted. If fire doors or fire barriers are subjected to positive pressure during a fire, some smoke and fire will leak to the unexposed side. The research on responses of equipment to fire related environments wil1 lead to additional insights on the importance l of such effects.
We agree with Sandia that if ventilation systems are to be designed and proposed for use to handle smoke and other combustible products so as to not jeopardize vital cooling functions, then these must be shown to be capable of performing this intended function. It is the staff's experience, however, that ventilation systems are not usually used in this way. In most cases, the ventilation system is isolated and the staff requires alternative means (e.g., portabla blowers) to be available for this function.
I i
6
- 2. Design' Bases (Acceotance Criteria) Have Shortcominos SNL Concern l Based on the following examples, we are led to believe that there are l
some shortcomings in design bases:
A. Qualification test procedures a'nd/or requirements do not always reflect application conditions, such as (a) Acceptance criteria for terminal blocks under certain use conditions as identified in SAND 83-1965C. ~
(b) Acceptance criteria for coaxial and triaxial cables are not documented as related to use conditions as identified in Vendor Inspection Program Occket 99900277.
(c) Interface conditions during testing do not always reflect use interface conditions. An example is venting of the internals of limit switches during qualification testing (as identified by FRC evaluation of 79-013 submittals
- p. 75 of TR-C5257-532).
B. Type testing reporting does not insure full reporting of all test results. An example was identified in the Vencer Inspection Program t
participation associated with Docket 99900277. J C. Fire protection guidelines do not specifically require equipment qualification for the environments expected during a fire (e.g.,
HC1, humidity, sprays).
D. Fire protection guidelines permit the use of spatial separation as a fire protection measure, despite evidence that separation alone may be inadequate to ensure fire safety.
i l
4., ,
7 NRC Response-(Design Bases Have Shortcomings) ,,
NRR is aware of the concerns expressed by Sandia and they are being ,
addressed in NTOL and OR equipment qualification reviews (ref: SER 1 for Byron /Braidwood, Callaway/ Wolf Creek). "
Applicants and Licensees in their review must ascertain that the test acceptance criteria is applicable to the end use of the equipment.
Specifically:
Item 2A(a) - Insulation resistance and leakage current values are reviewed in the acceptance of terminal block qualification.
Item 2A(b) - If the acceptance criteria are not documented nor reviewed then the equipment (coaxial and triaxial cables) is not properly qualified. The final EQ Rule, NUREG 0588, and R.G. 1.89 (which generally endorse IEEE 323-74) require that the safety-related equipment must perform its safety function. Other regulatory guides covering qualification of specific equipment, for example cables, are daughter guides and are by themselves not adequate to demonstrate qualification. In all cases, the requirements of the Final Rule must be met, t
Item 2A(c) - In all licensing reviews the equipment qualification files
) are audited to assure that the equipment is tested in a manner repre- ;'
sentative of its installed configuration. IE/ Regional-inspection activities further ensure consistency between testing and installation configurations. ,
Item 29 - The staff is aware of concerns about the adequacy of require-ments and practices for reporting qualification test failures and is t
currently considering actions which should be taken to address this
- issue.
Items 2C a'nd 20 have been addressed in our response to Sandia's first concern ("Qualification Methodologies Have Shortcomings") in the dis-cussion relating to fire protection.
r l
l
8
- 3. Some Inadecuate Ecuiement is in Plants SNL Concern Varied evidence indicates inadequate equipment is in plants. This evidence includes the FRC reviews of utility submittals, I&E Information Notices, as well as the Sandia testing experiences. Further evidence has been identified via the NRC Region IV Equipment Qualification Section inspections of industry qualification activities (e.g. .. Docket No. 99900277/83-01 re-garding Rockbestos cables). Other examples are: terminal blocks which can be inadequate in certain applications; 0.G. O'Brien connectors; recent tests of EPR cable performance in simultaneous environment; behavior of ,
RTDs and pressure switches. Therefore, since all such equipment has not been removed from plants, they exist and are "inadequate." In all instances, NRC is aware of these test results and action has been taken.
NRC Resconse The FRC reviews identified the equipment in operating reactors which have not been demonstrated fully qualified. The percentage of equipment so identified is not a measure of equipment inadequacy. All equipment which has not been shown to be qualified must either be qualified, be replaced by qualifed equipment or be justified for continued operation.
The JCOs have addressed the requirements for plant safety.
A number of I&E notifications have identified specific cancerns with qualification of some components. The licensee is required to review the notification for applicability and take appropriate action.
NRC is aware of the test failures experienced by the Rockbestos cables cited and an information notice is being prepared. The safety implica-tions have been addressed and it was concluded that an immediate safety problem does not exist.
NRR is aware of the Sandia concerns regarding the items listed as examples:
- Insulation resisti.nce and leakage current values are reviewed in the acceptance of terainal block qualification. IE Information Notice 82-03 which originally notified licensees of this issue will be updated in the near future to further clarify the results of research.
- 0.G. O'Brien penetrations, including the connectors referred to by Sandia, have been retested at Wyle Labs by Duke Power Company. These penetrations were certified by Duke Power to have passed the tests.
- EPR Cables tested at Sandia, as a part of the research program, used a saturated steam LOCA profile. The staff has requested additional testing.
- Pressure Switches - Those models that failed were not. vendor qualified and are not to be used for safety related functions in applications wnere they would experience high pressure and steam / spray environments. An IE Infor-mation Notics has been issued giving the results of the'Sandia tests stating that the models that failed should be replacad with qualified models.
- RTOs - Only one plant has the affected model inside containment. The licensee has provided a JCO. The staff will discuss this item with the licensee in an upcoming meeting.
,y 9
Pressure SNL Concern We perceive that the decision process in the. regulatory environment is such that new observations and interpretations cannot be accommodated without simultaneously involving a commitment to initiate a change ("ratchet"). We perceive further that this condition tends to foreclose an adequate under-standing of technical issues.
NRC Resoonse The Commission's 1983 Policy and Planning Guidance to tha-staff includes the following guidance on the relation between research and NRC regulations:
l "The research rescurces identified in NRC's budget should be allocated to supp .t a balanced program between supportive research for regulatory needs, research to reinforce or revise the current regulatory base, and concept 6al research for. improved reactor safety. The staff should be alert to research which shows that we ought to change our regulations.
NRC regulations should be changed when research shows them to be either too stringent or not stringent enough."
Although new observations and interpretations evolving from research may result in plant changes or "rachets" in acceptance criteria, additional efforts to understand technical issues are not necessarily foreclosed once licensing decision.s are made. There are many examples where related research has continued. after significant regulatory decisions were made. Specific examples ir.clude the continued research on fire protection after Appendix R l was issueJ, the continued research on equipment qualification after the EQ rule was issued, and the continued rese ch on loss of coolant accident analysis after Appendix X was issued. 1. may be differing opinions re-garding wlat constitutes an adequate unden nding of a technical issue, but further study is never foreclosed whene, r a significant safety issue is identiff.id.
The staff 9ill try to assure that Sandia research perspectives and NRC licensing perspectivas are mutually understood.
l l
l
! 4
{
l
y
~ ~
10
. , Pressure, not to Imoact Previous / Current Licensing Oecisions SNL Co,cern .
Examole o The board notification procedure has a 48-hour notification requirement, upon identification of problems, and a tendency to inhibit in-depth and rational analysis of the information and its implication. This has occurred on terminal block tests, 0.G. O'Brien connector tests and EPR cable test results.
NRC Aesconse Sandia's interest in engaging in an in-depth analysis of identified problems is reasonable. 'The near-term process followed by the staff in evaluation of new information from research programs involves a rational and sufficiently complete technical analysis to make appropriate decisions regarding the immedians actions to be taken, whether notification of hearing boards or issuance of IE Information Notices. The necessity of rapid notification of hearing boards precluces lengthy deliberations in the initial phase. However, this does not preclude or inhibit development of information necessary to resolve the issue. In fact, most initial board notifications are followed with detailed analyses to resolve the issue and in some cases this process takes many moaths.
In addition, in each of the cases mentioned by Sandia, additional research or testing has been pursued. Sandia has continued to test and evaluate terminal blocks, and will re-test EPR cables in tne near future, and Duke Power has performed follow-on tests of 0.G. 0'Brien connectors.
The staff believes that what underlies Sandia's. concern is the rapidity with which initial decisions must be made and a lack of continuity of Sandia's involvement in the complete resolution process, including decisions about follow-on activities. On this basis, the staff will make special efforts to keep Sandia informed and to call on them for participation as appropriate in future deliberations regarding new information tney have provided.
4
4 4 . 11 Pressure to Resolve the Problem Today SNL Concern Pressure related to the resolutica of issues related to severe accidents has been quite high. The NRC has taken steps which have resolved the i: sue.
NRC Resoonse This issue apparently rdsulted from NRC requests that Sandia perform r.aoid turn-around reviews of analyses of severe accidents performed by industry (IDCOR).
The schedules for these reviews have been relaxed. It has been agreed that this' issue is not related to the subject of equipment qualification.
I 1
l l
I i
l l
12 Pressure to Control External Interactions l
SNL Concerns Sandia is subjected to NRC control of the timing and distribution of reports in the severe accident area which fall under international cooperative funding agreements. It is Sandia's opinion that this inhibits the free exchange of information which in turn reduces the opportunity for peer review and the related checks on the quality of the results.
Sandia and other laboratories.have work statements from NRO containing directions to have all travel plans pre-approved by the NRC sponsor. Sandia and other laboratories feel that this control should only be present if it is used and that use of this control is detrimental to the efficient conduct of its worn.
NRC Response (External Interactions)
This issue is to be the subject of a further discussion within the NRC.
NPC Resoonse (Travel)
The decision to have travel pre-approved by the NRC program manager is the prerogative of the cognizant NRC organization. For most of the research programs related to equipment qualification research at Sandia, such pre-approval has not been required; when the practice has been implemented, it has not resulted in disapproval of any travel requests. We believe that' NRC's limited practice of travel pre-approval has not had any adverse impact on the quality of research carried out by Sandia or on plant safety as re-lated to equipment qualification. Pre-approval of travel by NRC is not nieant to inhibit research or information exchange. We will discuss Sandia's concerns in this area further to try to arrive at a mutually satisfactory resolution.
,7 00&CIE0 '
inHRC .
SEABROOK SERVICE LIST - ONSITE LICENSF 'O BOARD
'Sheldon J. Wolfe, Chairman 155 Washington Road ' Office of Genera' Aunsele *as J1.15 P5!59 U.S. NRC Rye, New Hampshire 03870 U.S. NRC Washington, D.C. 20555 .
Washington, D.C. 20555 Richard E. Sullivan, Mayor gyginightNak, g
'Dr.Jercy Harbour City Hall R. Scott Hill-Whilton ggAgcq Newburyport,MA 01950 Lagoulis, Clarck, Hill-Whilton
- By Hand Delivery U.S. NRC Washington, D.C. 20555 & McGuire Alfred V. Sargent, Chairman 79 State Street
- Dr. Emmeth A. Luebke~ Board of Selectmen Newburyport, MA 01950 5500 Friendship Blvd.- Town of Salisbury, MA 01950 Apartment 1923N George Dana Bisbee, Esq.
Chevy Chase, MD 20815 Senator Gordon J. Humphrey Geoffrey M. Huntington, Esq.
U.S. Senate Office of the Attorney Ger.eral Atomic Safety and Licensing Washington, D.C. 20510 State House Annex Board Panel (Atta. Tom Burack) Concord,NH 03301 U.S. NRC Washington, D.C. 20555 Selectmen of Northampton Allen Lampert Northampton, New Hamp- Civil Defense Director
. Atomic Safety and Licensing shire 03826 Town of Brentowood Appeal Board Panet Exeter, NH 03833 U.S. NRC Senator Gordon J. Humphrey Washington, D.C. 20555 1 E:gle Square, Ste 507 Richard A. Hampe, Esq.
Concord,NH 03301 Hampe and McNicholas
Docketing and Service 35 Pleasant Street U.S. NRC Michael Santosuosso, Concord,NH 03301 Washington, D.C. 20555 Chairman Board of Selectmen Gary W. Holmes, Esq.
hfrs. Anne E. Goodman Jewell Street, RFD # 2 Holmes & Ellis Board of Selectmen South Hampton, NH 03842 47 Winnacurnent Road 13-15 New Market Road Hampton, NH 03842 Durham,NH 03842 Juditb H. Mizner, Esq.
William Armstrong Silverglate, Gertner, et al.
William S. Lord, Selectman 88 Broad Street Civil Defense Director Town Hall-- Friend Street Boston, MA 02110 10 Front Street Ama.sbury, MA 01913 Exeter,NH 03833 Rep. Roberta C. Pevear Jane Doughty Drinkwater Road CaMn A.Canney SAPL Hampton, Falls, NH 03844 City Manager 5 Market Street City Hall Portsmouth, NH 03801 Phillip Ahrens, Esq. 126 Daniel Street Assistant Attorney General Portsmouth, NH 03801 Carol S. Sneider, Esquire State House, Station # 6 ,
Augus'a, ME 04333 Matthew T. Brock, Esq.
Assistant Attorney General 1 Asnburton Place,19th Floor Shaines & McEachern Boston, MA 02108 "Thomaa G. Dignan, Esq. P.O. Box 360 R.K. Gad H, Esq. Maplewood Ave.
Ropes & Gray Portsmouth, NH 03801 Stanley W. Knowles Board of Selectmen 225 Franklin Street P.O. Box 710 Boston, MA 02110 Sandra Gavutis North Hampton, NH 03826 RFD 1 Box 115
- Robert A. Backus, Esq. East Kensington, NH 03827 J.P. Nadeau Backus, Meyer & Solomon 111 Lowell Street Charles P. Grahaa, Esq.
Town of Rye Manchester, NH 03105 McKay, Murphy and Graham 100 Main Street "Gregory A. Berry, Esq. Amesbury, MA 01913
s --
- st. .
.L. NUREG/CR-4728 SAND 86- 1938
] RV Printed February 1988 l
Equipment Qualification Research Test of a High-Range Radiation Monitor E. H. Richards, M. J. Jacobus, P. M. Drozda, J. A. Lewin Prepared by Sanca Natonal Lat,oratones Albuquerove. New Menco 87185 and Levermore. Cahforrva 94550 icr the Ursted States Department of Energy under Contract DE-AC04 760P00789 Prepared for U. S. NUCLEAR REGULATORY COMMISSION
o 6 .
, SAND 86-1938
- RV EQUIPMENT QUALIFICATION RESEARCH TEST OF A HIGH-RANGE RADIATION MONITOR E. H. Richards H. J. Jacobus
- 1. M. Drozda J. A. Lewin l
1 1
February 1988 Sandia National Laboratories Albuquerque, NM 87185 operated by Sandia Corporation
. for the l U. S. Department of Energy l
l Prepared for Electrical and Mechanical Engineering Branch Division of Engineering Office of Nuclear Regulatory Research U. S. Nuclear Regulatory Commission Washington, DC 20555 l Under Memorandum of Understanding DOE-40-550-75
' NRC FIN No. A-1051 l
l l
l
1 .
A .
Abstract A high-range radiation detector was tested in a simultaneous steam and radiation environment simulating a postulated loss-of-coolant accident (LOCA) to assess possible synergistic effects that may be important to its performancewas in an accident. The detector, manufactured by General Atomic, simultaneously subjected to a simulated accident environment s including '71*C (340*F) steam at 410 kPa gage (60 psig) and 4 Mrad /hr gamma radiation whila its performance was monitored.
Test results showed that the detector successfully operated at the high dose rate and temperature, without evidence of synergisms.. However, at reduced radiation levels in a saturated steam environment, the detector signal at the readout module deteriorated in accuracy or ceased altogether. The cause of these anomalies is attributed to leakage currents and/or possible galvanic action in the coaxial connections and/or cables.
)
-lii/iv-
s A,
Table of Contents Executive Summary.......................................' 1 1
- 1. Introduction........................................ 3 1.1 Test Objective................................ 3 1.2 Background.................................... 3 1.3 Scope......................................... .
4
- 2. Test Specimen....................................... 5 2.1 Specimen Description.......................... 5 2.2 Basis for Selection........................... 6
- 3. Test Description.................................... 7 3.1 Test Facility................................. 7 3.1.1 HIACA................................... 7 3.1.2 Steam System.................,.......... 7 3.1.3 Test Chamber............................ 7 3.1.4 Instrumentation......................... 11 3.2 Test Environment.............................. 12 3.2.1 Aging................................... 12 3.2.2 Accident Simulation..... .............. 13 3.2.3 Dry run tests........................... 15 3.3 Monitoring of Test Specimen......... ......... 17 3.3.1 Pretest Characterization.......... ..... 17 3.3.2 Functional Performance During Test...... 17 3.3.2.1 Output Signal..................... 17 3.3.2.2 Insulation Resistance............. 18 3.3.2.3 Effect of Cable Length............ 18 3.3.3 Post-test Characterization.............. 18 3.3.4 Acceptance Criteria..................... 18 4.0 Test Results........................................ 20 5.0 Analysis of Test Results........................... 38 5.1 Data from Bench Tests......................... 38 5.2 Analysis of Data from Bench Test.............. 40 l
6.0 Conclusions and Recommendations..................... 46 6.1 Detector Performance........................... 46 6.2 Recommendations for Radiation Monitor Users.... 46 i
6.3 Recommendations for Future Tests............... 47 l
! 7.0 References......................................... 48 APPENDIX A Discussion of Split in Cable Jacket......... 49 1
1 l -V-l l
'\
l
N >
. s
- List of Figures Figure 1. RD-23 Detector................................. 5 Figure 2. Radiation Film Detectors used to Calibrate HIACA............................. 8 Figure 3. \- Test Chamber Installed in HIACA Facility....... 9 Figure 4. Mounting Fixture for RD-23 Detector............ 10 Figure 5. Thermocouple Locations......................... 12 Figure 6. LOCA Environment Profile, Based on IEEE323-1974............................. 13 Figure 7. Dose R3tes Calculated from IEEh323-1974, Table A1...................... 14 Figure 8. Dose Rate Comparison: HIACA vs. IEEE323-1974.. 16 Figure 9. Dummy Test Specimen............................ 16 Figure 10. Chamber Pressure and Temperature and Detector Output............................ 21 Figure 11. Detector Output et Beginning of Test........... 31 Figure 12. Strip Chart Record of Detector Output.......... 34 Figure 13. Block Diagram of Operational Amplifier Circuit.......................... 41
-vi-
. \
List'of Tables Table'1. Dose Rates for HIACA Co-60 Groups................ 11 Table 2. IEEE 323-1974 Radiation Profile (PWR)............ 14 Table 3. HIACA Dose Rate Profile Used for Test............ 15 Table 4. Calibration Data................................. 32 Table-5. Cable Insulation Resistance During Test.......... 36 Table 6. Experimentally Determined Outputs of the Readout Module for Various Insulation' Resistances and Input Currents............................ 39 l
\ e l
l .
I i
l l
l l
-vil-1
)
. 4 j 1
l i
Acknowledgments 1 l
l The authors would like to thank Lloyd Bonzon for .his assistance in planning this test program; Tim Gilmore, Ron Garcia, Bob Padilla, and Craig Ginn for assisting in the execution of the i test; Frank Wyant for supplying the radiation dose rate l calibration used in the test; and Jeremy Sprung for providing
~
i information on the timing of the radiation profile. l t
t t
-viii-l 1
Executive Summary A high-range radiation detector was tested in a simultaneous steam and radiation environment simulating a postulated loss-of-coolant accident (LOCA) to assess possible synergistic effects that me.y be important to its performance in an accident. The 6etector, a General Atomic model RD-23, was simultaneously subjected to a simulated reactor accident environment including 171*C (340'F)' steam at 410 kPa gage (60 psig) and 4 Mrad /hr gamma radiation while its performance was monitored.
The-original concern was that the heating induced by the high radiation levels would cause the ion chamber to overheat when applied simultaneously with the LOCA steam environment.
Such effects would go undetected during typical qualification tests, which do not employ simultaneous exposures. We found that the detector operated successfully without any evidence of synergisms at-high dose rates while exposed to the most severe environmental conditions. However, at reduced radiation levels l in a saturated steam environment, the detector signal at the readout module deteriorated in accuiacy or ceased altogether.
We at first believed that leakage currents in the cables
- and/or connections caused the anomalies, so we conducted additional electrical tests after completion of the LOCA simulation to assess the response of the readout module to variations in the insulation resistance of the interconnections. These tests demonstrated that the insulation resistance is critical at low radiation dose rates, but becomes less significant as the dose rate is increased.
The tests also established that the loss of signal at the readout module, manifested by the readings going off-scale at the low end of the ranga, was not caused by decreased insulation resistance. We found that the effects of decreased l
insulation resistance, both as determin'ed by theoretical l
considerations as wall as by test data, would have caused our particular detector to read abnormally high. The only explanation we could postulate for the detector's behavior during the LOCA simulation was galvanic action in the coaxial
- connections and/or cables, although we were unable to confirm l this theory.
! The results indicate that "low" dose rates cannot be accurately meesured under the conditions used in this test.
However, higher dose rates can be monitored accurately even l though the green "operate" light on the readout module may go off prior to the presence of high dose rate radiation. (If the light does go off, resetting the light illuminates it again.)
We concluded that i a) At low radiation dose rates, insulation resistance l effects of the interconnections may cause detector accuracy to l
, s degrade beyond tho' factor of two allowed by Regulatory Guide 1.97.- The magnitude and direction of the observed error depend on the interconnection insulation resistance, the readout module input amplifier's input offset voltage, and the detector signal level.
b) A second effect, perhaps galvanic action, caused the J detector signal to b.e lost at low levels of detector output.
We were unable to hypothesize any other mechanism which would explain the observed loss of signal during the accident test.
We also ~ noted that for our test, whatever this effect was, it appeared to dominate at low detector currents. After the test, we discovered a small split in the cable jacket which may have provided a convenient path for moisture intrusion, facilitating a galvanic reaction (although some moisure intrusion through the cable jacket would be expected even without the split).
We recommend that users of the RD-23 detector assess the results of this test program as it relates to their detector installation. We also recommend that licensees with other types of high-range radiation detectors review their
_ qualification test data to check for similar concerns.
i 2
6 .
- 1. Introduction 1.1 Test Objective The purpose of this testing program was to investigate synergistic effects of Loss-of-Coolant Accident (LOCA) steam conditions and high level radiation environments on a high-range radiation detector. The test was designed to assess the importance of qualifying high-range radiation detectors using simultaneous LOCA and radiation exposures as opposed to qualifying the detectors using sequential exposures. We were concerned that the high radiation dose rates early in an accident might cause excessive radiation heating when coupled w i *. '.i the LOCA steam conditions, causing damage to the detector. The test was not run with the intent of duplicating the vendor qualification test, but rather to subject the detector to a more representative design basis accident environment than that generally used during qualification.
Specifically, the radiation detector was exposed simultaneously to a LOCA steam environment and a radiation profile i
representative of a design basis accident. By comparison, l typical qualification tests include irradiation of a piece of l eqpipment to a given dose (usually at a much lower dose rate
<than that predicted for peak accident conditions) followed by exposure to a LOCA steam environment (i.e., sequential 1 testing).
1.2 Background
The Iguclear Regulatory Commission (NRC), via NUREG-0737 and Regulatory Guide (RG) 1.97, sets forth requirements for monitoring radiation levels after an accident at a nuclear power generating station. One of the requirements specifies that plants must have at least two high-range radiation detectors installgd in containment to monitor the high levels (greater than 10 rad /hr) of radiation associated with accidents. It is further required that these detectors or l monitors be Class 1E qualified to withstand the LOCA and radiation environments. Based on our knowledge of plant
- installations from equipment qualification inspections, three manufacturers' detectors have been tested and installed in l None of these detectors were qualified using l plants.
simultaneous exposures of radiation and the LOCA environment.
Questions have been raised regarding possible synergistic effects of the LOCA and gigh radiation environments. An NRC memorandum dated 11/27/84 requested testing of at least one high-range radiation detector using simultaneous exposures to LOCA steam and high radiation dose rates. The request was based on an NRC staff ' Evaluation of Presently Installed High-Range Containment Monitors in Nuclear Power Reactors,"
l (included with the memorandum). The evaluation's primary concern stemmed from an article in Health Physics.4 I In the article, the authors calculated a significant temperature rise 3
s !
for gas-filled radiation detectors (such as gas ionization chambers) when exposed to high levels of radiation. The staff was concerned that this temperature rise, if it actually occurs, combined with a LOCA steam exposure temperature could cause thermal damage or failure of the detector. Such effects would go undetected during qualification tests which do not employ simultaneous exposures.
Other types of failures not due to synergistic effects were also possible, such as failures due to moisture intrusion or other consequences of the steam environment alone. In addition, although we were primarily interested in the detector's response, the monitoring system cannot function unless the cable and connections also survive the accident conditions.
As it turned out, while the system operated satisfactorily at the high dose rates, it experienced some anomalous behavior at lower dose rates. We believe this anomalous behavior was not a result of any synergistic effects, but was a result of the response of the cable and/or connections to the LOCA steam
, environment.
E 1. 3 Scope This report covers testing of an unaged General Atomic (GA) RD-13 radiation detector which was exposed to accident radiation doses and dose rates simultaneously with the LOCA steam environment. Again, the intent of the test was not to qualify the detector by putting it through the standard set of qualification tests used in the industry, but to understand how it responds to a more realistic simulation of a design-basis accident environment. Sprays were not simulated in the test.
5 originally projected additional tests of The test plan the same detector, as well as other detectors, depending on the results obtained from the test documented here. However, while the detector signal reaching the readout module did experience some anomalous behavior, the anomalies do not appear to be the result of a synergistic effect on the detector itself. Based on the results of this test and discussions with the NRC, we do not see a need for further testing of the RD-23. We believe that the test anomalies were due to the cable and/or connections and that it would be more beneficial to direct future studies toward examining the behavior of coaxial cables and connections. A radiation monitor might be needed in such a test to produce and monitor the low level signals associated with the overall system.
4
1 .
- 2. Test Specimen 2.1 Specimen Description The item tested was General Atomic's model RD-23 radiation detector, a gas ionization chamber, which is part of GA's model RS-23 high-range radiation monitoring system. The RS-23 system can be purchased,as either an analog or digital system. We purchased one complete RS-23 analog system (RS-23A) with one RD-23 gamma detector.
A diagram (taken from GA promotional literature 6 ) of the RD-23 radiation detector is shown in Figure 1. It consists of a gas ionization chamber, stainless steel housing, _
integral cables, connections, _._q _ m om, and pull box with cover. The ,,,,,,
ionization chamber has two / ,io w irin caimiin integral mineral-oxide 85 * /
insulated cables, one for ,
//!'a%'II'Eilitovsmo signal, the other for high +
voltage. LOCA-qualified / "" **
connections are included for /
connecting the in-containment 2:a rM*
cables to the integral --g' f siii m nousm cables. The ionization f'iennc 8"5ai,te,u l
chamber is electrically isolated from the detector ,,,, 7",o',,n,';t st o ns' g,",tj,1?5 housing to minimize noise and w ber c is mechanically isolated from 7} A ,g oj y p,,cozia the detector housing to p j'"5""'"#"""
~
protect it from thermal and
--(
l mechanical shock. A more /$
detailed drawing can be obtained from References 7 jMjw'-
i and 8; a proprietary can otucion information agreement l prevents us from including it here. Figure 1. RD-23 Detector.
During operation, gamma radiation inside containment is sensed by the RD-23 detector. The ionization chamber produces a direct current proportional to the gamma flux. The current travels over coaxial cable, is sensed by a log picoammeter that is temperature-compensated to minimize drift, and is displayed in roentgens per hour on the front panel meter. To maintain a signal at all times, a coating of U-234 is applied to the inside of the ion chamber housing. This generates a sustained signal within the detector corresponding to approximately 1 rad /hr. A failure alarm will occur if the signal from the detector falls (sufficiently) below this value. This feature is intended to assure knowledge of the monitor's integrity at all times. Detailed information concerning the RD-23's 5
4
, s specifications, installation, and operation can be obtained in References 6, 7, and S.
i The GA requirements for interconnecting the detector ;
spgcify 10 that a cableSimilar ohm be used.
with an insulation resistance of at least requirements are specified for the i electrical penetration, with even more stringent requirements if the detector is expected to operate under station fault conditions. As can be shown using the equation develoged in ohm Section 5.2, an gnsulation resistance of 5x10 I (accougting for 10 ohm for the cable and interconnections and 10 ohm for the penetration, connected in parallel) will maintain the overall igstrument accuracy within the
- factor of two allowed by RG 1.97 if a reasonable worst case amplifier offset voltage is assumed and if other effects are neglected.
The error due to any given insulation resistance will be shown to be a function of the combined (parallel) insulation resistances of all interconnecting devices, the detector signal level, and the input offset voltage of the readout module input operational amplifier.
2.2 Basis for Selection We chose to test this particular radiation detector for
,i?several reasons. One is that, based on equipment qualificatiog
' inspections and the EPRI Equipment Qualification Data Bank, this model is currently installed in many reactor containmengs. GA has LOCA qualified the RD-23 to IEEE 323-1974, o and claims that the monitor meets the requirements of RG 1.97. Another reason for testing the RD-23 is that the dimensions of this particular detector are such tnat it will fit into the size test chamber required to achieve the high dose rates in Sandia's test facility.
6
- 3. Test Description 3.1 Test Facility 3.1.1 HIACA The accident simulation tests took place in Sandia's High Intensity Adjustable Cobalt Array (HIACA). . A complete description of the HIACA is available in Reference 11; it will be described only briefly here.
The HIACA consists of 32, 24-in. Cobalt-60 source pencils. The pencils are arranged in a circle or circles, giving a cylindrical test volume that supplies a uniform dose to test specimens, thus eliminating the need to rotate samples to obtain uniform exposure. The array is adjustable: the pencils can be moved to accommodate larger test specimens at lower dose rates or smaller specimens at higher dose rates.
The ' f acility design also allows the source to be configured either one or two pencils high. For this test, the smallest radius was used with the 24-in pencil configuration to obtain dose rates up to about 4 Mrad /hr. (Throughout this report, all dose rates are reported in rad /hr (air) or Mrad /hr (air). The 7
output actually reads in R/hr, but the two are
}[equivalent.)
detector Before installing the detector in the HIACA, we calibrated the radiation levels using film detectors (Figure 2). Table 1 shows the average radiation dose rates for the different groups of rods in the configuration used in the test.
3.1.2 Steam System Sandia has the capability of being able to apply the LOCA steam exposure and the accident radiation dose simultaneously.
The steam system is designed to accommodate severe accident testing; the test of the radiation detector did not require such extreme conditions, and the facility had no trouble meeting the desired conditions. The boiler shut down once during the test, but the associated temporary drop in pressure should not affect the test results.
3.1.3 Test Chamber Figure 3 shows the test chamber installed in the test facility. It was constructed using a stainless steel pipe with a standard pipe end cap forming the bottom. The RD-23 detector was installed in the test chamber via a mounting fixture attached to the chamber head (Figure 4). The simplest way to mount the detector would have been to mount it upside down in the chamber so that the cable (which exits the detector through the botton) would not have to be doubled back to penetrate the test vessel. The installation instructions for the detector do not specify that the detector be mounted in a particular 7
l 1 -
l l -
l w,. ,'
, k' fj ~
o s t ,. , . .;, , i' l ' ('f ', ;.? V j. ' -
- i4 l
j.
e
}?3 .:b
= r =b
. * ;,c
.- n,">g.y,v ..'
a-
.ri : s
. . I'j l
%'. c dp M 3 r+a :a. .+
.; . C.w , 4 ,
Yff'.t
'.MQ I, y;,
' j!}yf".l'f
$ , @-s
[l (G;/J 4 i:)$ z, sq. q; ; -
g:" M,'
- A $ h ., ,.
,q. . .
j
.N .
pt r ;p . : .,> , 1 t ,, .
1 , r. ;
- 0. !! .!.
i i
- I i e i J i
1
[ . d WM?$ c (h
.?CLS:@ -
+( )s
- Nl .
- s.,jGk i
'ly -
as ! ;s l 4 'l
?
c.2; , -
tA
',b Q..
t.W
.p; .
. ,r d" k .
, f'
[:
f , [h t , , .
~ M ;;g%. . - __ , _ -
. ::c c t '
( 2:O-9' l
.,,c
.4.
< e.M s
h' ...
+:
v: v.
. 4
$lfif~- s ;.<,
t s, a,1.Y ?
- _:lA -
> . ,t:
j l l s '- .
.;f c0
' g, .
Y ;*y -f 'g l f'
- i ,? s i C' ' '__ . 4 s X.G;
[4';,;f ' Y.M- 2 X',) .,
.%, ?s:
.v.; ,' ;
,b,',,,,. ..~.,>
.w s+c .sw .' . .
.a
.w ,
l<f j L % j;'lp ,,
f(ljy; ' X ..,;q tp.9; yj ;&, J i,,. % I4;.jl$
r? . , yL 8)dlll,w;wa
- .F
- l } "
- /r -4
- A'
~ DafWid emp C" yh A ' '
hl j h y'(Q+ k",,[ _ _ , _ , . -
Figure 2. Radiation Film Detectors used to Calibrate HIACA.
8 i i
l
~
g
. .w.
, c.i.9'J.} j47. '
)
, 3 ,
y 2- -
_ yn , + . . ,
}
~ ' '
- I/I' . .
> yq , _. kj II - *
,[
l /
_, .' e4
'A' .e, g.- ,
9 :. Q
+ -
s -<y., < -
.g
' ~
.['
3
~
7 ' Y. , j -f := r . ,4
/ W6.' s :,,J , .
- ,,, f- .
p-l
,; r. ,
y ,
i
, ~y
~ > ~ . , f .k l! nd s ,[ ' - ' ,& -' 5 y4.ln, .~ {s - ,,. . ,& -
3 ~ .
_7
( ".-
- [ l ' 's?. _
l 1
dl0 .t - w,t.---w
.i N6; .t: ', q
,/,5 , )
.y s
i ., .. g " ,7 fa .- ',
' '[ .2 's i /
h
,)
p #, #s -
i n .- _
. w*9. .:m, gd.s /
. .- s .; . . , , . -.
,, '- .. ~ r ) .
)
n : .;- ^7 /'L .
[-
- 7) %.,
tqm . e' gj+
R
- y. . -Eg .- , .b ; _, ~
. . , ,p .e.-(
,, . %%g -- - . ,,.
. ,f ....,
,A, .
,- u .,z y y Qt -__ . ;
a
,j....
~
ee
.. . , ' ,.. !Tl 'j ,
1 ,; v a ..
, o3- , ; *-
I
,f ? w 'j ny R l- 6-
- l . s . 7,a.g . , ,,-ow,.wgjf,
. . -., ..u. -
a, e . ,
l ( -m <
.n' ov.s
" "o l y12l , -) ' %:
j ., ; -,
l j I 'I f. 9 F j f f 4' t [ h . [ f . i p',
g ^ " " v .o e ,. l ,
? , ;[ >
l
.:} ;r;?
y
-i , s.,
3
, j-
<; =
~.)oG@
w31
,;j _ .,M [ U., t f' t j-; ,
g>, m. e. j ...
cn+
U 6 2i f
.g , b .:
J, ;y
- b,Y Y ;[?
- y
~*
fi ,:2:m 4 ; p s ,,y:;
=
.- - ' s u .a . , . -
, 4
.{
- y:' h q'
.,, , .. J7
- v-.~
~
, ; _ ->. ':?37:.m . ,
- rpm. , Gg.h ; -
Q;.
- m. :: s
'~~~.,.=
c ;. :, __c - m 7
. . ; Q.. k .tu; -t
= p[, 5 .; .
' - " ' C_ m j
.,c a
,i
~ .'e' y.
W;- e .;
Yp; c '- 7
{
. . . ,,: 4. .
- Wy~' .~.;, L
- c.,, ! ' '
_.,.. Wic ;
- ?%
3 x . .llL 1 A%.a a . .
I Figure 3. Test Chanber Installed in HIACA Facility.
1 9
l 1
S 6 9
. . = . -
3
^
gy _ _. . y_
/W . s*- .@ -
%% ,Y , ..-~-. a. __
4.,a.pw.
'? '
- p. -
m '
n
=~
b, s cf' 4..
o j 1, .
u l
y.:
i .
} ' '
i 1 .
.; b A
i
- l,%. ;* . . . N' s '
N
( ,
,..,- M I
"hk4.wwa =~ 4 s. j 'OU ' ' ". -
m
~ ' - *?. ,'
$* *f- ? *J 't, .
- .~
t v,\. [ ..,
-d,*.*.J .< i' ,5 *
, - s.)
j~i :.M ,
- (,! '= ' . .' .. ,,
. g . , .
- 7. *, ' . ,. ,
h3Y'j ]~, ' , *k ~--
) .g - ,
k
. . 3 l
l$v+ yi' , ., y.. A i * ..
.L.;y,~'.q
- t -
e Mg .
E h.
e 5 .z. . . ,
l r3 N (( i, :.,j .
augm-e ~. .
.' ~ '
b
( (,f
- ^
' g g ,
l h
ps ia , . , .
~
ar
. 27' '_M. . . .. .._' .
a
'.ffe J. ,,
} .- ,,
u . .. +. .
en . . _
., Y '% ;
NR . ; ~ *, l w . lhi[
.~: ;>. ,
L
+ waa ^. .- . .
- s. e _
~ 4..
.n < * ,.,
=;
_ ; . j *"
.~
,i g
p { . ' - b u.
. ~
~- " ';, ;<
n- ~$s A e %
1 1
10 1
Table 1. Dose Rates for HIACA Co-60 Groups.
Group Calculated (Mrad /hr) Measured (Mrad /hr) 1* 0.19 0.13 2 0.66 0.61 3 0.21 0.15 4 0.63 0.58 5 0.20 0.14 6 0.64 0.59 7 0.19 0.13 8 0.61 0.56
- To determine the dose rate from a combination c f groups, sum the calculated values, then determine the corresponding actual value from the equation b A =1. 015 ([bc )-0. 060 orientation. However, information from NRC Office of Inspection and Enforcement inspections indicated that the in-service orientation of the detector may be important because i
the detector was qualified for sprays " r ight-side-up . "
l' Therefore, the detector was tested right-side-up, with the cable doubled back to the chamber penetrations. Because we were interested in the detector performance, not the cable performance, most of the cable was enclosed in sealed conduit l
to protect it as well as possible during the test (the short
! cable segment in the detector junction box was exposed to the l environment).
l l
We were very careful to follow the instructions 7 ,8
' provided with the detector to ensure that its Class 1E qualification was not compromised during installation. The instructions are the same as those provided to the utilities, and include directions for installing, connecting, and operating the detector.
3.1.4 Instrumentation Pressure and temperature instrumentation were employed to monitor the test conditions given in Section 3.2, and the output of the test specimen was rer.crded digitally as well as with a strip chart recorder. The analog monitor provided with the detector also provided a visual indication of the detector output. The test facility is equipped with a variety of data acquisition systems capable of monitoring and recording the i necessary pressures, temperatures, ramp times, and other pertinent data. The instrumentation channels for monitoring the test specimen's response were on a separate data acquisition system from the environmental monitoring. Both
- 11 l
l
1 I
l
__ _ l g -
% s i
@ 7 @l ~~ f jf~ gg i6 ,,
REF. MAAK g $
L N r,F MARK R TYPICAL
^
n $gE' r A I
E. _. J .
@ @c 3 @ B '
8 '
@@@ O
.@%@ C @@ '
i@' E ig l ' g@! B t. C 8 @
,@ @. g Figure 5. Thermocouple Locations, acquisition systems were connected to a cotaputer to facilitate analysis.
Figure 5 shows the locations of the thermocouples used to monitor the temperature inside the chamber. We computed average chamber temperatures using thermocouple numbers 9 through 19 and 9 through 24. The first average indicated how the temperature immediately around the ionization chamber compared with the overall chamber average. We used the second average to control the environment in the chamber; it is also the chamber temperature reported in the figures is: section 4.
The two averages compared well, differing by only about a degree throughout the test. It did not prove feasible to mount thermocouples inside the detector to monitor the temperature of the ionization chamber, as we did not want to interfere with its Class 1E qualification status.
3.2 Test Environment 3.2.1 Aging The test specimen was not aged in this test. There are no age sensitive materials in the detector itself; the only such 12 l
.\ .
s7o 340 F
~
d l320F
'80 l300F
- 280 F y 60 psg aro -
55 psg (sau 250 F I
88' 200 F E P sattrated corduons e 110 -
120 -
ra
.L ..o . ..o u s o.; ~ 4 o.v. so o.v.
o.o io ...
Time (not to scale)
Figure 6. LOCA Environment Profile, Based on IEEE323-1974.
materials are contained in the Rockbestos cable and interconnections. We were most interested in the detector's behavior, and did not want the cable or connector to obscure the detector's response by failing during the test.
3.2.2 Accident Simulation The LOCA test was based on the profile given in IEEE
, 323-1974, except that we used only one temperature ramp (Figure
! 6). Most Class 1E equipment (including high-range radiation i
detectors) is tested using the two-ramp profile. Most l qualification programs apply radiation (aging and accident) before the LOCA steam simulation. However the intent of this test was to subject the detector to a more realistic simulaticn of the design-basis accident, so we applied the accident radiation dose simultaneously with the LOCA conditions. Also, most qualification programs specify a total integrated dose (TID) , but do not specify accident dose rates. We simulated, within the capabilities of the test facility, the radiation profile given in IEEE 323-1974, Appendix A (Table 2 and Figure 7 in this report).
' We wanted this test to be as realistic as possible within the given constraints. One of the questions we researched before the test was, "When during the LOCA profile should the maximum radiation dose rate occur?" We did not want to expose the detector to a maximum dose rate during the maximum l
temperature and pressure if that was not predicted to be the l
l 13 I
.J
Table 2. IEEE 323-1974 Radiation Profile (PWR).
IEEE 323 specification Calculated dose rate 4 Mrad after 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> 4.0 Mrad /hr 20 Mrad after 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> 1.45 Mrad /hr 24 Mrad after i day 0.33 Mrad /hr 40 Mrad after 10 days 0.074 Mrad /hr 55 Mrad after 1 month 0.031 Mrad /hr 110 Mrad after 6 months 0.015 Mrad /hr 150 Mrad after 1 year 0.0092 Mrad /hr 1Et -
5 Eo suo :
6 E 4 :
II -
a i
~ ' E E
~
e 3 :
E 1E-2 r
$ 5 i o'o o
~
o.i i io ioo i o'o o o Time (hours)
Figure 7. Dose Rates Calculated from IEEE323-1974, Table A1.
actual case during an accident. To answer the qugtion, we looked at some results from the MARCH computer code . (Note that radiation doses for design basis accidents are based on predictions of severe acc:ident environments.) The MARCH predictions for the accident environments varied: the dose rates were predicted to peak between a few minutes and several hours after the LOCA ramp, depending on the specific plant and accident. We chose a representative time, starting the radiation profile about 15 minutes after the LOCA ramp and reaching a peak at about 50 minutes. The steam pressure and i temperature were still at a maximum at this time (Figure 6). l 1
Thus, about 15 minutes after the temperature ramp, wg began stepping up tne radiation dose rate to about 4 x 10 14
rad /hr, the maximum that the HIACA could supply, by raising the cobalt pencils up in groups. Enough time was left between groups so that the effect of the radiation dose on the detector output could be assessed. After the maximum dose rate was reached, the dose rate was stepped back down by lowering the cobalt rods in groups g to simulate the radiation profile given in IEEE 323-1974 and achieve the total integrated dose of 150 Mrad. ^
not match The radiation dose rates available in th$0 the levels calculated from IEEE 323-1974 exactly, so the exposure time for each dose rate was adjusted accordingly to achieve the proper TID (Table 3 and Figure 8).
Table 3. HIACA Dose Rate Profile Used for Test.
Groups Dose Rate Dose Cumulative Dose (Mrad /hr) (Mrad) (Mrad) 0-5 min 1 0.13 0.011 0.011 5-10 1,6 0.34 0.028 0.039 10-15 ?.,5,2 1.01 0.084 0.123 15-20 1,5,2,6 1.66 0.138 0.261 20-25 1,5,2,6,4 2.29 0.191 0.452 25-30 1,5,2,6,4,8 2.91 0.243 0.695 30-35 1,5,2,6,4,8,3 3.13 0.261 0.956 35-90 1,5,2,6,4,8,3,7 3.32 3.043 3.999 1.5-11.1 hr 1,4,7,8 1.58 15.221 19.22 11.1 145.7 1,7 0.33 44.4 63.6 6.1-30 da 3 0.15 86.4 150 Notes:
The actual radiation profile was begun about 15 minutes into the LOCA test.
The times were adjusted slightly during the actual test to account for change in dose rates during calibration l checka.
I 3.2.3 Dry run tests Before any testing of the actual detector took place, several "dry runs" were made to ensure that the test chamber and penetrations were adequate and that the steam environment could be simulated. A dummy test specimen (Figure 9) that duplicated the geometry and thermal mass of the test specimen was used during the dry runs to make sure the test conditions would be met in the real test. Samples of the Rockbestos coaxial cable ( RS S- 6-10 4 / LD ) were included to assess the effects that the steam environment might have on their ability to transmit the low-current signal. One of the cables was installed in sealed conduit (as in the real test), and one sample was directly exposed to the environment. Both samples penetrated the chamber and were connected to a current source arrangement designed to detect differences between input 15
I 1E1.
- tseer pset of ororess' peeta s
g.
r.:
I h !
I a
o SEo r -- ,, .
I I
fe E
.. ; ie !
[ 1E-1 -
'o o.s o 1.00 1.80 8.00
- The ers) 2* : .
- i \
1E-t = l E
o :
o : l
- I l
IE-8 3 ,', ,
- i, ,,, ,,,,,
c,t Time (hours)
IEEE323-1974 HlACA ;
i (calculated)*
- trne shited by 0.5 hour5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> to concde mth HACA 1
Figure 8. Dose Rate Comparison: HIACA vs. IEEE323-1974. l u-
..; .n
,3 ,7 *!' . 4 i
s' k b',I
~
n; l 1Q .
L s .- j 3 ,
- - i
' -,. l l
- r. .
I ;W5 m
~
M .y1h, e e .,r' ~
,5a
.' .. k
.ssww Figure 9. Dummy Test Specimen.
16 l
[ _.
' ]
g 5, 'g L > current and output current. No changes in signal were.
detected, although the insulation resistance of the cables l
decreased with increasing temperature as expccted. U~ also noted that the cable's insulation extruded through the'b-a.s in some spots, a condition that may apply to other coaxial cables as well and may warrant additional investigation.
l 3.3 Monitoring of Test arminen ,
f
- The detector puts out an analog signal and during~tha test-was connected to the RP-2C analog signal processor and readout
+ module that is part . of the overall- GA RS-2 3 A _ high-range radiation monitoring system. We recorded the readout module '
signal digitally with the test facility's data acquisition '
system, and we also used an analog strip chart recorder to facilitate execution of the test and to-provide continuous data between scans by the data acquisition system. The detector
' output was monitored before, during, and after the LOCA steam test.
3.3.1 Protest Characterisation l After the detector was installed in the chambar and the chamber was in place in the test f aci?.ity , but before the LOCA
, test was initiated, the cobalt rods teere brought up in groups, exposing the detector to the levels of radiation it would see durirg the actual test. The output of the detector was l
recorded for each of the levels, and this information was used l
as r, baseline comparison during the remainder of the test.
l This was done because the radiation calibration of the facility (see section 3 .1.1) was accurate only to about 15%, and we wanted to be able to detect smaller changes in the detector's l
l performance. Table'4 in section 4 includes these baseline l radiation values. The first two entries of the calibration, L
1.5 rad /hr and 218 rad /hr were obtained with the cobalt fixture When the elevator is down, elevator down and up respectively.
the cobalt at the bottom of the HIACA pool gives essentially no radiation dose rate at the water surface; the 1.5 rad /hr signal is generated from the internal detector keep-alive source.
When the elevator is in the up position, the cobalt source is close to the water surface and the reduced shielding leads to a dose rate of 218 rad /hr (based on the detector reading).
3.3.2 Functional Performanco During Test 3.3.2.1 Output Signal During the test, the detector monitored the level of radiation in the test chamber, and the signal was recorded both on a strip chart recorder and digitally by the data acquisition system. The RP-2C readout module also provided a visual j output. The output was compared to the known levels based on j l
l 17 1 1
i
i the pretest calibration measurements to determine if it was j operating properly. l I
As shown in Table 3, the radiation test profile was )
lengthy; it required a given dose rate to be maintained for l hours or days. This condition would only allow the detector to j be checked ae the existing dose rate, so we ran calibration {
checks to check the detector's performance over its , entire j' range. During these checks the cobalt rods were lowered and then brought up in the prescribed sequence as done in the pretest calibration check. The calibration checks took only a short time to conduct, and we adjusted the timing of the radiation profile accordingly so that the overall dose to the
! detector was not significantly altered.
3.3.2.2 Insulation Resistance 1
During the test, the detector signal at the readout module showed some anomalous behavior at low dose rates. To try to determine the cause of this behavior, we measured the insulation resistance (IR) between the center conductor and
, shield of both the signal and high-voltage cables at various times during the test. For these tests, we removed the radiation, turned the power to the detector off, disconnected the detector from the readouts, and measured the IR of the cables going into the detector. The measurements were made at l applied voltages of 50V, 100V, 500V, and 10LOV.
I 3.3.2.3 Effect of Cable Length Before the test there was some concern that, while the detector might operate satisfactorily with the relatively short cable between it and the readout, a longer cable more representative of containment applications might cause l difficulty due to capacitance or insulation resistance effects. To address the capacitive effects question, we connected an additional 250 feet of cable (not exposed to the accident environment) into the system at appropriate times to deternine if there was any effect on the keep-alive signal. No l
effectc were apparent, but it should be pointed out that this small ?xperiment does not address the cable behavior in an accident environment, which was beyond the scope of the overall test progra.. Further, it did essentially nothing to a.ddress the insulc.cion resistance question (since the cable was not exposed to the harsh environment).
3.3.3 Post-test Characterization Afmer the LOCA exposure was complete, the rods were brought up once more, in the "me order as before, to assess any change in calibration that t. : / have occurred during or as a result of the test. Af w the test, the electrical characteristics of the detec ? were analyzed in detail (see sectica 5).
18
3.3.4 Acceptance Criteria This was a research test, so we were most interested in how the detector behaved and why, rather than establishing Class 1E qualification. As such, we did got set specific acceptance criteria, but we did use RG 1.97 to assess the results. Regulatory Guide 1.97 states that the detector must be accurate within 20%, and the system (including cable, connections, readout) must over the entire range.
be accurate within g factor of The GA specifications also list two the system accuracy as "a factor of 2 over the entire range of the monitor." Although we attempted during the dry run tests to characterize the cable, we really did not have enough information on the connections' characteristics to evaluate whethar the detector itself was within the prescribed 20%.
l l
l l
l l
l l
l l
l i
19 1 l
i ,
t i
~
3 ,
4.0 Test Results Figures 10a-j show the output of the radiation detector as a function of time, along with plots of the temperature and pressure inside the chamber. Figure 10a, the first hour of the test, shows the pretest calibration of the detector (see section 3.3.1), the LOCA ramp, and the beginning of the soecified radiation profile (section 3.2.2). Note the steps in the radiction profile, the result of raising up the cobalt rods in groups. (The calibration run also contains steps, but they I are not evident on this time scale.) Figure 10b shoNs the l first 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> of the test, Figure 10c the firnt 96 hours0.00111 days <br />0.0267 hours <br />1.587302e-4 weeks <br />3.6528e-5 months <br />. l Figures 10d-j then show the remainder of the test, in 96 hour0.00111 days <br />0.0267 hours <br />1.587302e-4 weeks <br />3.6528e-5 months <br /> l increments. The spikes in the . radiation profiles are due to the calibration checks that we made throughout the test.
Although during most of the test the detector provided an adequate measure of the radiation dose rates present in the i
chamber, sometimes it failed to do so, particularly at the ,
l lower dose rates. The first anomaly occurred during the {
l pressure / temperature ramp. The HIACA elevator had not yet been brought up, so the only radiation at that time was the 1 rad /hr {
keep-alive signal. Figure 11 shows the erratic output of the j detector during this initial portion of the LOCA exposure. {
Initially, coincident with the ramp, the output went high, to !
about 300 rad /hr, followed by a decay until the output was lost at about five minutes. It then came back and fluctuated significantly rangipg from a loss of output on the low end to readings above 10 rad /hr at the high end (e.g., at around l
O.18 hr). It disappeared again, but returned fourteen minutes into the test, when the HIACA elevator was raised to give a dose rate in the chamber of 218 rad /hr (based on the pretest calibration). It should be noted that some behavior noted in the early part of the test might have been caused by piezoelectric cffects in the cable.
At this point the detector began satisfactory operation that lasted about two days. Three low-level signal checks were made during this time (by lowering the HIACA rods and elevator), one 2.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> into the test, one about 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> into the test, and one after about 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />. Some variation in the accuracy of the signal was evident, but the values were still very cloge to The RG 1.97 .
within a factor of two, the accuracy specified in check at 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> gave a detector output corresponding to 3.8 rad /hr, slightly greater than a factor of two above the baseline of 1.5 rad /hr; the reading was well within a factor of two at 218 rad /hr (222 rad /hr reading). It should be noted that the LOCA conditions went from superheated to saturated between 6 and 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> into the test.
About 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> into the test, the strip chart recorded a blip in the detector signal (Figure 12a), and around 52 hours6.018519e-4 days <br />0.0144 hours <br />8.597884e-5 weeks <br />1.9786e-5 months <br /> some 10% variations were evident (Figure 12b). A fourth low-level check, about 53 hours6.134259e-4 days <br />0.0147 hours <br />8.763227e-5 weeks <br />2.01665e-5 months <br /> into the test, showed a loss of signal from the detector with all rods down, but the HIACA 20
+,-e
. . (;. N. .
s s
a, h
~
J s' - ' '
- n 3ee -
- - ~---
60
- l. \f~ ,
. g I Temperature -
40 -
2es -
C - I _ . . . _
- Pressure
?'
y
~ .
Q
- -. l -
h - -
g
( -
20 ;
t les -
I -
g
.t - -
o .
a.
g .
- I
' 1 '-' " ' ' " - ' ' ' ' ' ' ' ' ' ' ' ' ' ' '
-0 8
.4 .2 8 '.'2 ' . '4 .6 .'8 1 Time (hr).
4.E+6 -
/
- Detector Output f 3' ~ - --
Reference I E 1 - I b -
g 2.E+6 -
e 8 -
~
5 - '
t" 1.E+6 -
a :
t t . . .
q ~. t . . . 1 . . l . . .At t . . t . . . . . .
.4 .2 8 .2 .4 .6 .8 1 l
Time (hr) l i-Figure 10a. Chamber Pressure and Temperature and Detector Output.
l 21 i
-?; , .
, c.
l' C l
- - - ^y 3, 4, ,- ---
60
.I l ---_\ -
g Temperature -
o 2e -
- Pressure I
~
7,3 o '64
- . I -
n.
l ~g -
\ . 4 l
L t les
- \ -
20 s
l cI t ~
'-t =
c g l , , i , , , i , , , t , , , i , , . I . - '
O O 2 4 6 8 18 12 Time (hr) l l
Detector 4.E+6 -
, Output Reference n -
p 3.E+6 -
2 . ~
b -
g, 2.E+6 -
- c. - _- -__- -e
- l 8 -
g : , 'l t* 1.E+6 -
a : .
e i2--- L a 2--- L a- ' '
l ' - < - ' ' ' ' ' ' ' ' '=
8 2 4 6 8 18 12 Time (hr)
Figure 10b. Chamber Pressure and Temperature and Detector output.
22
+
4
+
60 3se L~ ~
- k
[
~
L
~l .
2ee - I Temperature 40 -
- ._m O.
I - - - Pressure a
=_ .
o.
. l m .
= _ u,
% - s l a 20 a les -
- - - ~ - - - - < -
y _
qu _ _ , .
o-2 .
~
0 8 80
- e 2e 48 68 I Time (hr) l
\ .
l 4.E+6 7 Detector l Output l
- - Reference 3.E+6 4'
2 b
\ 2o. 2.E+6 - ,
E>
c O
t" 1.E+6 -
l
! r3 q_ _ _ _ _ _
l b' ' ' ' '
8 8 28 40 60 80 Time (hr) l i
I l
Figure 10c. Chamber Pressure and Temperature
) and Detector Output.
l.
l 23 ,
i
. .. 3; 3
a .,
s
.T o 3m -
'60 t 1 288 -
Ag ,
._m
.u Temperature ; &"
- - - Pressure E
- u s e 20 a
= lee -
U E
1 E , -- , - - :- -r , ,'
-t - t- , , , -r : , , i g 8
100 120 148 160 180 Time (hr) 1.E+6 Detector 808800 -
Output Reference p688800 '
36 .
%se
' "' ~-- -~
" 4eesee 8
f--
6 .
$298888
?!
- -\
d -
g I , , , t , , , I , , , i , , , 1 , ,
100 128 140 ISO 188 Time (br)
Figure 10d. Chamber Pressure and Temperature and Detector Output.
24
~~
388 - -
60 288 -
40 -
m C .
'G t
3--
t
. Temperature 6 _
3 . - - - Pressure .
U 188 - 20 $
=
3 et I e v -- -- --,---1 .,-t- .- --
ri7m~" '
g 0 288 228 248 268 288 Time thr) l 580888 i .
Detector 488888 -
l . Output I -
- - - Reference n .
N'"
2
~
7 b -
l 1288888 p 5
a ti 188888 i c8 I i i . i . . i .
i g . t . . . . . . . ,
288 228 248 268 298 Time (hr)
Figure 10e. Chamber Pressure and Temperature and Detector output.
25 l
l
)
4
\
60 3ee 2ee -
40 -
-G .
.?
1
. --- Temperature .
g - e s
. - - - Pressure t ise -
20 :
& W -
I' A .
r~~7 7 ~"r - ' " ~'~
0 e' 300 328 348 360 38e Time thr)
Seeees iseRee -
Latector
- Output n -
Reference k*
T
~
3= :___________
3 28 eses .
3 -
[-
g :
- isesse -
3 -
a :
,. . i . . . i . . . i . . . i . . . i 3ee 320 348 36e 388 Time (hr)
Figure 10f. Chamber Prer.aure and Temperature and Detector output.
26
5 .f A- t w .
e s
'60 388 -
40 -
288 -
m .
.v Temperature -
g .
- - - Pressure -
g y .
b tes --
20 0*
t . *-
3 .
y wm , _ , - i _ , -r----.,-t.'~.- g g.y . - -r-448 468 489 400 428 Time (hr) 588000 488888 -
Detector
- Output Reference l s k
1
(
a ;_ _ _ _ _ _ _ _ _ _ _. _ _ _ _ _ _ _ _ _ _
t 2eeeee -
.? - __
8 -
I 8 .
I t lessee -
l 3 -
a :
i t . .
g ". . , t , , , t , , , , , , ,
498 420 440 460 489 Time (hr) i l
Figure 10g. Chamber Pressure and Temperature and Detector output.
27 b
1 L .. i
.. i k' <
1 1
I 388 - --
60 288 -
40 _
- v. . ' en
- . T.
Temperature ; &
b "
L. %u .
- - - Pressure -
s 4
. Ise - -
20
i: .B..
. - U a.
. . . r -- m . i . r. --. v. - . , - i,,~,'
e 8'
, 488 50s 528 540 568
.. Time (hr) 5081e8 4essee 1 Detector
- Output
--~
Reference n .
N*
2'
.~
,sressee -
B- --
8 -
, u O -
t; Isasse -
3 -
a :
. . . i . . . i . . . i . . . , .
e . .
488 5ee 529 540 560 Time (hr)
Figure 10h. Chamber Pressure and Temperature and Detector output.
28 j l
t i.
3ee -
60 288 -
40 -
.?
M w
~
Temperature ; 3-u .
- - --Pressure E
$u o
20 a e lee -
U N .
ct l g 1 i . r-- -
1- 7 - ~, - i , . . i .- - --- r -- - 0
!- 58e See 628 648 668 l
Time (hr) 588888 4essee Detector
- Output Reference f' 3seges .
R -
a ;_ _ _ _ __ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _
g 298888 -
e 8 -
~
. t .
l t; lessel; -
l 3 -
S ".
~ t , , t . . . i , , , i . . . i , .
g ,
500 See 628 648 668 Time (hr)
Figure 101. Chamber Pressure and Temperature and Detector Output.
29
~
3es - -
60
~
288 ~ -
40 -
G . .
.?
- t ~
Temper ature 3-u , .
a e
- - Pressure -
g b tes - -
20 _
[ .
w a.
e ' T ~ '~ ' ~' 7 7 ~~
' ~ ~ ' ~'
O 680 690 798 718 720
, Time (br) 588880 4 esses h - Detector
. Output Reference N 38"" .-
2 e .-
a - -
______1,
& gaggge .
8 - Il l
,1 o -
t lessee -
it a :
g . . . i . . . . . . . . . i , . . . i . . . . i 688 690 700 710 720 Time (br) i Figure 10j. Chamber Pressure and Temperature and Detector Output.
l 30 l 1
258 g e- -
Detector
- Output 3 ig ; --
Reference b
2, .
1 les 5 -
$ 58 1 it -
S
'-""l - - ' ' ' '" - ^- ' I' e
8 .85 .1 .15 .2 ' .?5 Time (hr)
Figure 11. Detector Output at Beginning of Test.
l elevator still up (dose rate about 218 rad /hr). A calibration check (#2) was then made; the results are given in Table 4.
The test was continued, and more variations in the detector signal were evident around 56 hours6.481481e-4 days <br />0.0156 hours <br />9.259259e-5 weeks <br />2.1308e-5 months <br /> (Figure 12c). The calibration procedure was repeated at about 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> (#3) with the same results as #2. Again, the accuracy appeared to be decreasing somewhat, but was still within a factor of two at the higher levels. Unfortunately, the HIACA did not have the capability tg provide dose rates between 218 rad /hr (elevator up) and 2x10 rad /hr (one rod group up), so we were unable to establish thresholds at which the signal disappeared.
About 75 hours8.680556e-4 days <br />0.0208 hours <br />1.240079e-4 weeks <br />2.85375e-5 months <br /> into the test, the strip chart recorded an aberration in the detector signal (Figure 12d), and we ran a calibration check (#4 on Table 4) an hour later. We then decided to check the insulation resistance (IR) of the cables leading into the detector (see section 3.2.2.2); the results are given in Table 5 (#1). The signal cable showed reduced IR, while the high voltage cable showed relatively high IR (perhaps because of the high voltage keeping moisture out).
At 161 hours0.00186 days <br />0.0447 hours <br />2.662037e-4 weeks <br />6.12605e-5 months <br />, the strip chart recorded another blip (Figure 12e). A calibration run (#7) was made at 168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br />.
For the next5 24tohours, 2.0x10 ye signal varied somewhat, increasing rad /hr in 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />, holding for 3
,. from 1.7x10 hours, then decreasing to 4.7x10 4 rad /hr in 9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br />, at which point (192 hours0.00222 days <br />0.0533 hours <br />3.174603e-4 weeks <br />7.3056e-5 months <br /> into the test) calibration #8 was done.
l
- Following an IR check at 193 hours0.00223 days <br />0.0536 hours <br />3.191138e-4 weeks <br />7.34365e-5 months <br />, the detector signag
! was very slow to respond when a nominal dose rate of 2.26x10 rad /hr was applied. The readout slowly increased from about 31
s,-
Table 4. Calibration Data.
- 1 #2 #3 #4 #5 #6 #7 #8 Original 53 71 74 75 102 168 192 Calibration- hr. hr. hr. hr. hr. hr. hr.
T-Ambient' T-122*C T-123*C T-123*C T-123,'c T-100*C T-100'C T-99'C 1.5 0
- 0 0
-1004 -1004 -1004 218 0 0 0 0 .
0 0 0
-1004 -100% -100% -100% -100% -100% .-100%
2.03E+05 1.55E+05 1.94E+05 7.46E+04.3.00E+04 1.97E+05 1.44E+05 2.44E+04
-23.6% -4.4% -63.3% -85.2% -3.0% -29.1% -88.04 4.11E+05 3.69E+05 3.90E+05 3.03E+05 2.28E+05 4.17E+05 4.22E+05 2.05E+05
-10.2% 5.1% -26.3% -44.5% "1.5% 2.7% -50.1%
1.21E+06 ****
1.18E+06 1.08E+06 9.56E+05 1.23E+06 1.11E+06 8.87E+05
-2.5% 10.7% 21.0% 1.7% -8.3% -26.7%
- 2.00E+06 2.02E+06 1.97E+06 1.85E+06 1.74E+06 2.05E+06 1.86E+06 1.58E+06 1.0% -1.5% 7.5% -13% 2.54 -7.0% -21.04 2.77E+06 2.79E+06 2.71E+06 2.57E+06 2.44E+06 2.86E+06 2.59E+06 2.24E+06 0.7% -2.2% -7.2% 11.9% 3.2% -6.5% -19.14 3.49E+06 3.52E+06 3.45E+06 3.23E+06 3.09E+06 3.58E+06 3.33E+06 2.87E+06 0.9% -1.1% -7.4% -11.5% 2.6% -4.6% -17.8%
3.71E+06 3.75E+06 3.65E+06 3.46E+06 3.29E+06 3.85E+06 3.54E+06 3.04E+06 1.1% -1.6% 6.7% -11.3% 3.8% -4.6% -18.1%
3.89E+06 3.95E+06 3.87E+06 3.60E+06 3.48E+06 4.05E+06 3.73E+06 3.23E+06 1.5% -0.5% -7.5% -10.5% 4.1% -4.1% ',7.0%
- Values in the tables are arranged as follows:
Indicated dose rate (rad /hr) from detector output.
Error compared to the baseline value given in the first column.
32
Table 4. Calibration Data. (cont.)
- 1 #9 #10 #11 #12 #13 #14 #15
- Original 193 .338 433 529 606 697 Post-Calibration hr. hr. hr. hr. hr. hr. -Test T-Ambient T-99'C T-100*C T-99'c T-99'C T-99'C T-99'C T-Ambient
'1.5 0 0 0 0
-1006- -100% -1004 -1004 218 0 0 0 0 0 0
- -1004 -100% -100% 1004 -100% -1004 2.03E+05 1.59E+05 1.47E+05 1.57E+05 1.55E+05 1.73E+05 1.87E+05 1.28E+05
-21.7% -27.6% -22.7% -23.6% -14.8% 7.9% -36.9%
4.11E+05 3.63E+05 3.52E+05 3.60E+05 3.58E+05 3.73E+05 3.92E+05 3.27E+05
-11.7% -14.4%. -12.4% .12.9% -9.2% -4.6% -20.4%
1.21E+06 1.14E+06 1.13E+06 1.13E+06 1.13E+06 1.14E+06 1.17E+06 1.09E+06
-5.8% -6.6% -6.6% -6.6% -5.8% -3.3% -9.9%
2.00E+06 1.92E+06 1.90E+06 1.90E+06 1.90E+06 1.92E+06 1.96E+06 1.86E+06
-4.04 -5.0% -5.0% -5.0% 4.04 -2.0% -7.0%
2.77E+06 2.67E+06 2.64E+06 2.36E+06 2.64E+06 2.66E+06 2.71E+06 2.58E+06
-3.6% -4.7% 14.8% -4.7% 4.0% -2.2% -6.9%
3.49E+06 3.37E+06 3.35E+06 3.34E+06 3.34E+06 3.37E+06 3.44E+06 3.28E+06
-3.4% -4.04 4.3% 4.3% -3.4% -1.4% -6.0%
3.71E+06 3.58E+06 3.55E+06 3.55E+06 3.54E+06 3.57E+06 3.64E+06 3.49E+06
-3.5% -4.3% -4.3% -4.6% -3.8% -1.9% -5.9%
3.89E+06 3.79E+06 3.74E+06 3.68E+06 3.73E+06 3.77E+06 3.84E+06 3.60E+06
-2.6% 3.9% -5.4% -4.1% -3.1% -1.3% -7.5%
- The post test calibration was run after about 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> of exposure to ambient air. Note also that Co-60 decays by approximately 1% over a 30 day period, accounting for a small portion of the observed dose rate decreases.
33
. e IE9 ,
IE9 r E
- d .. . . .
- ..I ._ . .,i . . 1
. .j l
.a. cb' . i 00 .
1E. r - l to i 4 - ... -
iE. r g .c.. q . ..,
i y
1E7 r-4e nes- - - -
,- -h IE7 r
. f ..
es hre . --
_.-f-- '
s i ..i. . ~ --.,- i .
~ .
..i.,. s _.- .3 ...i..T . ,: . .. i .,... i
~~
3 . ' b:
e* .. ! . .
I
._ l' . . . . l .I . . .. '
. - . . .I
'- .. ..i',.
.. ., i..
\ ^~"_....i t.
.$3 1E5 r'
- i . i 1E5 r' .
.' ' h, ._ - .
! ' I
. .i . .i.. .~ -
.. -. E i .
i
. .p Q -
al t I i -
I i w
1E4 r .. ; . t [~j . . .i l 1E4 r 4
~ ~~
o 3
_j ._
i .q. .; . . ,-
r i .. - . i. . 3
- 7..
. .l .
z, q,,. = 'a t
O
. .l- ..I ..j. .1.. (I 1ES r _.! . I IE3 r .. - Q
$, E l J. E l
J. .
I I i :
_i i 6 :
._I_
0 1E2 I.- * ~* ~ ~ ~ ~~
1E2
N ~ ~~
l
.A -
"j*l
!. p~;
I '"
.-}_. . I. ..
.h _; ..
I 3 6 I
,j ]
1El - I i i l .- I : '
IE1 -
=
a_ I I~ . f 3
]
-E I
j.-q I
l l I
I t o mie
-E . . . . .
l
.g._
1,0 ,m i n_. _
j' L ,
1EC IE0 I {.__ . . _ I
- a. < Time b. < Time r 'S .
1E0 r
' ! _. 1E9 -- . - - -- - . - -
. c3 .. . ... . -.-
.-f- li E. g ..
IES r 7 '" .t - --
l
_j( i i i 1E8 r ~1 i g ; -j- , ,
l ; ; 3 -
Og Q--
i
_.-.c4. g i' e
i 1E7 r -+t 4-~ i- ,i- - --
i ! l IE7 r,[ t ,<. I *
, i p ! I -] - ' l I ' '
% - , I i i ! ! 1 I I 5
~~~ ~ . l l G' 1Et r "~i ~ i i i 1E6 r 3 _ , . _I '. _~_.l _.
. - ' i. m,
. i
... ] ~ I m.
3 j.. . . ii . .. ....
. r . .l - - !.- E I
i 3
1ES rl..}'I -},._ ll -
l
- ' 5
- I I 1ES rl _l
' i
- l. 6
.C.L s
g i i
_4..
i i i
=
.._ T' ;
., :; , .l .
O -
u IE4 r l
I Tf*6
.se. nee .8--
i i I
-p IE4 r "-~ ~ ~' i ' ~ '~ - ,
! * ' , , . E s.s bro. I
.o.
+ -
I ,
t -_ _*____.1_-.. l _ .I g 1E3 . . ;- .. . . . . . .! _ .] ...L .h- --. 1E3 p' l. l rl
.,_i ._L . : i 6 i ; l ...
c - ; i .
i i # -
O ' ,
i .
l ,
1E2 r!'T-
-I 't i l ,]e~
l i 1E2 g -
,I , , , ~ " j' l '
l
._1
. ~
, . . I , - -
l ..
l
. ).
1Et r I
- j. - . .
. ; . ..- .... j j ..; ..; g g ,
.. ...l..._.1; IE1 [ l I_
T- 3-7j E
i
~L i i ! l i ' t t o ata i i
, co T 1 i 1 I i i i t, i 1 T-t"I- .
~I I
- c. < Time d. < Time l
Figure 12. Strip Chart Record of Detector Output.
34
4 1Et r. 1E9 r
- a j +
_j. ... V 5 .j )
. . -: a
- l t -t - : i 1ES r: iQ _
i.- i-. :H. 1E8 r E
T 7" [
,i. ~ Loi n
,> ~.
, ..i...
2 ^
1E7 r ~l' 1E7 r =-- -_.
p [
. i .. _! F b-.- t . , i
% . 4
.i. .; . . . -
q .- -m-,e z . l 9 ; ;
- ~~~
1**
~
E 1E6
[
.; .:. 7 z.i ccg: . 1E6 y , 'l 1
" 'l ~-
- s i ,
g _
E
.i . . _. l. _ .!..
. ., l. l
' 1 I" . _ I. . : .
r*
CL 1El 1E5 . -.
r'I l' 'l i.Y. -
3 ' t E i i t 1 E
~ '; " i ' 'I' 0 i
+ ,
. . i i "
'*1 1E4 r, 1 i
). . i 1E4
- r. ..,
.. l.
4 . .
O
.. , e is1 nre.. t . , .
I t 707 hre ! >
o.
i
- 0 1E3 r I I I I i f I ! 1E3 r < l i I
.e g
' e i
i :
6 e i l t l O 1E2
[- d ! 1E2 r l
i l
j'~ *i'"
.l
~l i i
- I _ ,;_.. e l..
l lEl r- W 1 1El r I i i l .e-- !
- 10 min : t 10, min >
1 l l 1E0
' I I -3 I' 1E0 i I. I 'I l
- e. < Time f. < Time 1El IES I I I
_.[ E _I_. a 4
- IC : iy I !
1ES _ . __ . . $ IES
[ ,% _,_ 9 _j , _ ,
l
.k- --l.
!.- .d Ly,,, ,g -t w
IE7 r - .. - -. - . - .
g, l i
1E7
_i_5' f s g i . - .d. -
j i.
l .c
- 1 rpk.4 5 7-
"j-
~l' & .'
N - I i t i 1 I l ~~i i
'{[ T r ~7~ l(T ' .~
l l $ 1E6 r. ' i~ 'l I ' }i 1E4 l
./ '
t !' 3
!.. l j .. . . l g :: 5 j._ '
CL 1E5 r..... i . .- --
IES r' .
l .'
o e
.. --*l . . l : t .. . . . ' ,
- i. .
i p..i '
i t 0 -- i_,_ ..'_ ,i _ _ . . _
l_.
,g, w
' ' ' ~
- l i 7 31 hre
.... 714 hre. 1 l !
4
.o. _ ...q l .
l g 1E3 rl ..!. .l ! .l ! . 1E3 rl . l .
l
' i
- E L.L J _ .' . . . . .___ !
__ J _ .. . .' . ' . . _ . _ _ _ _
l 0 1Et l l-i l l i
i iE2 r s
'J.. _' ,
. . ,! . ._I
- , 4 p I
-rl p. .)
t _ ,j
- ; j 1Et .l to . .M. ein
!. IE1 -
t o min
=
..,..}..'.
- ' i i i '
.,q..-a
-l, : ....i..
- q. l.-.p. j!
- l. 1E.
i q 1 l-
- , . t i. _
- g. < Time --
- h. < Time Figur:e 12 (continued).
35 1
h
1 i l
Table 5. Cable Insulation Resistance During Tast.
I
- a. Signal Cable Time into Test Insulation Resistance (0)
- at (hr) 50 V 100 V 500 V 1000 V 74 9.5E+05 3.3E+05 ** ***
192 1.1E+06 2.2E+06 1-2E+07# ***
720 2.4E+05# 3.5E+05# 4.5E+06# 3.5E+06#
6.0E+06##
6.0E+06###
3 wk after test end 7.5E+05 1.1E+06 1.5E+07 6.0E+07
- b. High-Voltage Cable Time into Test Insulation Resistance (0)
- at (hr) 50 V 100'V 500 V 1000 V 74 6.5E+10 5.0E+10 1.1E+10 ***
1 192 6.8E+09 1.2E+10 2.8E+10 ***
720 9.0E+06 1.7E+07 8.0E+07 8.0E+07
.3 wk after test end 5.0E+11 2.0E+12 5.0E+12 5.0E+10 All measurements taken at one minute unless otherwise noted
- Reading varied excessively and was not recorded
- Measurement not made at this voltage
- Reading was unsteady
- Measurement made at 6 minutes, reading more steady
- Measurement at 11 minutes, after reading was stable 1.2x10 4 rad /hr to 1.15x105 rad /hr (within a factor of two of the nominal value) in 30 minutes, followed by essentially complete recovery (at this dose rate) about 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br /> later.
For the remaindy of the 30-day steam exposure (dose rate at nominally 2.26x10 gave low readings, except during calibrations), the system but within a factor of two. The error varied somewhat, but was typically between about -5 and -25%.
At 702 hours0.00813 days <br />0.195 hours <br />0.00116 weeks <br />2.67111e-4 months <br />, the radiation source was removed and the detector signal was lost. The steam system was then shut down and an air line was connected to the chamber to dry it out The radiation was again applied at nominally 2.26x10 g rad /hr. During the next 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br />, the detector signal accuracy worsened in the negative direction and was almost precisely a factor of two low when it stabilized. The decreases in accuracy were due mostly to two discrete steps, rather than gradual degradation (Figures 12f and 12g).
At 721 hours0.00834 days <br />0.2 hours <br />0.00119 weeks <br />2.743405e-4 months <br />, we again measured the IR of the connecting cables with the results shown in Table 5. The results were 36 l
I low, considering the ambient- temperature and that the chamber had been drying oilt for almost a day. We noted significant fluctuations during the measurements, but they would eventually stabilize, indicating that moisture was still present in the cables / connections and was probably being driven off by the applied voltage. When power was restored to the detector, the accuracy of thesygtemhadimprovedsignificantly--theoutpug went from 1.1x10 rad /hr before the IR checks to 1.7x10 rad /hr immediately after (Figure 12h). At this point, a final calibration check was run with the results given in Table 4, calibration #15.
Three weeks later, the detector was checked for the low-level keep-alive signal, but it had not returned. We then repeated IR measurements on the cables and found they had improved significantly as shown in Table 5. We observed that the readings were steady, unlike the readings taken at 721 hours0.00834 days <br />0.2 hours <br />0.00119 weeks <br />2.743405e-4 months <br />. At this point, we removed the detector from the chamber to conduct post-test analysis. We noted that the jacket of the signal cable had a small split where the cable entered the sealed conduit (see section 5).
I We began the post-test assessment of the detector about two months after removing it from the chamber. At this point, we discovered that the keep-alive signal had returned. The next section discusses our findings from the subsequent examinations and measurements that we made on the detector and readout module.
l l
l 37 l
l
5.0 Analysis of Test Results 5.1 Data from Bench Tests The major anomaly noted during the test was the loss of the detector signal at low dose rates at different points during the test. The keep-alive signal loss occurred soon after introduction of steam into the test chamber, followed by return of the signal after several minutes when we increased the dose rate to 218 rad /hr by raising the elevator. The keep-alive signal was then detectable for the duration of the
.superheated steam exposure and for some time during the saturated steam exposure (based on three checks as stated in Sec. 4.0 above). It disappeared again some time between 30 and 53 hours6.134259e-4 days <br />0.0147 hours <br />8.763227e-5 weeks <br />2.01665e-5 months <br /> into the test (saturated steam conditions). At 53 hours6.134259e-4 days <br />0.0147 hours <br />8.763227e-5 weeks <br />2.01665e-5 months <br />, neither the keep-alive signal nor the 218 rad /hr nominal dose rate would produce a reading from the detector.
Several observations during the test led us to believe that the signal loss had been caused by . low-level leakage currents in the signal cable: (a) the signal was lost during the time that moisture was most prevalent in the chamber, i.e.,
during saturated steam conditions and in the early part of the superheat conditions when moisture would have been condensing on the relatively cooler components in the chamber (higher moisture conditions are typically associated with increased leakage currents) ; (b) insulation resistance measurements made on the signal cable indicated decreased resistance values; (c) we reasoned that leakage currents in the circuit would cause a path for the detector signal to leak to the circuit common and hence never be seen at the detector readout module; and (d) the small split in the jacket could have provided a convenient entry point for moisture, although we would have expected moisture intrusion through the polymeric cable jacket anyway.
We conducted post-test analyses in an attempt to verify the cause of the loss of the keep-alive signal and determine experimentally the magnitude of the readout module error at various input current levels and various cable insulation resistances. We used a picoamp current source to simulate the detector signal, various resistors between the coaxial center conductor and shield to simulate lowered IR values, and a voltmeter to monitor the readout module output. The voltmeter was used for indication only and was not calibrated. (Lack of calibration is unimportant for three reasons: (a) the primary importance of the readings is how they compare in a relative sense to baseline readings, (b) the voltmeter readings corresponded very well to the readings on the readout module meter, but were much easier to read and record, and (c) the calibration curve for the detector, based on the voltmeter readings, was very nearly linear.)
The results of this experiment are shown in Table 6.
Because this experiment was looking only for trends and causes, 38
.' o Table 6. Experimentally Determined Outputs of the Readout Module for Various Insulation Resistances and Input Currents.
Input Simulated Leakage Resistance Current Dose Rate
,(A), (rad /hr) Infinite 22_HQ 19.JQ 1.1 MD 105 kn R hD Short 10E-11 10E0 0.000
- 0.251 0.375 0.449 0.471 l'. 0 1.1E2 1.1E3 4.5E3 6.8E3
+1.1E4% +1.1E54 +4.5E5% +6.8E5%
10E-10 10E1 0.120 0.147 0.164 0.257 0.471 9.6 16 22 1.2E2 6.8E3
+67% +129% +1.2E3% +7.1E44 10E 9 10E2 0.244 0.248 0.251 0.285 0.471 97 1.1E2 1.1E2 2.1E2 6.8E3
+134 +13% +116% +6.9E34 10E-8 10E3 0.369 0.369 0.370 0.374 0.471 1.0E3 1.0E3 1.0E3 1.1E3 6.8E3 04 04 +10% +5804 10E 7 10E4 0.496 0.496 0.493 0.483 0.471 1.1E4 1.1E4 1.0E4 8.6E3 6.8E3 On 94 224 -384 10E-6 10E5 0.619 0.610 0.567 0.471 1.1E5 9.2E4 4.1E4 6.8E3
-16% 634 -944 10E-5 10E6 0.734 0.733 0.726 0.682 0.471 9.4E5 9.2E5 8.1E5 3.5ES 6.8E3
-24 -144 10E 4 10E7 0.858 0.858 0.851 0.806 0.471 l (9.6E6) (9.6E6) (8.4E6) (3.6E6) (6.8E3) 0.0 ** *** 0.087 0.125 0.245 0.368 0.450 0.471 (5.2) (11) (99) (9.9E2) (4.6E3) (6.8E3) l
- Values in the table are arranged as follows:
a) Voltage measured at the readout output (0 1 V) b) Corresponding radiation dose rate from calibration curve of best fit semilog line of voltage vs. simulated dose rate:
RC -1.0122*exp(18.72*V) where V is voltage and DRC is the calculated dose rate. (Correlation coefficient-0.99984.)
c) Percent error calculated as the ratio of the calculated dose rate with the given leakage resistor minus the calculated dose rate with infinite leakage resistance to the calculated dose rate with infinite leakage resistance times 100%.
- Simulates a loss of the detector signal. Percent error meaningless.
l
- Detector went into fault mode signalling loss of keep alive signal.
1 39
we did not make measurements at all values of detector current l and insulation resistanco. Table 6 shows that, as expected, I the errors at the low end of the range have the most severe accuracy degradation. However, the error at the low end of the range was n21 in the direction we expected (i.e. had the opposite sign), nor in the direction we had observed during the steam exposure (loss of signal). At the high end of the range, the error was in the expected direction with less severe accuracy loss than at the lower end of the range.
i Because of the unexpected results, we decided to submerge !
the detector, cables, and connections to see how the detector keep-alive signal would respond. We found that the signal drifted uoward while the assembly was submerged overnight.
This observation is consistent with a decreased IR of the signal cable based on our observations in Table 6; however, we were unable to cause the signal to go off-scale at the low end by inserting any type of leakage resistance in the signal cable. The submergence test also failed to cause the signal to go off-scale at the low end.
I l
We also did tests with resistors between the center conductor and the shield of the high voltage cable to assess the impact of degraded insulation resistance on the high voltage lead. We found that with no shunt resistance, the high voltage supply produced a nominal voltage of 875 V. The voltage degraded to about 800 V when we added a 1 MO resistor across the high voltage cable, to about 650 V when we added 422 kn, to about 460 V when we added 211 ka, and to about 300 V when we added 105 kn. The detector keep-alive signal decreased with lowered resistances, and the operation indicator light went off with 211 kn across the high-voltage cable. We did not assess accuracy over the entire instrument range under degraded high voltages, but we feel that a criterion of 1 Ma for the parallel combination of insulation resistances of the cable, penetration, connections, etc., would be a reasonable specification for the hich-voltace side interconnection.
5.2 Analysis of Data from Bench Test Wo investigated the cause of the unexpected response to insulation resistance effects by examining the circuit schematic for the detector. As a result, we believe that knowledge of the offset voltage characteristics of the readout module's input operational amplifier is critical to assessing the loss of accuracy of the readout module caused by insulation resistance effects. To illustrate the point, an operational amplifier circuit is shown in Figure 13. In this circuit, under ideal conditions, all of the input current is diverted around the input amplifier and through select feedback elements. Under these conditions, the negative terminal of the input amplifier acts as a "virtual" ground, i.e., the voltage across the amplifier inputs, V 1, is very nearly at ground potential. The output voltage of the amplifier must be 40
Fee @ack -
l 13 Rin _y l
1
> 0 M 'O -
o Scpal Condtioning g
y det - + 0-1 V Ostput II Rins o V V V .
Figure 13. Block Diagram of Operational Amplifier Circuit.
l V t
" "A*Y1, where A is the open loop amplifier gain.
v but a typical
( Ne amplifier value might be gain could vary around 2x105. ) er a wide The range, actual voltage V 1 is "adjusted" by the feedback elements so that the desired closed loop properties of the output voltage are achieved. The amplifier voltage output range is approximately 12 V to 0 V, I decreasing in proportion to the logarithm of the input
! current. With no input signal, the amplifier output voltage is about 12 V, while with an input signal of 10 mA, the amplifier output voltage is about 0 V. The voltage at the 0-1 V output of the readout module is scaled and offset from the amplifier outgtA so that an input current to the readout module of gives an output of 0 V and an input of 1 mA gives an 10-i output of 1.0 v. The voltage measurements shown in Table 6 are
! taken from the 0-1 V output. As an example of the voltage at V,1 consider the above open loop amplifier gain and an output voltage of 5.0 V. The resulting voltage V is easily calculated as 0.025 mV (indeed a "virtual" ground)1 . Any input offset voltage is automatically compensated for by the feedback elements since the input current is completely controlling the feedback characteristics. This compensation is manifested as the above calculated voltage "floating" on the input offset voltage. In the above example, if the input offset voltage were +1.5 mV, the actual voltage V, would be -1.5 mV + 0.025 mV = -1.5 mV, or essentially just the offset voltage (negative offset voltage since V 1 is located at the negative amplifier input). In fact, regardless of the amplifier output over a wide range, V i remains at approximately the negative of the offset voltage as long as the amplifier open loop gain is high.
I Next, consider the effect of finite insulation resistance i and nonzero input offset voltage on the circuit of Figure 13.
Finite insulation resistance will exist from cables, connectors, penetrations, or other interconnecting devices.
The voltage V 1 is nearly at ground potential (-1.5 mV for the case described above). Based on a center conductor resistance 41 i
of 20 ohms, it can easily be shown that the voltage along the center conductor of the coaxial cable is essentially uniform along the length of the cable at any given detector current.
The following analysis dencnstrates how the errors shown in Table 6 may be predicted analytically if the input offset voltage is known. Ac discussed above, V is essentially constant over much of the range of detectori currents if the amplifier gain is high. Thus the input offset voltage can be easily measured as the voltage V 1 with a small input current. For our detector, the input offset voltage was measured as about 1.5 mV (V1 =
-1.5 mV). The meagurement uag checked at several input currents from about 10- A to 10-A with only a slight change in V 1, indicating a high enough amplifier gain to keep V i approximately constant over that range. Referring to Figure 13 and summing currents at node 1 gives:
Il+Idet " Iin (1)
With Il= V i /Rins, (1) becomes:
V
+I R
ins det " Iin (2)
With I d (Y 1 + V o ) / R in, where V os is the amplifier inpuItoffset voltage, s(2) becomes:
V , V +V os 7
ins
=
Q in Solving (3) for V1 gives:
V = R V gg in ins (4) 7 Rin + ins , ,
in _ Rg Finally, using (4) in (2) and rearranging gives:
=I R ins Y I + os (5)
Rin + Rins, Rin + Rins Equation 5 gives an expression for the current actually measured by the readout module for a given offset voltage and a given insulation resistance between the center conductor and shield of the coaxial cable.
To illustrate how well equation 5 works, we will use the input offset voltage of our readout module's input amplifier (approximately 1.5 mV). As a first case, consider an ins 10-gation resistance of 1.1 Ma and a detector current of A. The value of Rin was verified by measurement to be 20 ko. Substituting into equation 5 gives an input current ;
42 l
of 1.46x10-9 A which corresponds to a dose rate of 146 rad /hr. Using the equation from the footnotes of Table 6 gives a readout module output voltage of 0.266 V, which compares extremely well to the measured value of 0.257 V given in Table 6. A second example consider a higher detector current of 10-g aA with an insulation resistance of 10 0.
4 We recognize that this tends to be a low value for coaxial cable in a steam environment, but we use it here for illustration purposes to get a large enough error so that equation 5 can be meaningfully verified. For this gecond case, equation 5 gives an input current gf 3.33x10- A. This corresponds to a dose rate of 3.33x10 rad /hr and a readout module voltage output of 0.679 V, which compares very favorably with the measured val.te of 0.682 V given in Table 6. It should be emphasized that this assessment required the knowledge of the input offset voltage of the readout module input operational amplifier. The offset voltage is a random parameter and might typically be within the range of -3.0 mV to
+3.0 mV. Consequently, without knowing the input offset voltage for a given device, neither the magnitude nor even the direction of the error is predictable. However, given the manufacturer's specifications for the input amplifier, bounds can be put on the IR-induced error as a function of the l detector current and the interconnection insulation resistance by using equation 5.
Equation 5 can also be used to give a qualitative assessment of the readout module's behavior by considering the two terms of the equation separately. The first term represents the loss of signal generated by leakage of detector current to ground; the factor in paren.hasis is always a positive quantity less than 1.0. The second term represents the contribution of the amplifier input offset voltage to the input current and may be either positive or negative. If the input offset voltage is positive (as in our case), it causes additional current to flow into the readout input because V i l
is approximately the negative of the offset voltage and is thus l
below ground potential, causing current to be drawn from ground. A reverse argument holds for a negative offset voltage, but the result is current drawn from the readout module. In this second case, at low detector currents, the readout module will tend toward going off-scale on the low end (due to both terms) . At low detector currents, the second term of equation 5 tends to control readout behavior, while at high detector currents, the first term tends to control the behavior. For any given interconnection insulation resistance, the ur. desirable effects modelled by equation 5 are much more l pronounced for the low detector currents (mainly the second l
term of equation 5).
At this point, we have a model which appears to accurately l
l I predict the effect of interconnection insulation resistance on the readout module's error. This prediction was verified over a range of values for our readout module. We have one 43 l
1
significant problem remaining, that of explaining why in the test the detector signal was lost altogether during periods of high moisture conditions. Obviously, the effect of insulation resistance, which prior to post-test analysis seemed to be the certain explanation, is not the proper explanation. We postulated only one theory which might explain the detector's behavior: galvanic action between two dissimilar materials somewhere in the interconnections to the detector.
Theoreticslly, a voltage induced on the coaxial ~ cable's center conductor which has the effect of reducing the voltage at the readout module input would tend to allow less current to flow into the readout module. At some lower potential which could be positive or negative, the induced voltage at the input would equal the negative of the amplifier offset voltage (-1.5 mV in our case) and would result in no current into the readout module. At even lower potentials, current would be drawn out of the readout module. Any reduced voltage causes the readout module to go in the negative direction, resulting in the eventual loss of detector signal at some level. For our detector, the measured loss of signal occurred very near -1,5 mV, confirming the theoretical discussion at a voltage above. (This observation may provide an additional means of measuring offset voltage by inducing a reduced voltage on the center conductor of the cable until the signal is lost.)
summary, an induced voltage on the center conductor of the In cable approximately equal to the input offset voltage co' tid account for the observed loss of detector signal.
The next question is whether a possible galvanic reaction could supply the necessary voltage and power to account for the observed effects. Certainly a voltage of greater than a few mV is possible with almost any galvanic reaction. Based on the observed lack of signal at a dose rate of 218 rad /hr during our
- test, 1000 we rad might guess that it would stgil be lost as high as
/hr (it did between 218 and 2x10cgme back at 2x10 , but we have no data rad /hr because of the discrete steps available from our radiation source). A dose rate rad /hr corresponds to a detector current of 10-8 of 1000 A. For our system, a pot gal about 1.5x10-gntialV x 10"ganic A = reaction 1.5x10- gould W = need to supply 15 pW, which does not seem to be excessive for a galvanic reaction over a fairly small area. However, we can only very tentatively give this explanation since we did not investigate the root cause completely. We did contact the connector manufacturer and found that the connector uses a brass contact with either silver or gold plating and a brass body with silver plating.
The coaxial cable uses tin-coated copper for both the cr iter conductor and the shield. Thus, diss'-ilar metals do exist and could provide the components for a galvanic reaction.
Galvanic action was the only explanation we could postulate that had a reasonable basis to explain the observed behavior of the detector in the test. In any case, for our system, whatever caused the loss of detector signal at the low 44
6.2 Recommerdations for Radiation Monitor Users Based on the results of this test, we feel that the detector will operate properly under the most severe design-basis accident conditicns; that is, when radiation is at a high level during a LOCA. The problems occurred when the radiation dose rate was relatively low, prcducing only a small signal current from the detector. '1hu s , we recommend that users of the RD-23 detector assess the results' of this test program as it relates to their detector installation. We also recommend that licensees with other types of high-range radiation detectors review their qualification test data to check for similar concerns.
We feel that utilities need to be aware of the following specifics of the observed GA detector response:
- a. Readings at levels below about 1000 rad /hr were not reliable dur'ig accident conditions.
1 i b. Readings at levels above about 1000 rad /hr under accident conditions were accurate even though the green "operate" light on the readout module went off several times (it did come back on when it was reset l- by pushing it).
l l In addition, utilities should be reminded that installations of coniponents and systems qualified by parts may not constitute a qualified system.
6.3 Recommendations for Future Tests As stated earlier, we do not feel additional tests of the l RD-23 detector itself are warranted, as the detector itself I
appeared to operate satisfactorily. However, it may be l advisable to test other types of radiation detectors that could i
exhibit similar problems. We do think the coaxial cable / connection systam merits further attention; perhaps a I general investigation of Class 1E coaxial cable and connections L exposed to harsh environmants should be undertaken.
l l
l l
1 47 j
j I
6.0 Conclusions and Recommendations f 6.1 Detector Performance i
General Atomic's RD-23 high-range radiation detgctor is designed to operate in a range of 1 rad /hr to 10 l
rad /hr !
under ~ extreme pressure, temperature, and moisture conditions.
We found that the detector had no trouble "surviving" these conditions, and' as long as there was a sufficient radiation i dose ra the detector met the accuracy conditions in RG 1.97.pe, However, at "lower" dose rates the detector signal l
at the readout module became inaccurate or was lost altogether due to the reasons discussed in section 5. Based on the I results, we conclude that while the detector "survived" the '
test conditions, the overall system (detector, cable, connections) did not meet the requirements in RG 1.97 fqr thg lower pose rates only.
Two concerns identified with the detector operation are as follows: t
- a. At low radiation dose rates, insulation resistance effects of the interconnections may cause detector
, accuracy to degrade beyond the factor of two allowed j by RG 1.97. The magnitude and direction of the l observed error were shown to depend on the interconnection insulation resistance, the readout <
module input amplifier's input offset voltage, and the detector signal level.
I b. A seccnd phenomenon, perhaps galvanic action, caused f the detector signal to be lost at low levels of i
detector output. We were unable to hypothesize any l f mechanism which would explain the observed loss of l signal other than galvanic action. We also noted that l l
l for our test, this phenomenon appeared to obscure insulation resistance effects at low detector currents. Thus, analysis techniques considering only l
j insulation resistance effects to assess detector accuracy may be of somewhat questionable value. It: is I possible that the split in the cable jacket I contributed to the loss of signal by providing an l
additional path for moisture intrusion into the connector.
Although the potting and conduit system used in our test does not necessarily represent actual installations, the cable split again raises the question of whether separate component qualification is sufficient to ensure tht" a complete, installed system is qualified. We used "qualified" components, I
but the system we created by combining these components created a failure mode which apparently did not show up in any individual component qualifications.
l l 46 1
(
I .
g .
4 APPENDIX A Discussion of split in cable Jacket Upon remeval of the radiation detector from the test chamber, we noted a small split (approximately 3 mm (1/8 in.)
long) in the cable jacket where the cable entered the sealed conduit. The split appeared to have been caused by material degradation resulting from exposure to the harsh environment (high temperature, steam, and radiation) coupled with internal stresses due to thermal expansion of the coaxial cable. At the beginning of the test, the cable went straight from the connector to the potting; after the test, the cable had a (longer) curved path from the connector to the potting, clearly indicating that the cable's length had increased.
The split in the jacket may have contributed to allowing moisture to enter the connector via the braid, causing reduced IR as well as a path for galvanic action. However, there are a number of reasons that indicate that even if moisture entered the connector through the crack that moisture would have penetrated to the connector anyway. Moisture is known to
' permeate polymer materials such as the coaxial cable jacket, and moisture penetration in this fashion takes time. We saw the most significant signal loss somewhere between 30 and 53 hours6.134259e-4 days <br />0.0147 hours <br />8.763227e-5 weeks <br />2.01665e-5 months <br /> into the test (first verified at 53 hours6.134259e-4 days <br />0.0147 hours <br />8.763227e-5 weeks <br />2.01665e-5 months <br />). At those times the total dose to the cable was between about 25 and 33 MR. The cable had not been thermally aged priot to the LOCA exposure. Jhe jacket material is an irradiation cross-linked polyethy: and the cable has been tested to 200 MR with thermal ag..g included. Cross-linked pclyethylene in general is conu 1ered a very good material for radiation resistance and would ne expected to retain good elongation properties after only 30 MR exposure. Additionally, if the major stress causing the crack were the severe transient temperature exposure, we would have expected to see the effects of the crack earlier than 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />, since at 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br /> a long term steady-state exposure at 250*F had been going on for 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />, and this exposure remained constant to the 53 hour6.134259e-4 days <br />0.0147 hours <br />8.763227e-5 weeks <br />2.01665e-5 months <br /> point. Thus, it is very possible that the crack did not appear until later in the test when higher radiation doses were attained. However, we cannot be sure that this was indeed the case. We should also note that signal losses similar to those we observed have been observed in industry tests of radiation monitors, although not in the final "qualified" configuration of the monitors. The signal losses were usually attributed to connection problems.
Finally, we note that the errors we observed early in the test were much worse than those reported in industy tests.
If we assumed that the crack was in fact what allowed moisture penetration into the connector, then the following questions would be relevant:
i 1) The shorter lengths or cables typically used in qualification testing can tend to restrict the amount of 49 i
7.0 References
- 1. "Instrumentation for Light-Water-Cooled Nuclear Power Plants to Assess Plant and Environs Conditions During and Following an Accident," Regulatory Guide 1.97 (Washington, D.C., U.S. Nuclear Regulatory Commission, Revision 2, Dec. 1980).
- 2. "Clarification (Washington, of TMI Action Plan Requirements,"
D.C.: U.S. Nuclear Regulatory Commission, Nov. 1980). l
- 3. Oliver D. T. Lynch Jr. to William S. Farmer, "Modification of FIN-A1051 to Include Instrumentation Required I
by NUREG-0737, II.F.1(3)" (Washington, D.C.: US Nuclear I f Regulatory Commission, Nov 27, 1984). Memorandum, i i
I 4. Dennis E. Hadlock and Ronald L. Kathren, "Thermal Limitations of Detectors in High-Level Radiation Fields,"
Health Physics, 43:1, July 1982. '
- 5. Elizabeth H. Richards and John A. Lewin, "Test Plan -
High-Range Radiation Monitor" (unpublished)
July 12, 1985.)
(Albuquerque: Sandia National Laboratories, )
{
- r. "Specifications for Backfit Digital Radiation Monitoring Systems" (San Diego: GA Technologies E-162-1147, 1983).
- 7. "High-Range Gamma Radiation Monitoring System --
Operation and Maintenance Manual" (San Diego: GA Technologies E-115-870 Rev. 2, November 1981).
- 8. "High-Range Containment and Steam Line Radiation Monitors --
Equipment Manual" (San Diego: GA Technologies E-115-1049, November 1981).
9.
(Clearwater, "Equipment Qualification Data Bank Quarterly Report" FL: Electric Power Research Institute, April 1984).
- 10. "IEEE Standard 2.or Qualifying Class 1E Equipment for Nuclear Poewr Generating Stations," IEEE 323-1974 (The Institute of Electrical and Electronic Engineers, 1974).
- 11. William H. Buckalew and Frank V. Thome, "Radiation capabilities of the Sandia High Intensity Adjustable Cobalt Array," NUREG/CR-2582, SAND 81-2655 (Albuquerque: Sandia National Laboratories, March 1982).
- 12. J. A. Gieske, et al., "Radionuclide Release Under Specific LWR Accident Conditions," BMI-2104, Vol. 1-6 (Colombus, OH: Battelle Columbus Laboratories, 1984).
48 1
r I
- DISIRIH EION:
U.S. Government Printing Office Atmic Dwrgy of Canada, Ltd.
Receiving Branch (Attn: NRC Stock) 1600 Dorchester Boulevard West 8610 Cherry Iane Mont'al, Quebec H3H 1P9 Laurel, MD 20707 CANADA 330 copies for RV Attn: S. Nish
~
Araldo Inplanti At mic Energy Research Centro Sperimentale del Boschetto Establishment Corso F.!!. Perrone, 118 Building 47, Division M.D.D.
16161 Genova Harwell, Oxfordshire ITALY OX11 ORA, -
. Attn: C. Bozzolo DCIAND Attn: S. G. Burnay Ansaldo Inpluti Via Gabriele D'Annunzio, 113 Bhabha Atomic Research Centre 16121 Genova Health Physics Division ITALY BARC Attn: S. Grifoni Batay-85 INDIA ASEA-AKM Attn: S. K. Mehta Department KRD Box 53 British Nuclear Fuels Ltd.
4 S-721 04 Springfields Works Vasteras Salwick, Preston SWEDEN Iancs Attn: A. Kjellberg DCLAND Attn: W. G. Cunliff, Bldg 334 ASEA-AKM Department 'IQD Brown Boveri Reaktor GMBH Box 53 Postfach 5143 S-721 14 D-6800 Mannheim 1 Vasteras WEST GERMANY SWEDDI Attn: R. Schemmel Attn: T. Granberg Bundesanstalt fur Materialprufung ASEA KABEL AB Unter den Eichen 87 P.O. Box 42 108 D-1000 Berlin 45 S-126 12 WEST GERMANY Stockholm Attn: K. Wundrich SWEDDI Attn: B. Dellby CEA/ chi-FAR Departement de Surete Nucleaire Atomic Energy of Canada, Ltd. Service d' Analyse Fonctionnelle Instrument and Control Branch B.P. 6 Chalk River Nuclear Laboratories 92260 Fontanay-aux-Roses Chalk River, Ontario KQ7 1JO FPRICE CANADA Attn: M. Le Meur Attn: E. C. Davey J. Henry D-1
h 1
moisture getting to the colinector by permeation through cable jackets. How'much moisture would have migrated to the connector if a representative length of cable had been l connected to the detector?
- 2) Will jacket integrity be maintained in actual l l
installations which may inc]ude different cable products, additional environmental exposures, and various aspects unique to the particular installation? "
l The answers to these questioris are beyond the scope of this study, but perhaps they should be investigated in future studies. .
l l
i i
l
> 1 l
l
]
,t r DISIRIBUI' ION:
U.S. Government Printing Office Atmic Energy of Canada, Ltd.
Receiving Branch (Attn: NRC Stock) 1600 Dorchester Boulevard West 8610 Cherry Iane Montreal, Quebec H3H 1P9 Iaurel, MD 20707 u CANADA 330 copies for RV Attn: 'S. Nish Ansaldo Inplanti Atcmic Energy Researth Centro Sperinentale del Boschetto Establishment Corso F.M. Perrone, 118 Building 47, Division M.D.D.
16161 Genova Harwell, Oxfordshire ITALY OX11 ORA, +
Attn: C. Bozzolo DIGIAND Attn: S. G. Burnay Ansaldo Inplanti Via Gabriele D'Annunzio, 113 Bhabha Atcmic Research Centre 16121 Genova Health Ihysics Division ITALY BARC Attn: S. Grifoni Bombay-85 INDIA ASEA-A'ICH Attn: S. K. Mehta
. Department KRD Box 53 British Nuclear Fuels Ltd.
, S-721 04 Springfields Works N
Vasteras Salwick, Preston SWEDDI Iancs Attn: A. Kjellberg DIGIAND ,
Attn: W. G. Cunliff, Bldg 334 ASEA-A'ICH Department 'IQD BIw n Boveri Reaktor GMBH Box 53 Postfach 5143 S-721 04 D-6800 Mannheim 1 Vasteras WEST GERMANY SWEDDT Attn: R. Schemmel Attn: T. Granberg Bundesanstalt fur Materialprufung ASEA KABEL AB Unter den Eichen 87 P.O. Box 42 108 D-1000 Berlin 45 S-126 12 WEST GH M lY
, Stockholm Attn: K. Wundrich SWEDDI Attn: B. Dellby CEA/CDi-FAR Departemant de Surete Nucleaire Atomic Energy of Canada, Ltd. Service d' Analyse Fonctionnelle Instrument aM Control Branch B.P. 6 Chalk River Nuclear Iaboratories 92260 Fontenay-aux-Roses Chalk River, Ontario K0J IJO FPRICE CANADA Attn: M. In Meur Attn: E. C. Davey J. Henry 1
1 l
l D-1
CERN, Electricite de France Iaboratorie 1 Direction des Etudes et Recherches 01-1211 Geneve 23 1, Avenue du General de Gaulle SWuztalAND 92141 CIAMART T DEX Atta: H. Schonbacher FRANCE Attn: J. Roubault ;
Canada Wire and Cable Limited L. Deschanps l Power & Control Products Division .
l 22 Ccmnercial Road Electricite de France
'Ibronto, Ontario Dir~ tion des Etudes et Recherches CANADA M4G 1Z4 L., h rdieres Attn: Z. S. Paniri Beite Postale n' 1
- 77250 FCRUr SUR IORDIG Centro Elettrotecnico FRANCE Sperimentale Italiano Attn: Eh. Roussarie )
Research and Develognent V. Deglon '
Via Rubattino 54 J. Ribot 20134 Milan, '
ITALY Energia Nucleare e delle
, Attn: Carlo Masa.tti Energie Alternative
! CKE Casaccia h i e riat a l'Energie Atomique 1-0060 Rcme ORIS/IABRA ITALY i BP N' 21 Attn: A. Cabrini 91190 Gif-Sur-Yvette FRANCE EURA'KM Attn: G. Gaussens Ccanission of European Camunities J. Chenion C.E.C. J.R.C.
F. Carlin 21020 Ispra (Varese)
ITALY Ccmd m riat a l'Energie Atomique Attn: G. Mancini CDI Cadarche DRE/ SIRE BP N' 1 FRAMA'ICME 13115 Saint Paul Inz Durance Tour Fiat - Cedex 16 j FRANCE 92084 Paris Ia Defense l Attn: J. Campan FRANCE Attn: G. Chauvin Conductores Monterrey, S. A. E. Raimondo i P.O. Box 2039 Monterrey, N. L. Furukawa Electric Co., Ltd.
MEXICO Hiratsuka Wire Works Attn: P. G. Murga 1-9 Higashi Yawata - 5 Chome Hiratsuka, Kmagawa Pref Electricite de France JAPAN 254 (S.E.P.T.E.N.) Attn: E. Oda 12, 14 Ave. Dubrieroz 69628 Villeurbanlie Gesellschaft fur Reaktorsicherheit Paris, FRANCE (GRS) rbH Attn: H. Herouard Glockengasse 2 M. Hermnt D-5000 Koln 1 WEST GER4ANY Attn: Library D-2
l l
Health & Safety Executive Japan'Atcmic Energy Research l Baynards House Instituta l 1 Chepstow P1, Westbourne Grove Tokai-Mura Iondon W2 4TF Naka-Gun ENGIAND Ibaraki-Fen Attn: W. W. Ascroft-Hutton 319-11, JAPAN Attn: Y. Foizumi ITT Cannon Electric Canada Fotar Cannon Court- Japan Atcnic Energy Research 1 Whitby, Ontario LIN SV8 Institute ,
l CANADA Osaka Iaboratory for Radiation Attn: B. D. Vallillee Chemistry
. 25-1 Mii-Minami machi, Imatran Voima Oy Neyagawa-shi Electrotechn. Department Osaka 572 P.O. Box 138 JAPAN '
SF-00101 Helsinki 10 Attn: Y. Nakase FINLAND Attn: B. Regnell Kalle Niederlassung der Hcechst AG K. Koskinen Postfach 3540 6200 Wiesbaden 1, Institute of Radiation Protection WEST GEENANY Department of Reactor Safety Biebrich P.O. Box 268 Attn: Dr. H. Wilski
- 00101 Helsinki 10 .
1 FINLAND Kraftwerk Union AG Attn: L. Reiman Department R361 Hanmerbacherstrasse 12 + 14 ISMES S.p.A. D-8524 Erlangen Viale G. W re, 29 WEST GFRMANY 24100 BERGAMO, Italy Attn: I. Terry Attn: A. Castoldi
) M. Salvetti Kraftwerk Union AG Section R541 Japan Atcmic Energy Research Postfach: 1240 Institute D-8757 Karlstein Takasaki Radiation Chemistry WEST GERMANY Pesearch Establishment Attn: W. Siegler Watanuki-machi Takasaki, Gunma-ken Kraftwerk Union AG JAPAN Hammerbacherstrasse 12 + 14 Attn: N. Tamura Pestfach: 3220 K. Yoshida D-8520 Etlcngen T. Seguchi WEST GERMANY Attn: W. Morell Motor Colu:rbus Parkstrasse 27 CH-5401 Baden SWITZERINID Attn: H. Fuchs D-3 1
w
. .a .
l l
National Nuclear Corporation Rheinisch-Westfallscher Canbridge Road, Nhetstone Technischer Uberwachunge-Verein j Iaicester II8 3IH e.V.
DUIAND s Postfach 10 32 61 l Attn: A. D. Hayward D-4300 Essen 1 J. V. Tindale WESP GERMANY Attn: R. Sartori NOK AG Baden Beznau Nuclear Power Plant Sydkraft G-5312 Doettingen Southern Sweden Power Supply ,;
Sh m rxi.AND 21701 Malmo Attn: O. Tatti SIEDEN Attn: 0. Grondalen -
Norsk Kabelfabrik 3000 Dtauuen Technical University Munich NORWAY Institut for Radiochemie Attn: C. T. Jacobsen D-8046 Garching WESP GERMANY Nuclear Power Engineeriry 7bst Attn: Dr. H. Heusinger Center 6-2, Ibranctnon, 3-Crrne UKAEA Minato-ku, #2 Akiyana Bldg. Materials Develegnent Division Tokyo 105 Building 47 JAPAN AERE Harwell Attn: K. Takumi OXON OX11 ORA DUIAND Ontario Hydro Attn: D. C. Willips 700 University Avenue Toronto, Ontario MSG 1X6 United Kingdan Atcraic Energy CANAIR Authority Attn: R. Wong Safety & Reliability Directorate B. 141kreti '
Wigshaw Iane Culcheth (
Oy Strcanberg Ab Warrington WA3 4NE Helsinki Works DGIAND Box 118 Attn: M. A. H. G. Alderson FI-00101. Helsinki 10 FINIAND Waseda University Attn: P. Paloniemi Department of Electrical Engineering Radiation Center of 3-4-1 Ohkubo, Shinjuku-ku Osaka Prefecture Tokyo 160 Radiation Application- JAPAN l mysics Division Attn: Y. Ohki Shinke-G o, Sakai Osaka, 593, JAPAN Sorrento Electronics Attn: S. Okancto General Atanic Cottpany P.O. Box 81608 Rappinl San Diego, CA 92138 DEA-PEC Attn: T. Moshenrose i Via Arcov&Jgio 56/23 l Bologna ITALY l
Attn: Irg. Ruggero D-4
Toledo Edison Ocupany NFED-EQ Mail Stop 3040 5501 North S.R. 2 Oak Harbor, OH 43449 Attn: Peter W. Jacobsen Office of Scientific and Technical Information ~
P.O. Box 62 Oak Ridge, 'IN 37831 (8)
I 1200 J. P. VanDevender l 1234 G. J. Iockwood -
1512 J. A. Iswin 1800 R. L. Schwoebel 1810 R. G. Fepler l 1811 R. L. Clough 1812 J. M. Zeigler 1812 K. T. Gillen 1813 J. G. Curro 2126 J. E. Gover 2126 O. M. Stuetzer 2514 L. L. Bonzon 3141 S. A. Iandenberger (5) 3 3151 W. L. Garner 6200 V. L. Dugan 6222 E. H. Richards (7) 6258 P. M. Drozda 6300 R. W. Lynch 6400 D. J. McCloskey 6420 J. V. Walker 6429 K. D. Bergeron l 6440 D. A. Dahlgren 6442 W. A. Von Riesemann l 6447 W. H. Buckalew 6447 L.' D. Bustard 6447 M. J. Jacobus (10)
,. 6447 D. L. Berry l 6447 M. P. Bohn l 6511 F. V. 'Ihcme l 6512 F. J.Wyant l 8524 P. W. Dean l
D-5
- i. .-. . ..
~
a
- s. s u s vema a uufm co ... . .t rou w..u m.~ rioc - v., u,. ., ,
j an,c ,,oa= =
%"'E BIBLIOGRAPHIC DATA SHEET NUREG/CR-4728 SAND 86-1938 us sfavef.oaso 7 e nvene 3 TITL4 AMo lveisitt JLlaveSLA%s EQUIPMENT QUALIFICATION RESEARCH TEST OF A h HIGH-RANGE RADIATION MONITOR 4 o ATE RE#o#T COMP (ET(o
~
uo*e T ** tlan i Avi,soam January 1988 E. H. Richards, M. J. Jacobus .oAri n , cat.uvio woa'- 'a^a P. M. Drozda and J. A. Lewin l February 1988
- 7. Pf a*omesessG oar,Ames2 A7 igg NAest Ano esAstewo Aoomass riaews,to c.ari e PaoJtet/T ASKmont v'eil Nueses a Sandia National Laboratories ,,,,,,,,,A,, ,,,,,
Albuquerque, NM 87185-5800 A1051 10 SPO8eloni46 3RGami2 Af tom NAWt Aho WalklNG AooR{$$ tter&de gg Capet 113 TVPtofStPoAT Division of Engineering Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission o 'imoo cov e"o ua<-~ ~~
Washington, DC 20555 12 SUPPLgwEs eT AA v peoTE S 13 A487 m A GT d200 eeres ., 'essi A high-range radiation datector was tested in a simultaneous steam and radiation environment simulating e postulated loss-of-coolant accident (LOCA) to assess possible synergistic effects:that may be important to its performance in an accident. The detector, manufactured by General Atomic, was simultaneously subjected to a simulated accident environ-ment including 171 C (3400F) steam at 410 kPa gage (60 psig) and'4 Mrad /hr gamma radiation while its performance was monitored. Test results showed that the detector successfully operated at the high dose rate and temperature, without evidence of synergisms. However, I
at reduced radiation levels in a saturated steam environment, the detector signal at the readout module deteriorated in accuracy or ceased altogether. The cause of these anomalies is attributed to leakage currents and/or possible galvanic action in the coaxial connections and/or cables.
l it A A At Tv 54 DocVM6 4T A8eakv$il - e EE'Fromol'otsca PToas Un1imited
,. u cu.... c6 A n,,,e AT,oN l <r. n
. os=t.,iine,i i eio tia-s Unclassified
,r..,,,
Unclassified
,, w n. o, , Aon 64
, . . ,c e
=
- u s GovtaNutNi eamTwo omet two-stwamoA4
. .