ML20128J212
| ML20128J212 | |
| Person / Time | |
|---|---|
| Issue date: | 10/01/1996 |
| From: | Lohaus P NRC OFFICE OF STATE PROGRAMS (OSP) |
| To: | Ortciger T ILLINOIS, STATE OF |
| References | |
| NUDOCS 9610100222 | |
| Download: ML20128J212 (3) | |
Text
{{#Wiki_filter:l 1 OCT - 1 1996 i ! Mr. Thomas W. Ortciger, Director ! Illinois Department of Nuclear Safety l 1035 Outer Park Drive l Springfield, IL 62704
Dear Mr. Ortciger:
This is in response to your August 1,1996 letter on reciprocal recognition of International Standards. In your letter, you provided examples of competent foreign authority documents for approval / certification of sealed sources and discussed use of these documents. You indicated that you would appreciate receiving any comments on "these issues." NRC staff has reviewed your letter and documents and a copy , of the staff's comments is enclosed. ' We appreciate the examples you and your staff provided on this issue. l Sincerely, ' OriginalSigned By: PAUL H.LOHAUS Paul H. Lohaus, Deputy Director Office of State Programs
Enclosure:
As stated l Distribution: ! DIR RF (6S-205) RBangart PLohaus SDroggitis LBolling , DCool, IMNS ' Illinois File DCD (SP07) . I PDR (YES f N0_) 1 DOCUMENT NAME: G:\TSK6S205. LAB *See previous concurrence.
** Memorandum concurrence.
To secolve e oopy af this alocument, indicate in the bos: 'C" = Co f without ettechment/ enclosure 'E' = Copy with ettechment/ enclosure *
'N' = No copy [' ,t[g 4 0FFICE OSP lE OSP{hl41 'T IMNS:D C OSP:( byC NAME LBolling:gd:kk PLohaus DCool RBangaR W DATE 09/26/96* 09/26/96* 09/11/96* S9-/pl/96 4 g 'g OSP CODE: SP-AG-8 l0 g% h bbhbb g 9610100222 961001 PDR STPRG ESGIL PDR
. , e i Thomas W. Ortciger We appreciate the examples you'and your staff provided on this issue.
\
Sincerely,
\;
Paul H. Lohaus, Deputy Director Office of State Programs
Enclosure:
As stated \
\
\
\
\
\
\,
\
Distribution: \ DIR RF (6S-205) RBangart {; PLohaus ( SDroggitis I LBolling i DCool, IMNS
\
Illinois File DCD (SP03) i PDR (YES / NO ) i
\
DOCUMENT NAME: G:\TSK65205. LAB
- r. . . . , ., m. 4.. . i.ac.t. in ts. i,..: c . copy without .it. chm.ntl.ncio.ur. "E" . Copy with .tt. chm.nt/.ncio.ur. -
r . wo..py '
.et.ch.4 _ ,
0FFICE -f%0SP lE OSP:W] C IMNS:D* C OSP:D lC NAME (;5#1Hng:gd:kk PLohifufFi DCool RBangart DATE 09/J.6/96 09/26/96 09/11/96 09/ /96 OSP CODE: SP-AG-8
W *% p 4 UNITED STATES g j t NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 2006H1001
's . . . . . ,o! October 1, 1996 Mr. Thomas W. Ortciger, Director Illinois Department of Nuclear Safety 1035 Outer Park Drive Springfield, IL 62704
Dear Mr. Ortciger:
This is in respense to your August 1,1996 letter on reciprocal recognition of International Standards. In your letter, you provided examples of competent foreign authority documents for approval / certification of sealed sources and discussed use of these i. documents. You indicated that you would appreciate receiving any comments on i
"these issues." NRC staff has reviewed your letter and documents and a copy of the staff's comments is enclosed.
We appreciate the examples you and your staff provided on this issue. erely, M>W Paul H. Lohaus, Deputy Director Office of State Programs i
Enclosure:
As stated i s
. i eg
[pu %" UNITED STATES NUCLEAR REGULATORY COMMISSION
'f WASHINGTON, D.C. 30666-0001 d September 12, 1996 \ . . . . . ,o 1
MEMORANDUM TO: Paul H. L,e/ny, Deputy Director Office of State Prrrrams FROM: Donald A. Cool, Directo 1 / Division of industrial an Medical Nuclear Safety { Office of Nuclear Material Safety and Safeguards
SUBJECT:
RECOGNITION OF INTERNATIONAL STANDARDS. My staff has reviewed the August 1,1996, let'er from the State of Illinois Department of Nuclear Safety (IDNS) and has the following comments. in some cases, an applicant for evaluation and registratkn of a sealed source or device may indicate that a product has been tested in accordance with, and meets the requirements of, an internstional or foreign standard. However, in order for NRC to find this acceptable, the person j reviewing the sealed source or device application must first ensure that standard meets or exceeds any specific regulatory requirements (e.g., compliance with ANSI N432-1980 for radiography equipment). The reviewer must then review the requirements and acceptance criteria of the international or foreign standard based on the normal and likely accident conditions associated with use, handling, storage, and transport of the product to determine if the standard is acceptable. The reviewer may compare the foreign or international standard with applicable U.S. standards in determining the acceptance of the standard. This may include professionaljudgement on the part of the reviewer. IDNS provided the following specific examples of documentation that were provided to support source and device evaluations. My ,taff provides the following comments concerning each document.
- 1. German Special Form Certificate from Bundesanstalt Fur Materialforschung Und-Prufung (BAM).
Typically, NRC requires sources to meet the requirements of ANSI N542, "Seded Radioactive Sources, Classification." Special form testing alone is not typicall/ acceptable since special form testing is designed for accident conditions associated with transport of the source. The referenced standard should be compared with ANSI N542. The reviewer must determine, for any tests not meeting the test criteria listed in ANSI N542, whether additional testing is necessary. Final determination of the acceptance of the standard should be based on the normal and likely accident conditions associated with use, handling, storage, and transport of the source. 4 py.looo ? *57p
- t. .L.
2
- 2. German Sealed Source Registry from the Sentate Administration for Science and Research, Berlin.
NRC does not recognize sealed source or device approval issued by foreign regulatory agencies. If such an approvalis submitted in support of an application,it should be treated as any other information included in the application.
- 3. Leak test certificate from the Physical Technical Federal Institution, Berlin.
This document does not appear to be a leak test certificate. It appears to be a certification provided by the Physical-Technical Federal Institution that indicates the source model has been tested in accordance with international Organization of Standardization (ISO) 2919 and DIN 25426 Part 1, and, in accordance with these standards, has achieved a classification of ISO / C 66646. The certification states that the testing was performed by the manufacturer on eight prototype sources. However, no details of the specific tests performed are provided. It may be necessary to obtain details of the specific tests performed by the manufacturer to ensure the manufacturer did not misinterpret the testing or ecceptance criteria requirements of the standard. If the reviewer has confidence that the testing was , performed in accordance with the standard, it may not be necessary to obtain any additional information since ISO 2919 meets the specifications of ANSI N542. With respect to leak test certificates, NRC does not typically evaluate the person performing the tests and has accepted leak test certificates from foreign authorities. l , Acceptance is based on the fact that the document certifies that a particular source has I been tested. However, NRC may ask for additional information on leak testing procedures if there is a particular concern, such as the void volume within the source capsule. Void volume requirements are discussed further in response to item 4.
- 4. Report EUR 8053 EN regarding IAEA Special Form Testing.
As stated in response to item 1, NRC typically will not accept special form testing alone since the special form testing is designed for accident conditions associated with transport of the source. ' With respect to procedures for leak testing prototype cources to ensure acceptance of the source design, NRC typically requires the procedures to meet the guidance included in ANSI N542. However, if the applicant provides informatiori that shows the testing procedures are acceptable, the reviewer may accept the testing as an adequate substitute for the testing included in ANSI N542. The above referenced document appears to be a study that evaluates an intemationally recognized stcadard. It provides information concerning why attemate testing should be considered as an acceptable substitute. NRC may accept this type of document if it provides an adequate argument for why the testing is acceptable. 1
t, b 3 IDNS indicates that this document provides "some very compelling information regarding the quality of the sources." Specifically, IDNS notes the information concerning leak testing of source having less than 0.1 ml (0.006 in') of void volume within the capsule. NRC agrees that the study provides a compelling argument for the acceptance of leak testing of sources with a void volume less than 0.1 ml(0.006 in'). However, as stated on page 5, Appendix 3, of the study, the goal of the study "was to establish that tests using reduced void volumes are valid under specified conditions." Therefore, the use of the leak testing procedures would only be valid in limited situations, such as those where source design and the leak testing procedures are in accordance with the information provided in the study.
- 5. Quality Assurance: Regulatory Guide 6.9
- Establishing Quality Assurance Programs for the Manufacture and Distribution of Sealed Sources and Devices Containing Byproduct Material
- and ISO 9000.
NRC has routinely accepted the submiscion of, and commitment to, a quality assurance (QA) program in lieu of the quality control (OC) information cited in 10 CFR 32.210. Typically, details of the QA program are submitted in the form of a QA manual. To assist persons in the development and implementation of an acceptable QA program, NRC issued regulatory guide 6.9. The details of an applicant's QA program are evaluated against the guidance provided in regulatory guide 6.9. Development of regulatory guide 6.9 was based on existing QA stcndards, including the ISO 9000 series QA guides. The intent of the regulatory guide was to be less restrictive than the ISO 9000 series guides to ensure that persons implementing an ISO 9000 program would meet the guidance. Therefore, persons implementing the appropriate ISO 9000 series program should meet the guidance provided in regulatory guide 6.9. Historically, NRC has not accepted a statement that a program meetc regulatory guide 6.9 or ISO 9000 as being acceptable nor has NRC accepted, a face value, E third party certification. NRC has requested the submission of the program, usually in the form of a manual, and has evaluated the program against the guidance provided in regulatory guide 6.9. IDNS indicates that it believes NRC requests " extensive additional information" from persons implementing ISO 9000 series programs. As stated, NRC typically request a copy of the QA manual for comparison with regulatory guide 6.9. Typically, the only additional information that NRC may request from any persons that submit a QA program is a commitment to perform a leak test and an operational test on each product prior to distribution. If you have any questions conceming the above information, please contact John Lubinski of my staff at (301) 415-7868. i
_ kky (JL 5 4 l Mh EXECUTIVE TASK MANAGEMENT SYSTEM
<<< PRINT SCREEN UPDATE FORM >>>
TASK # - 6S205
, DATE- 08/13/96 MAIL CTRL. - 1996 TASK STARTED - 08/09/96 TASK DUE -
/ / TASK COMPLETED -
//
TASK DESCRIPTION - 8/1/96 LTR RE RECIPROCAL RECOGNIATION OF INTERNATIONAL l STANDARDS REQUESTING OFF. - IL REQUESTER
- T. ORTICIGER WITS - 0 FYP - N I PROG.- LAB PERSON - STAFF LEAD - LAB PROG. AREA -
PROJECT STATUS - OSP DUE DATE: PLANNED ACC. -N 1 LEVEL CODE - 1
/
l
L' . :gm a STATE OF ILLINOIS e DEPARTMEN,TOF NUCLEAR SAFETY 1035; OUTER PARK DRIVE SPRINGFIELD, ILLINOIS 62704 2 1 Jim Edgar 217 5[990f+. Thomas W. Ortciger Governor 217-782-6133 Director mog (TDD) I August 1,1996 U.S. Nuclear Regulatory Commission Document Control Desk t*l-37 Washington, DC 20555 Attn: Paul Lohaus, Deputy Director RE: Reciprocal Recognition of International Standards
Dear Mr. Lohaus:
Thank you for your comments on our letter of February 27,1996. We look forward to continued Agreement State involvement in the evolution of the Sealed Source & Device Evaluation Program and the Sealed Source Standard Review Plan. Regarding your request for examples of other competent authority documents for approval / certification of sealed sources, I have enclosed the following attachments for your review: (1) German Special Form Certificate from Bundesanstalt Fur Materialforschung Und-Prufung (BAM); (2) A copy of a German Sealed Source Registry from the Senate Administration for Science and Research, Berlin; (3) A leak , testing certificate from the Physical Technical Federal Institution, Berlin; and, finally, (4) a copy of Report EUR 8053 EN regarding IAEA Special Form Testing. During any review process, we require our licensees to adhere to the standards of the Department, the U.S. NRC and the U.S. DOT, as suggested in your letter. However, the above documents contain some very compelling information regarding the quality of the sources in question. One of the more controverdrJ compatibility issues we have encountered involved the void spaces required for bubble testing. Our current standards require a void space greater than 0.1 ml. Appare:itly, the international community allows a much smaller void space, depend <ng on the test medium used for these tests, as is evident in Item 4 above, Appendix 3, pages 5 - 11. l
-9408040229 PDR 960801 //7 y !
STPRC ESGIL t PDR , bl 1 nm
l O tr U.S. Nuclear Regulatory Commission I Fage 2 August 1,1996
- Another ' :a of concern is quality assurance (QA). With the advent of NRC Regulatory Guide 6.9, it appears that the NRC no longer accepts International l Organization for Standardization (ISO) 9001 approved programs without requesting l extensive additional information. At one time, ISO 9001 approved programs were readily accepted. The Department is in the process of trying to amend our regulations / guidance in this area to accommodate both the ISO standards and the key elements of NRC Guide 6.9. Effectively, we are prepared to accept ISO programs during the licensing process in combination with on-site inspections performed by the Department or other responsible third party entity to ensure that the details of ;
Regulatory Guide 6.9 exist within the framework ofISO, as we believe is the case. l The Department would appreciate any comments on these issues. Ideally, we would like to see either the U.S. or international standards revised and some type of reciprocal recognition adopted for registries / certificates / programs approved by j competent foreign authorities. If you have any questions on this matter, do not hesitate to contact Gibb Vinson of our Licensing Staff for the Division of Radioactive Materials at (217) 785-9947. l l Sincerely 3 p> \ 1 k Thomas W. Ortcig Director i TWO:cgv Attachments cc: Jim Lynch, NRC Region III State Agreement Program Officer l i l l
l I' BUNDESANSTALT FOR MATERIALFORSCHUNG UND -PROFUNG o ,
* (FEDERAL AGENCY FOR MATERIALS RESEARCH AND TESTING)
(BAM)
)
APPROVAL CERTIFICATE i for the design of a special-form radioactive material D/0054/S-85 L 12ral basis: This approval is given in conformity with the requirements for "special-form radioactive material" as specified in Safety Series No. 6 Regulations for the Safe Transpon of Radioactive Materials,1985 Edition (As Amended 1990) of the International Atomic Energy Agency (IAEA) l I and relevant national and international regulations such as the Ordinance on the Domestic and Cross border Carriage of Dangerous Goods by Road (GGVS) in the version of the official announcement of 26 Nov.1993, as amended by the Railway Reorganization Act of 27 Dec.1993, Annex A, marginal number 3731; the European Agreement Concerning the International Carriage of Dangerous Goods by Road (ADR Agreement) of 30 Sept.1957, as amended by the 12th ADR amendmg regulation of 2 Dec. 1994, Annex A, marginal number 3731; the Ordinance on the Domestic and Cross-bonier Carriage of Dangerous Goods by Railways (GGVE) of 22 July 1985, as' amended by the Railway Reorgaruzation Act of 27 Dec.1993, Annex, margmal number 1731; the Ordinance on the laternational Carnage of Dagerous Goods by Railways (PJD rules)'- Annex
! to Appendix B of the Convention on International Railway Transport (COTIF Convennon) of 9 May 1980, as amended by the 5th RID amending regulation of 8 March 1995. Annex, marginal number 1731; This Approval Ceruficate comprises rive pages and the assembly drawmg and shall be copied only m unabridged form.
Pubicatsons of Approval Cemfrates, even af relating to exuacts only, any reference to testing for advemsement purposes and the use of the content of teports require in each case the approval in writag of BAM whicts may be revoked at any time.
! I
, 1 Page 2 of the Approval Certificate No. u/0054/S-85 ;
'- - the general authorization of the air transport of dangerous goods including weapons by airlines of
- 15 June 1988, in connection with the ICAO Technical Instructions for the Safe Transport of Dangerous Goods by Air and the IATA Dangerous Goods Regulations (currently 36th issue, valid as of ist Jan.1995) l l
2 Apolicant. manufacturer of the radiation sourc3s and holder of this acoroval 1 BEBIG Isotopentechmk und Umweltdiagnostik GmbH ) 1- l l Robert Rossle-Sti.10 i D-13125 Berlin-Buch 1 l
- 3. Desienation of desian l Doubly encapsulated radioactive material (radiation source) of type Cs7.PO2 with l a maximum of 18.5 GBq Cs 137 bour.d in ceramics.
l 4 m Drawines: l L a) Assembly drawing "Cs-137 source type Cs7.PO2" 1.203.03.02-00:00(4) of l 17 March 1995; I b) Drawing of capsule 1.203.03.02-00:01(4) of 17 March /5 July 1995; I c) Drawing of window 1.203.03.01-00:02(4) of 3 March /9 May 1995; d) Assembly drawing " Primary capsule Cs7.K01" 1.203.03.01-01:00(4) of 7 ; March /24 July 1995; e) Drawing of external body of the primary capsule 1.203.03.01-01:01(4) of 7 , March /9 May 1995; f) Drawing of external cap of the primary capsule 1.203.03.01-01:02(4) of 7 March /9 May 1995. L Submitted desien umnles:
- Two test saraples, filled with inactive ceramics 3mm x 3mm, weight: about 2.21 g each.
! L Dggriotion of the design: The doubly encapsulated radioactive material (radiation source) consists of a capsule which is scal-welded using tungsten inert arc welding (TIG), is made up of ' 1
Page 3 of the Approval Certificate No. D/0054/S-85 stainless austenitic steel (special steel) and features a 0.4mm-thick window; it also contains the TIG-scal-welded primary capsule made of the same special steel and the radiation source made of ceramics. t l ' The external body is a cylindrical part with a 6.4mm diameter and a 7mm shaft to hold the radiation source. l The radiation soune has a total length of 15.9 mm. The shaft has been fitted with an M4 thread which comes next to a short exposed part. Except for the differing shafts, the two designs, Cs7.P02 / formerly: design HB / Approval Certificate No. D/0054/S-85 and Cs7.P01/ formerly: design H / Approval Certificate No. D/0053/S-85 are identical. The primary capsule is a cylindrical part (diameter: 4mm, height: 6mm) made of special steel to which the inserted cap is scal welded by means of TIG welding. The scurce contained in it is a cylindrical body (diameter: 3mm, height: 3mm) made of ceramic material which was impregnated with cesium-137 nitrate, dried and sintered. Because of its thermal treatment the ceramic material has a vitreous appearance and a glazed surface. L Descriotion and results of the te=t= nerformed on the desien samoles-The legal regulations which apply currently to the various common cr.rriers are based or %e IAEA Safety Series No. 6, edition 1985/1990 concerning the requirements for "special-form radioactive materials". They provide for the following design tests: para.607: Eall of a radiation source sample from a height of 9m on to an inflexible foundation. para.608: Impact of the flat base (diameter: 19mm) of a steel rod weighing 1.4 kg from a height of im on a radiation source sampic placed on a lead pad.
.. - - .. - - . - . ~ . - . - - . _ _.-. ~
~
Page 4 of the Approval Certificate No. D/0054/S-85 para.609: Bending test, only for particularly slender radiation sources, not applicable. para,610: Maintaining the radiation source sample at a temperature of 800*C ! for ten minutes, then slowly cooling it down; see temperature class l ! 6 as set out in ISO 2919 or DIN 25426-1: maintaining the radiation source sample at 600'C for one hour with subsequent quenching in l
\
water of ambient temperature, i.e. a more severe stress than J required by IAEA. para.612. Leaching test for non-encapsulated radioactive materials, not applicable. para.613: Leak tightness test following radiation source stresses as specified by the ISO Technical Report 4826-1979, which has been replaced in the meantime by the more advanced international standard ISO 9978-1992. In a first step, one radiation source sample was submitted to the falling test from 9m. then to the impact test in the lying position. The other radiation cource sample was submitted to the impact test, the threaded part being in the upright position and then to the test conditions satisfying ISO temperature class 6. Both samples ! remained intact. As a result of the impact the lengths were reduced from 15.9mm j to 15 7mm and the diameters from 6.4mm to 6.25mm. The samples were then opened up so that, after a thin copper plate was glued into the bores, the welding areas of the external capsules were directly exposed to helium, which was measured against a vacuum in a helium mass spectrometer. Both radiation source samples proved to be leakproof, the minimum detectable limit was 5x10* mbar x 1/s. EL Ouality assurance: In a letter dated 26 June 1995, the applicant and manufacturer of the radiation source submitted to BAM " Regulations on quality assurance measures and tests in the manufacture of the encapsulated special-form radioactive material of designs Cs7.PO1 and Cs7.PO2" that suggest consistent high quality of all radiation sources
..- --..- .- - . - . - - - . - - - . - - .. . . . _ - . _ _ - - . - - . ~ . - . . . -
i- i Page 5 of the Approval Certificate No. D/0054/S-85
, of this design to be manufactured at a later time. BAM reserves the right to verify at the expense of the applicant whether the radiation sources to be manufactured in ;
j future are in conformity with the approved design. 4 1 Aooroval of the desien ; ! On the strength of the documents submitted and the test results described.in paragraph 7, the radiation sources of design Cs7.PO2 made by BEBIG as specified
)
in paragraphs 3,4 and 6 comply with the relevant provisions set out in paragraph ! l
- 1. i This approval is valid for an initial period of five years; it shall be submitted ;
1 unsolicited to BAM for revision in good time before the expiry of this period. l l l 12205 Berlin,25 July 1995 j Unter den Fichen 87 l BUNDESANSTALT FOR MATERIALFORSCHUNG UND -PR6TUNG Specialized group 111.3 Laboratory 111.35 Safety of Transport and Storage Special Issues of Encapsulation : Vessels By order: Prof. Bernhard Droste (Dr,-Ing.) Dipl-Ing. Lothar Buhlemann Director Employee in charge: Dipl.-Phys. Helmut Kowalewsky Oberregierungsrat [ Stamp) Bundesa:ntalt & Material-forschung und prufung (BAM)
'N 74.12.95 16:23 BEBIG 150TCPENTEONIK . '. Attachment 2 i i I
Senatsverwaltung fur Wissenschaft und Forschung BERLIN ( - , III D als stellvertretende Strahlenschutzbeauftragte 1 l ekM.N74/eIsN ' BEBIG 36096E '9 *~ Isotopentechnik und s,II';e,D Umweltdiagnostik GmbH 1iAug.133b I Robert-R5eale-Str. 10 zyg;pu Kladky 13125 Berlin pj l \ g g% 30 32-
. ev**AAAe (80l@32.92 09 05-47 N: hbggg o9 05-22 ostum M. 8,1994 Betr. Erlaubnis zur Nutzung der unbefristeten Stellungnahme -
der PTB 6.22 - G S vom 25.07.1991
~ Als Eigentilmer der Stellungnahme 6.22 - G 5 der Physikalisch-Technischen Bundesanstalt (PTB) vom 25.07.1991 zu den Priif-fristen fiir Dichtheitspriifungen an der radioaktiven Strahlen-quelle mit der Typenbezeichnung EB gestatten wir Ihnen die Benutzung der o.g. Stellungnahme filr die von Ihnen nach glei-cher Technologie hergestel2 ten Produkte gleichen Typa.
Fur die Einhaltung aller technischen Parameter gem &B Typen-beschreibung sind Sie vorantwortlich. Mit freundlichen GruBen Im Auftrag l Hladky Vr.2_- .- go,echronen: Zahngen see an ce Lanonenauptkaane Benin,107se Ber+e VBahn (aneertsamm Montag, benstag Komanummer Osamshtut 8anauenzahl ' Bus 104.105 saa. 204 von 9 Die 12 Uhr. 53 100 #estbeck Barhn too 10010 Damerstag 99I9 260 800 gm Bertner Sann 143 200 00 %, ,em, von 16 bis 18 Uhr 0990 007 soo Lanoescana Benin 100 500 00 una emon vereinoerung 10 00i S20 Lancearentrashana f 0 C00 000
. . . - .- - -- . . . - . . . . - . _ - .. .~ -. . . . . . . . - . _
l Senate Administration for Science and Research BERLIN . III D as acting Radiation Safety Officer Senate Administrahon for Scaance and Research Brodtschneiderstraae 5. D-14057 Berhn oste 8/l1/94 BEBIG l Isotope Technology and Environmental diagnostics GmbH Roben-Rossle-Str.10 t 13125 Berlin i j l l i Es: Permission to use the unlimited opinion of the PTB 6.22 - G 5 of 7/25/91 l As owner of the opinion 6.22 - G 5 of the Physical-Technical Bundesanstalt (PTB) of 7/25/91 for the test period for the leak testing on radioactive sealed sources of the model type HB we permit l you to use the opinion for products of the same type which are manufa:tured by you with the l same technology. You are responsible for the keeping of all technical parameters according to the specifications. With friendly greetings On behalf . l Hladky
- j. . .
- 04.12.95 16:24 BE3!G ISOTOPENTECHNIK GO5 I Typenbeschreibung fur umschlossene Strahlenquellen Gammastrahlenquelle fur industrielle Anwendung 4
Typ H, Cs-137 1 t
- 1. Herstellerland Deutsche Demokratische Republik e
- 2. Lieferbetrieb Isocommerz GmbH 3 Binnen- und AuSenhandelsunternehmen fur radioaktive und stabile Isotope . !
j 1115 Berlin-Buch, Robert-R5ssle-Stra3e 10
- 3. Herstellerbetrieb i Akademie der Wissenschaften der DDR I Zentralinstitut fur Isotopen- und Strahlenforschung -
Bereich Strahlenquellen und Nuklearpharmaka 1115 Berlin-Buch, Robert-R5ssle-Stra3e 10 I i )
- 4. Bezeichnung der Quelle i
Gammastrahlenquelle fur indust **1elle Anwendung .
- 5. Radionuklid, AktivitXtsbereich l Cs-137 bis 18,5 GBq (500 mC1)
- 6. Typ der Quelle H
- 7. Kennzeichnung Typ und Seriennummer
- 8. Strahlenphysikalische charakteristika der Quelle Die Betastrahlung des Cs-137 wird durch die Kapsel voll-stEndig absorbiert.
Pur Strahlenschutzzwecke kann die Energiedesialeistung in Luft aus der spezifischen Gammastrahlungskonstanten berech-net werden. Die Energiedosialeistung in Luft betr8gt fur eine Aktivit8t von 3,.7 GBq ' (100 mC1): Abstand mGy / . (R/h) 27 (3,1 ) 0,1 m (0,34 )
"*0,?'m 3 0,3 (0,031) 1 m 2
, 04.12.95 16:25 EEBIG ISOTOFB5TECKNIK D277 H 2
- 9. Konstruktion der Quelle 9.1 Zeichnung alicemeiner Art i
verxhweht! 3, y rxhwaTit , S __ \
- Ip_fas l k i
f_ _/ H H H H f , I h _ r l i 't m __ _j _ _ _ . .< _ m ' . M
.I 4 b { ! {% .
) h W'dM
! e fj 2
Bakre Kap. ret
, g {
innereKapsel radioaktiver Teil ( 'gg, g 1 1
~
l 9.2 Physikalisch-chemische Eicenschaften des radioaktiven Teils ! In einem getemperten Keramik:ylinder 9 3 x 3 mm ist das Cs-137 in sehr abriebfester Form verteilt. Der radioaktive Stoff ist in gewdhnlicher Atmosphure oder gegennber Wasser inaktiv. Von 1 g des radioaktiven Stoffes ) werden innerhalb von 48 Stunden weniger als 0,01 % in 100 ml ! destilliertem Wasser bei 20 *C ohne Ruhren geldst oder dis- i pergiert. 9.3 Beschreibune der HUlle 9.3.1 Benennung oder Typ 9.3.2 Material Edelstahl der Sorte X8 CrN1Ti 18.10 nach TGL 7143 i Wanddicke allseitig der 4nneren Kapsel .: 0,4 mm j der RuBeren Ka/sel, radial : 1,1 m= 0,4 mm
, axial fansterseitig: i 1
9.3.3 Verschlu8 art verschweiBt
- 9. 3. 4 Anf orderungen be:uglich der HuSeren BErschaf fenheit
- Mittlere Rauheit der KapseloberflKehe: R2 20
- Schwe18naht gratfrei und homogen
- Kapseloberfluche, Form des Halters und Kennzeichnung ohne sichtbare M1ngel.
3
.. ._" T
. _ - . - - . ~ - . - . - . . ~ _ . . _
'04.12.95 16:25 BEBIG ISOTOFENTECW IK DOB
,y.
~ i Q .
,3 -
3 ' H 10, Arten und Methoden der Kont:911e der Quellen beim Herstellert Normen und Kriterien fur die Einschutzung der Resultate , 10.1 Dichtigkeit Innere Kapse:.: Immersionstest unter 51 eden in de'stilliertem Wasser *C. cher 2 mal 30 Minuten mit einer zwischenahkuhlung ~440 . Eluierbare Aktiyic1t: (185 Sq (5 nC1) XuSere Kapsel:
- Blasentest durch Eintauchen der,umschlossenen Strahlenquelle in Xthylenglykol und Senkung des Druckes Cher der Flussig-keit auf 15 kPa (150 mbar) Ober >1 Minute.
Innerhalb der Prufzeit durfen keine Luftblasen aus der um-schlossenen Quelle entweichen. l 10.2 Oberf1Rchenkontamination . ,*,r Innere und RuSere Kapsel 7L: Q. Feuchter Wischtest. Abwischbare AktivitEt <185 sq ' (5 nC1) . ..... Egk - Y ., 10.3 Prufung der Unversehrtheit .
,5 E Visuelle, stereemikroskopische Segutachtung bei ca. 10facher ~ t. ,..
Vergr5Serung. .V Einhaltung der Anforderungen an die Xu6ere
~'
Kriterien: Beschaffenheit entsprechend Punkt 9.3.4 o
- 11. Klassifizierung der Quelle gemE8 RGW-Standard 3839-82 und TGL 25294 ,
ISO /RGW/ C 6 6 6 4 6 , x "I,
^
- 12. Einstufung der Quelle als "stoff is bescnderer-Form" . .~.
entspricht v - ut .y s,. t Empfohlene Anwendungsbedingungen_ fur die Quelle,_. 13. Einsatzklasse nach Projektierte Erfahrungszeitraum " DDR-Standard Einsatsdauer der Einsatsdauer .; TGL 9200 Blatt 3 a'
-70/+200/+30/95//21-3 20 Jahre 15 Jahre ,
-70/+250/+30/95//22-3 '1 Jahr
- 14. Einsatzdauer der Quelle h nter Anwendungsbedingungen gemE8 Punkt 13 14.1 Projektierte (vorgesehene) : s'iehe Punkt 13 14.2 Auf der Grundlage von Erfahrungen bei der Anwendung ge-i wonnene: .
f,
- siehe Punkt 13.
f
..i.
1 g
..I' i
_. : t.,' i y v 3 o/e:
, .f '*
s . y. g. . e e
, - - - - , , - y,- ,-,r- -
-,q , - - , ,+m
. , 04.12,95 16420 BEBIG ISOTCFENTECHNIK Da9 H, 4
- 15. Zusutzliche Angaben I 15.1 Radioaktive Begleitstoffe Bezogen auf den Cs-137-Gehalt:
Cs-134 f5 %; Ru-106 + Rh-106 f1 %: andere il % 15.2 Cmgebungsprufuncen Die umschlossene Strahlanquelle bleibt nach folgenden Pruf-beanspruchungen dicht und funktionsf1hig: K11te. '
- 78 *C Uber 24 Stunden a (TGL 9204 Prufung Aa)
-190 *C " 18 (getaucht in flussigen Stickstoff),
zwischenzeitlich 4maliger thermischer Schock von -190 *C auf. 20 'C. Trockene Wkrme:
+200 *C Uber 24 Stunden a (TGL 9205 Prufung Ba) . Nach
+300 'C " 96 (TGL 9205 Prufung Ba) zeigen die umschlossenen Quellen leichte VerfErbungen.
(Einweis: Nach
+900 *C uber 96 Stunden (TGL 9205 Prufung Ba) bleiben die Quellen dicht, die Oberfluche ist allseitig leicht vorzun-dert, die Quellen-Nummer ist nur noch schlecht lesbar.)
Feuchte WErme: 56 Tage bei +40 *C und 93 % rel. Luftfeuchte (TGL 9206 Blatt 1, Prufung Ca) entsprechende Prufklasse nach TGL 9200 Blatt 2 75/200/56 SO -haltige Atmosphere mit hoher relativer Luftfeuchte: 2 30 Tage bei +40 'C, 93 % relativer Luftfeuchte und einer 80 -X nzentration von 10 mg m'3 (TGL 9209 - Blatt 5,2Prufung Ki).
- 16. Nummer und Datum der Zulassung der Produktion der Strahlen-quelle~ der verliegenden Konstruktion vom Staatlichen Amt fur Atomsicherheit und Strahlenschutz der DDR wurde am 27.10.1983 die Strahlenschutz-Bauart-zulassung Nr. 121083 und am 4.3.1988 deren ErgEnzung Nr. 01 erteilt.
.i Berlin, den 17.3.1988 .
n.d i. d., w na.n si, DOa Zonenlinsmut IUr holopo u.5:miden!crs6ung ' his 3raNenquden und Mephemats ?rof D: G' Y **"*" DDR - 1113 7,w Bens-8id,R.b ntassi+5v.10 sereichsleiter u so (gerste11er) l 4
.e
- . - . . .. .-. . . - _ . . - - - _ ~ - . - . - - - . , _ . ~ . , - . - .~-- . - . . - . . . .
i
. : I I-
- Type description for sealed sources l i '
Gamma radiation source for industrial applications L Type H, Cs-137 ! l ! Country of manufacturing l German Democratic Republic 1 Distributor - , j Isocommerz GmbH j Domestic Commerce and Export Company for Radioactive and Nonradioactive Isotopes ! . I115 Berlin-Buch, Robert-R6ssle-Str.10 ) 1 1 i ' Manufacturer i
- Academy der Wissenschaften der DDR i- Central Institute for Isotope- and Radiation Research j Section Radiating Sources and Nuclear Pharmacy
.; 1115 Berlin-Buch, Robert-Rossle-Str.10 Source Name I Gamma radiation source for industrial applications 1 j Radionuclide. Activity Range j' Cs-137 up to 18.5 GBq (500 mci) ,' Source Tvoe H Labeling Type and serial number Radiochvsical Characteristic of the Source The beta radiation of the Cs-137 is completely absorbed by the capsule. For radiation safety purposes the radiation level can be calculated with the gamma-radiation constant. The radiation level in air for an activity of 3.7 GBq (100 mci) is as follows: Construction General Drawing
. - - - . - ~ - .. . - . .-. . . - - . - - - . - . - . . - . - - - - . . .-. e Physical-Chemical Characteristic of the Radioactive Pan The Cs-137 is spread in a tempered ceramic cylinder, with dimensions of 3 x 3 mm, so that it is nearly abrasion resistant. The radioactive material is inen to water or usual atmosphere. Less than 0.01% of the active material are solved when putting lg ofit into 100ml distilled water of 20 degree Celsius for 48 hours without stirring. l Descriotion of the Caosule ' i Name or Tvoe Material i Stainless steel X8 CrNiTi 18.10 according to TGL 7143 l Wall thickness inner capsule: 0.4 mm l Wall thickness outer capsule radial: 1.1 mm Wall thickness outer capsule axial (window side): 0.4 mm Closing Welded Reauirements to the outside Roughness of the surface: Rz 20 Welding seam bur-free and homogeneous Capsule surface, shape of the holder and marking without visible defects. Source Test Methods and Criteria for the Evaluation Leakage Inner capsule: Immersion test in boiling distilled water, twice for 30 minutes, cooling to <= 40 degrees Celsius during the tests. Elutable activity: <=185 Bq (5nCi) Outer capsule: Bubble test: The sealed source is immersed in Ethylglycol and the pressure lowered to j 15kPa (150 mbar) for at least one minute. No bubbles must be observed during the test. 1 Surface Contamination Inner and outer capsule: Moist wipe test. Activity wiped off <=185 Bq (5nCi) e
i i Damage Testing Visual testing via stereomicroscope (10 fold magnification). l Criteria: Meeting the requirements described in 9.3.4. ! i ! Gassification According to RGW standard 3839-82 and TGL 25294 ! ISO /RGW/ C66646 l l Classified as " material m soecial form" Met. Recommended Source Aoplications Application class according projected working life experienced working life to GDR standard TGL 9200 page 3
-70/+200/+30/95//21-3 20 years 15 years
-70/+250/+30/95//22-3 1 year Workina Life in Applications According to Point 13.
Projected (planed): see 13. Based on experiences while using the source: see 13. l Additional Information i Other Radioactive Material Included Related to the Cs-137 content: Cs-134 <= 5%, Ru-106 + Rh-106 <= 1%, other <=l% Test in Different Surroundings The sealed source remains leak free and functional after being tested under following conditions: Cold:
- 75 degree Celsius for more than 24 hours (TGL 9204 Test Aa)
-190 degree Celsius for more than 18 hours (immersed in liquid nitrogen)
In the meantime four thermal shocks from -190 degree Celsius to +20 degree Celsius. Dry Heat:
+200 degree Celsius for more than 24 hours (TGL 9205 Test Ba). After
+300 degree Celsius for more than 96 hours (TGL 9205 Test Ba) the sources were slightly discolored (To your attention: After
l
+900 degree Celsius for more than 96 hours (TGL 9205 Test Ba) the sources were still leak free but slightly oxidized, and the source number was almost illegible.
Humid Heat:
+40 dyree Celsius and 93% relative atmospheric humidity for 56 days (TGL 9206 page 1, test Ca)
Corresponding test classification according to TGL 9200 Page 2: 75/200/56 SO2 containing atmosphere and high relative atmospheric humidity:
+40 degree Celsius,93% rel. atmospheric humidity and a SO2 concentration of 10 mg/m' for 30 days (TGL 9209 Page 5, Test Ki).
Number and Date of the Production Admission for this sealed source The national bureau for atomic security and radiation safety of the GDR approved the radiation safety type license no. 121083 on Oct.10,1983 and the its supplement no. 01 on Mar. 4,1988.
/ . Attachment 3 4
l l Physical-Technical Federal Institution i Braunschweig and Berlin l Central Institute for Isotope and Radiation Research Bus.-No: 6.22 - G 5 ' Robert-R6ssle-StraBe 10 Tel.: 0531/ 592-6220 ! o - 1115 Berlin-Buch Eraunschweig, 7/25/91 ! Re.: Comments on the test period for leak testing of the radioactive sealed source with the type name HB ! l 1 l Covering: Application of 6/5/91 l l 1.) General Statement Manufacturer: CentralInstitute for Isotope and Radiation Research Division ofRadioactive Sources and Nuclear Pharmacy Robert-Rossle-StraBe 10 0-1115 Berlin-Buch Radioactive Material: Cesium-137 Activity: max.18.5 Gbg 2.) Radioactive Contents b Cs-137 is a component of a 1000.C-tempered ceramic cylinder. Exhaustive trials carried out by the manufacturer on unencapsulated ceramic cylinders for a 48 hour storage in still water with a temperature of 20 C released less than 10 EE -4 of the total activity, so that the radioactive content is to be seen as "slightly extractable" in accordance with DIN 25426 Part 1. 3.) Encaosulation The construction of the capsule follows the accompanying drawing number 6.134. 0101 (3) of 11/16/73 with the latest change of 1/12/88,149 - 6001 (4) of 4/3/91, and the description type HB. l The radioactive material is double-encapsulated. The inner capsule consists of a welded stainless ! steel capsule (Steel X6 Cr Ni Ti 18.10 according to DIN 17440) of 4 mm Diameter, 6 mm height and a wall thickness of 0.4 mm. The exterior capsule is prepared of the same material and has a diameter of 6.4 mm and a total length - including the holder fitting - of 15.9 mm. It is welded from the front. The smallest wall thickness of the outer capsule consists of a sunken front surface of 0.4 mm. .
- . , m _
i l Out of stress testing according to the ISO Standard 2919 and DIN 25426 Pan 1, which was ] carried out by the manufacturer on 8 sources of the model type H, whose encapsulation j corresponds to the type HB, allows for the derivation for the above described sealed source with model type HB of the class ," ISO / C 66646 4 which is sufficient for the designated use in gamma gauging devices.
- 4. Conclusion
, Each source is to be leak tested and tested for surface contamination before shipment to the ! customer (release test). With positive finding is the source to be looked at as sealed radioactive 4 material in the sense of Paragraph I of the StriSchV of 6/30/89. J j With use of the source in industrial gauging and calibration devices further leak testing can be
- done without if the part of the device that contains the source protects it from blows, pressure, or 3 abrasion and is so insulated that the invasion of aggressive steam, liquid and dust is prevented, and j if the radiation emission opening is locked with a metal plate.
j The cost information is enclosed. I 1' On behalf i j (Dr. U. Lauterbach) i l l i J i t 4 a 4 i
. Attachment 4 Commission of the European Communities nuclear science and technology e
THE SPECIFICATION ANbTESTING OF RADIOACTIVE SOURCES DESIGNATED AS ,,SPECIAL FORM" UNDER THE lAEA TRANSPORT REGULATIONS A D. ASTON, A.H. BODIMEADE, E.G. HALL,'C.B G. TAYLOR Amersnam International Ltc Contract No. XVll/322/80.5
-; e FINAL RE:CRT Cirec: orate-General for Researen. Sc:encs anc Ceveicoment 1982 EUR 8053 EN
-. - ..~ . - - . .. . . . - - - . . - - - - - . - . - - . . . . . - - ~ . . . - . ~ . . . . - . . , . ..a . ..-
i
'ITE S4 aITICATION AND '."EE*"INO OF RADIOAt .IVE SOURCES
. DESIGNATED AS "SPECIAL FORM" UNDER IAEA N; SPORT REGUI.ATICNS -
1 4 . 4 . l
~
i 1. Introductic.n j 2. Impact Test Study i , 3. Leak test Method Study J l 4. Discussion of Results y
- 5. Recommendations 4 4
1
- 6. Conclusions i
4 4 i
+
l e 1 . Appendir. 1 A comparision of the features of IAEA and ISO testi I Appendix 2 A study of impact testing 4 Appendix 3 A theoretical and experimental assessment of the vacuum bubble test i 1 1 1 Appendix 4 The pressurised bubble test I Appendix 5 Liquid nitrogen bubble test
' . Appendix 6 Belium pressurisation test 4 Appendix 7 Radioactive leak testing ! e 4
9
l ADSTRACT The. object of this study is to remove some of the uncertainties , , l associated with the application of the I AEA Regulations insofar as l they apply to Special Form materials. The first part of this project involved a comparison of the ISO and , IAEA Regulations and this is covered in Appendix 1. An analysis of - the physical tests has been carried out and in particular the crushing tests have been compared. The second and most important part of the project inv'olved an assess-
- ment of the leakage tests used to evaluate the capsules af ter each of the physical tests.
The review part of this work confirmed the need to clarify certain aspects of IAEA documents ard the need to harmonise with relevant ISO publications. ,p'
~+
The work has defined and confirmed by experiment the relationship between the IAEA and ISO impact and percussion tests. The practical application of the teste particularly with regard to specimen orientation will be aided by the data now available. The work has estab".ished the sensitivities of the primary volumetric , l leak test methods and practical procedures are outlined. Volumetric leak test methods are considered to be more reliable in detecting leakage paths in capsules than methods using solid leachable or non-leachable radioactive contents. Volumetric methods with sensitivities - 1'0" mbar 1/s are considered to be acceptable for establishing whether leakage paths exist after environmental tests. Further work is needed to investigate the equivalence of volumetric leak rates to release of leachable active material in an immersion test. f A further study of the sensitivity, reliability and application of immersion tests using leachable radioactive tracers is needed. f The work reported should assist in the updating and clarification
- and harmonisation of'IAEA Safety Series Nos 6 and 37 and ISO 4919
- and ISO TR 4826.
- i g
I
l l 1 . I
- 1. Introduction "he concept of Special Form radioactive =ater:.al is outlined in IAEA Safety l Series Nos 6 and 37 and it is defined as either indispersible solid radioactive material or as a sealed capsule containing radioact ive material. It will have properties such that even after a transport acciden:, dispersal of material by fracture, cru=bling, sublimation or ignition is improbable.
g The special form concept permits the inclusion of a greater activity (ie up to A instead of A 2) in a Type A package. Definitions of the latter terms are given 3 in Section 1 of Safety Series 6. l A series of environmental tests are specified with which special form material has ; to conform. A leachability test is used to assess specimens after environmental j tests. In the leachability tests the allowed release of activity is fixed at 1 0.05 Ci and this does not bear any relationship to A or A2 levels for various nuclides, 3 y In Safety Series No 37 supplementary infor=atieff is given which should assist in application of the Safety Series No 6 Regulations. It states that by agreement with the co=petent authority, a capsule designed in accordance with ISO Standard l 2919 and meeting a specified minimum ISO classification need not be actually submitted to the tests specified in IAEA regulations. It further advises that by agreement with the competent authority, a volumetric leakage assessment method may be used if tests of a capsule design are not carried out with radioactive l contents. The Iso Standard 2919 includes a nu=ber of environmental tests which are similar in principle to those in Safety Series 6. A further ISO document Technical Report 4826 describes a number of acceptable leak tests which include volumetric ; and radioactive methods. The object of this study is to remove some of the uncertainties associated with the application of the IAEA Regulations insofar as they apply to Special Form materials. In particular the nature of the leak tests to be applied after testing and the criteria for passing the tests need clarification. It is necessary to compare the IAEA and ISO tests and test criteria and consider how they may be better harmonised. Better understanding of these problems will lead to quicker and more certain procedures for the approval of Special Form designs. The first part of this project involved a ecznparison of the ISO and IAEA regulations and ,this is covered in Appendix 1. An analysis of the physical tests has been carried cut and in particular the crushing tests have been ecanpared. The second and most important part of the project involved an assessment of the leakage. tests used to evaluate the capsules after each of the physical tests. problems have arisen in the past with leak tests on inactive capsules regarding the pass rate to be e= ployed and the minimum void within the capsule necessary for a valid test. The basis for figures quoted for equivalence of release of e active material with volumetric leak rate and the validity of immersion tests using radioactive tracers needed clarification.
"o 1
.- . . - _ -. _-, _-_~__-- - __-__- - - . - _ _ _ - _ . -- _ . _ - . - . . . _ _ _ . - _ _ . .
Th2 thsoretical cnd **perimental work is presented in 'atail in a numb 2r of , l appendless. A summ4 ef tha two main parts of the p set is prescntsd in tha nzxt'szetions. Tnis is follow.d by a discussion of tha overall results in relation to their use in assisting in the preparation of revised versions of IAEA and ISO documents. Recomoendations are made concerning specific improvements which should be considered for incorporation by the relevant authorities, i l l
- 2. Impact Test Study -
l A series of over 100 capsules were tested in order to compare the effect of the l ISO Impact Test with the IAEA Impact and Percussion tests. It should be noted * ' that the data given in para 748 of Safety Series No 37 may be interpreted in - different ways. The document states that the IAEA Percussion and Impact tests need not lie performed if the capsule meets I
- Zapact Class 4 at least
)
- %e ISO hammer for impact has a mass 10 t2nes greater than the mass of the capsule plus content y'~
It is assimied in this' report that the' sources with capsule plus content mass less than 200 g should be tested with the ISO Class 4 hammer, mass 2 kg. There are very few designs of radioactive source weighing greater than 200 g, but in these cases the hamer weight would have to be increased. The other possible interpretation is that it is only necessary to use an ISO hammer of mass greater than 10 times the weight of the source plus contents.
; Many source derigns which weig. less than 5 g or 20 g would then only have to ;
meet ISO Class 2 and 3 respecti; ily. - 1 For this experimental work it was decided to select a simple cylindrical capsule ; wit' two designs of weld closure. The capsule void space was left empty or !
! filled with different types of metal inserts. !
It is recognised that there are many alternative source designs and that it may be argued that any conclusions drawn may not be applicable to the complete range of sources. 'However, many of these source designs have been approved as Special l Form material and there should be a considerable amount of test data arising frcan IAEA and ISO tests available with manufacturers and campetent authorities which should verify the conclusions. The results of the study are given in Appendix 2. They show that the effects of the 9 metre IAEA drop tests are negligible. This confirms experience of a very i wide variety of sources over a period of many years. . The ISO Clasa 4, 2 'kg hamer is clearly more severe than the IAEA percussion test, but the ISO Class 3, 200 g hammer was less severe. An ISO hamer weight of 1.2 kg - gives similar results to the IAEA percussion test. -
, All the capsules tested passed the ISO Class 4 test at each of the three
; orientations used. Tests at ISO Class 6 and with an intermediate hamer weight l
of 14 kg showed failure modes and gave some indication as to the most damaging 4 orientation. l 2
To investigate the size of leak occurring 30 capsules were tested at cr just below their failure weight. End welded capsules were tested in the end orientations using weights around 10.8 kg. Tne major finding was that no capsule that passed the bubble test
< 10 "mbar 1/s were found. failed the helium test ie no small leaks 4
- 3. Leak Test Study 3.1 Non-Radioactive Methods ,
.. The sensitivity and practical use of various bubbl' e tests and helium lehk detection methods has been investigated and a suitable test procedure outlined.
The full details of this work are reported in Appendices 3-6 and a su= mary is given below. 1 J 3.1.1 Vacuum Bubble Test w A theoretical assessment of this test and the results of an experimental programme are given in Ahendix,3.. This work demonstrates clearly that under specified and controlled conditions ~ the 10 mm tgst can be valid for capsules with minimum internal voids of if the licuid is glycol or iso-propyl alcohol or 40 m= if the liquid used is water. Leak rates of at least down to 10 mbar 1/s can be detected with this technique. The equi;xnent required is . l simple and cheap and the test procedure straightforward indicating l not only the presence of a leak but also its position. These general !
- points also refer to the variations described below in 3.1.2 and 3.1.3.
This test overlaps substantially, at the large leak end ' with visual examination by which it is estimated at least a 0.2 mm diameter hole can be detected. 3.1.2 Pressurised Bubble Test a A theoretical assessment of this test and the results of experimental work are given in Appedix 4. It is shown that with specified and controlled conditions it is a convenient and ieliable way of tesgng
. sources with volds down to 10 mm to sensitivities of down to 10 mbar 1/s. This test covers a wide range of leaks with the same upper limit 1
as the vacuum bubble test. 3.1.3 Licuid Nitrocen Bubble Test A theoretical assessment and results of experimental work including a comparison with the vacuum bubble test is given in Appendix 5. It is shown that this gst although inconvenient is capable of achieving a semigivity of 10 mbar 1/s with minimum internal void volume down d to 2 mm .
)
3.1.4 Belium Pressurisation Test i The helium pre.ssurisation test is considered from a theoretical stand point and the main points of known theory noted. The largest
' detectable leak, in conjunction with a 10 mm# internal void was !
determined experimentally. The effects of contamination by surface helium which may adversely affect the sensitivity of the test were 3
F 5 .
' investigated as a function of time. The effects, on surface helium contamination, of the nature of the surface and the applied pressure -
l are noted. The work shows that the test may be used with confidence l to esgablish leak rates from specimens with internal voids as small as
- 10 mm providing that correct procedures are followed. It was also clearly shown that the sensitivity range of this test overlaps
- 4 substantially with the range of the bubble test.
1 Practical aspects of carrying out the tests are discussed and a
- j tygcal procedure is outlined. It is shown that a sensitivity of ,
} 10 mbar 1/s can be achieved under a range of practical conditions. i Full details of the work are given in Appendix 6. , , 3 3.2 Radioactive Methods l When capsules containing active contents are tested the wipe test and the a imoersion test are the commonest leak tests employed. No experimental work I has been carried out on the wipe test which is ffisted in ISO TR 4826 as an appropriate test for dete. 6.g leakage afta [ envirdnmental tests, but an
; assessment has' been made of its value. "
l This test as a means of measuring leakage depends on transfer of active material through a leakage path to the external surface of the source. Providing the source was free from surface conemmination prior to any ' environmental test any subsequent wipe will show whether leakage has occurred and the actual value for the quantity of activity removed from the l surface can be measured. The test procedures specify a storage period of ' 7 days after any cleaning process. 'this is an arbitrary time chosen to allow l transfer of active material through a leakage path. Although not stated it is 1 4 assumed that the source is stored under ambient conditions. : This test is particularly appropriate for testing of sealed sources in use ; ) at regular intervals to detect any deterioration in service. It is also ' l very important in checking whether there is any extraneous contamination. } Bowever there are various aspects which make it difficult to quantify leak i a rates le transfer of active material will be affected by its form and properties, the type of leakage path and environmental conditions. It seems inappropriate to use this test as the sole criteria for assessing performance i i of encapsulated material after envi.ronmental tests. In Safety Series No 6 paras 736 and 737, inanersion test procedures are ' specified for "indispersible solid material" and " encapsulated material". 4 4 In the former' case the test detaminas the amount of activity transferred *
~ ;
- to water after the material is subjected to 7 day storage periods under '
j ' water and in humid conditions. It is thus an assessment of leac'umbility. . In the case of encapsulated material the active material has to transfer , i through a leakage path in the capsule to the water in which it is immersed. , ] The transfer of the active material may be facilitated by ingress of water through the capsule wall (s) to reach the active material inside. If fully active sources are tested the results of the immersion test give ( an actual value of the activity released for that particular design of source * ! j under the particular test conditions. Since the solubility and physical form of the active material and type of leak may affect the amount of active ! materia 1' transferred it is difficult to relate the activity release to j volumetric leakage rate. i 4 J
-v- -
,~, , ,n - -
l _ - . _ _._ _ I i The ISO Technica. Report 4826 specifies procedure. icr i=msrsion tests and l l states that the liquid used must have been de=enstrated to be effective in removing the radioactive material present. No attempt is =ade to quantify l the effectiveness of the liquid. l In the ISO standard 2919 the use of soluble tracers in simulated sources is allowed and acceptable tracer activity levels are stated. l The prime intention of the immersion tests described in the ISO documents is I clearly to determine whether the source is leaking. An acceptable threshold l l of 185 Bq (5 nci) is stated and this is related to volumetric leak rates although it is made clear that further werk is needed on this aspect. l i The experimental progra=ce reported in Appendix 7 atte=pted to assess the f sensitivity of the imoersion test and its relationship to volumetric leak tests. Considerable difficulties have been encounteredp devising suitable experimental work to investigate these aspects.mnd ~+ further work is needed. The experimental work censisted of preparation of capsules with known leaks containing active material in various forms, which were then. subjected to immersion tests. The results with capsules containing non-leachable material such as ceramics indicate that volumetric leak test methods are
- more reliable
'9" ^1* * *
* #*** * ""99"**
- 9" indetectingdeakagepaths.5 abar-1/s) volumetric leak rate to 0.005 pCi of 1.33 x 10 Pa.m /s (~ 10 activity release in the specified immersion test errs on the safe side le the ac.ivity release is likely to be less.
The results with active material in leachable form such as caesium chloride were not consistent and throw doubt on the validity of the use of solubleIt l tracers as a sole leakage assessment method after environmental tests. was not possible from the experimental work to confirm any relationship between volumetric leak rate and release of leachable active material in an immersion test. . Until Further work is being carried out to try to clarify these yPa.m" ints,/s to a further evidence is obtained the equivalence of 1.33 x 10 release of 0.005 pC1 of leachable content should be accepted. 1
- 4. Discussion The two sets of tests employed to evaluate sealed radioactive capsules for performance in use (ISO 2919 and ISO TR 4826) and for transport (IAEA Safety I Series 6 and 37) are compared in Appendix 1.
The ISO standards are fairly detailed and explicit inHowever, their description in TR Reportof the 4826 environmental stress tests and leak test procedures. attention is drawn to the fact that further work was needed to clarify minimum free voltIces necessary for bubble test methods and on correlation between activity release and volumetric leak rates. e The IAEA regulations require clarification in a nu=ber of areas and would benefit from incorporation of additional information at least in the Advisory document. 5
1 i A ccupariron of thm /ironstntal tests chows that th- 50 te=perature test Class 6 is et lonct ns stvere cs ths ISO haat test cnd is probably a more realistic one in assessing a transport accident. , The relationship between the IAEA 9 metre drcp test and percussien test an:' the ISO impact test needed investigation and clarification. Experimental work has shown that an ISO type impact test between Class 3 and 4 is needed to give 2 equivalent results to the IAEA test. There would also be benefits in introducing it into the ISO standard for use in perfornance classification since there is a . large difference in effects on sources caused by the current Class 3 (200 g) and -
)
i Class 4 (2000 g) tests, I The area where the need for clarification is greatest is that involved in leak testing. There is a clear need for precise procedures with defined acceptance thresholds. The results of the experimental and theoretical work are considered under three ; I main headings. Test Specimens -[ 1 The IAEA regulations and ISO standard specify that test specimen should be as near as practically possible identical to actual production sources. The use of specimens containing active material is discussed elswhere. It is clear frcan AppenMx 2 that the inner source construction is important. The use of empty capsules for evaluation of resistance to impact will err on the safe , side since contents will nomally increase strength but in temperature tests compatibility effects may be missed. It is therefore clear that test specimens { should be as near as practically possible identical to actual production sources. It appears preferable to use two sources for each test as specified in the ISO standard rather than one as normally used for testing to IAEA regulations. Different sources may be used for each test but the practice of using the same specimen is allowable since it errs on the safe side in that it increases the severity of the test programme. Environmental Tests The pressure and vibration tests ine'luded in the ISO standard are not considered to be relevant to transport operations. The temperature test specified in ISO is more severe that that in IAIA and is clearly an acceptable alternative. I Both sets of regulations contain crushing tests but there is no clearly defined relationship between them. The impact test study Append.1x 2 has one objective of detemining relationship between the impact, percussion and drop test. . . Tba presant regulations as stated in Section 2 are capable of misinterpretation I although. the generally accepted view is likely to be that ISO Class 4 must - I be used as an equivalent test to the IAEA 9 metre drop test and percussion ! test for capsules up to 200 g mass. Experimental work has shown that, for i the W of capsules used, an ISO impact test procedure using 1200 g is
- equivalent to the percussion test and that the effect of the 9 metre drop on specimens less than 200 g is negligible.
6
1 The results cen. .: an assessment sinet frc= a con .eration of =c=entu: at l' impact the 9 = drop becomes ecuivalent to the 1200 g ISO test at source weights ]* of 400 g. It see=s appropriate to recc= mend that an ~IO impact test with ha==er = ass
- 1000 g is used for sources up to 120 g mass. In cases where sources are ;
j between 120 g and 200 g the present ISO impact test Class 4 will be suitable and in the very few cases above this a special hammer mass 10 times the source l ! will be needed. 4 . In both the IAEA Regulations and ISO Standard it is specified that the 1
. source must be struck in the most vulnerable orientation. As discussed I - in Appendix 2, in many cases this will be obvious from a study of the design.
j Bowever, when doubt exists it will be necessary to test in more than ene j orientation. i !, Evaluation More confusion hAs eEisted with the leakage assessment methods carried out ! ~ - after the environmental tests than with the tests themselves. The leak tests. y, employed,have been used for many years, but ,t3eir sen'sitivities and precise procedures are ill-defined. A major part of this programme of work was
- concerned with a study of leak test methods. The aim was to examine the tests in depth, define their sensitivities, and specify recessnanded l
procedures. The results are presented in Appndices 3, 4, 5 and 6 for 3 volumetric leak tests and Appendix 7 for active leak tests. The first stage in the evaluation of any capsule after the environmental tests is a close visual examination. It is only after confirming that no l i major failure has occrred that the leakage assessment methods are carried I out. 4
^
[ In the IAEA Regulations a leaching assessment method involving inanersion l testing is described and an acceptance limit of 0.05 pci quoted. The l concept of approval of Special Form material is based on an acceptable f level of activity leakage following environmental tests. i
- In the ISO Standard 2919 the criteria is that the capsule should maintain i its. integrity and acceptable leak test methods are given in Technical Report 4826. The imunersion tests in this report quote an acceptance level
- of 0.005 pC1, but they are simply used as a method of determining whether the capsule has failed, it does not imply that a release of activity after l environmental tests is allowed.
j 1 f The IAEA Safety Series No 37 para 749 states that the following relationships between volumetric leak rate and activity releases of 0.05 pCi may in most cases be considered to apply. i e
- For solid content
-4 -5 3 I
> , , 10
-4 Torr 1/s (1.3 x 10 mbar 1/s) (1.3 x 10 Pa.m /s)
! For liquid and caseous content i .
-6 -6 -7 3 4
10 Tor: 1/s (1.3 x 10 mbar 1/s) (1.3 x 10 pa.m /s) The ISO Technical Report 4826 quotes an equivalence in a slightly different I way and relates it to 0.005 uC1. f a ! 7 1
,,,,.,,n.~r---
For non-i .:hable content
~
1.33 x 10 Pa.m /s (1.3 x 10 r.bar 1/s) (10 Torr 1/s) - l Fcr leachable or caseous content
~0 ~ ~
1.33 x 10 Pa.m /s (1.3 x 10 mbar 1/s) (10 Torr 1/s) The origin of the 0.05 pCi in Safety Series 6 and the equivalence figures *
~
is obscure. i i The work on volumetric leak test methods centred around their sensititivties ~ related to the free internal volumes of test specimens. It has been ! established thgt suitable test methods can be adopted, with gree volumes
~
as low as 5 mn which can give sensitivities better than 10 abaz- 1/s - . ; The helium pressurisation test is also shown to be capable, using acceptable practical procedures, of giving sensitivities considerably in excess of this. I y ! If we accept the equivalence figures quoted th'en the' sensitivity of any of
- the volumetric bubble test methods is adequate for evaluation of encapsulated i non-leachable active material after special form environmental tests. l l
Experimental work reported in Appendix 7 supports acceptance of this: approach. i This is further confirmed by the results of the impact study in Appendix 2 where it was found that gy leaks developed were all detected by a method , with a sensitivity of 10 abar 1/s. l The experimental work using tracer quantities of a leachable radioactive material has given erratic results. It has not been possible to establish a relationship between volumetric leak rate and activity release and has thrown doubt on the reliability of innersion test methods using tracers. Sirsee there already existed scoe doubts about scaling up of .results to take into account various quantities of radioactive material used in fully active sources it seems prudent not to use this test on its own to evaluate specimens after environmental testing. Further work is needed c.i this aspect but it has to be recorded that it is difficult to devise practical experiments'to provide firm evidence.
- 5. Recanmendations It is reccamended that IAEA are invited to review the work presented in this report and that consideration should be given to incorporating where appropriate the -
~
following in Lwre versions of Safety Series No 6 or No 37.
- Definitions of permitted forms of specimens similar to those given in ISO 2919.
, a Two specimens anould be used as required in ISO 2919. Different =-
specimens may continue to ba used for different tests.
- Detailed procedures for the impact test are specified as in .-
ISO 2919. A hanner weight of not less than 1.2 kg is specified for sources of mass less than 120 g and not less than 10 times the weight of sources of mass above 120 g. 8 l l i - _ _ _ _ _ _ _ _ _ _ _ _ _ . - . _ - --~ .-.- - = .
Tha IAEA 9 matre drop (I=pret) test is withdrawn. The heat test is specified exactly as ISO Class 6.
- The bending test is retained.
- The permitted activity release in the immersion tests is changed to 0.005 Ci as in the ISO Technical Report 4826.
. The equivalence of 1.33 x 10 Pa.m /s (1.3 x 10 ~ mbar 1/s)
- to a release of 0.005 Ci for non-leachable centent is accepted.
~
- The equivalence of 1.33 x 10' Pa.m /s (1.3 x 10 mbar 1/s) to a release of 0.005 Ci for leachable or gaseous material continues to be accepted u. .il further data is available.
- A list of acceged volumetrie leak test methods with sensitivities better than 10 mbar 1/s should be presented together with test procedures. The procedures should fhcorporate a requirement for primary visual inspection S-tor to any volumetric leak test method.
l NOTE: If a higher sensitivity than 10- mbar 1/s is regarded as being necessary then overlapping test procedures eg visual / bubble / helium spectroscopy will be required if the normal vacuum bubble test is used.
- A suggested form of application for Special Form appretal should be; included.
- Guidelines for pemitting approval of series of capsules of sirtilar shape but differing in dimension over a narrow range should be produced based on the assumption that. only the weakest me=ber needs testing.
It is reccc: mended that ISO TC 85(SC4) are invited to review this work and that consideration should be given to incorporating the following in future ISO publications.
- An additional Impact Test (eg Class 3A) is introduced into ISO 2919 with a haz::mer weight of 1.2 kg.
- Technical Report 4826 should be reviewed and re-issued taking into account the new data now available.
- 6. Conclusions
- 1. The review part gf this work confirmed the need to clarify certain aspects of IAEA docu=ents and the need to harmonise with relevant ISO
. , publications.
- 2. The work has defined and confir=ed by experiment the relationship
, between the IAEA and ISO impact and percussion tests. "*he practical application of the tests particularly with regard to specimen orientation will be aided by the data now available.
9
_._m._._. . . . . . . _ . _ . . _ _ _ .._ ____
. l
- 3. Thz work has establichsd the sensitivities of the primary volumetric I leak test. methods and practical procedures are outlined. !
- 4. Volumetric leak test methods are censidered to be more reliable in '
detecting leakage paths in capsules than methods using solid leachable or non-leachable radioactive contents.
~
- 5. Volumetric methods with sensitivities ~10 abar 1/s (1.33 x 10 Pa.m /s) are considered to be acceptable for establishing whether leakage paths exist after environmental tests. ""
- 6. Further work is needed to investigate the equivalence of volumetric .
leak rates to release of leachable active material in an innersion test.
- 7. A further study of the sensitivity, reliability and application of ,
immersion tests using leachable radioactive tracers is needed. 4
- 8. The work reported should assist in the updating, and clarification and harmonisation of IAEA Safety Series Nes,8 and 37 and ISO 4919 and ISO TR 4826. 4 i
l l 4 l e e g 9 10
APPC; DIX 1 A COMPAPlSON OF THE TEATURES OF IAEA AND ISO TESTS INTROD CTION TO IAEA AND ISO TESTS TABULATION OF DETA!!.,5
- 1. hvironmental tests
. 2. Specimens , , 3. Evaluation
- 4. General and Administrative aspects a
Y i l W g 0 s
. ~ . -- . .- . _ _ . _ . . _-
APPENDIX 1 - A COMPARISON OF FEATURES OF IAEA AND ISO TESTS INTRODUC* ION TO IAEA AND ICO TESTS gsrt.s for Special Form Radioactive Material a he regulations for the Safe Transport of Radioactive Materials IAEA Safety Series No 6 1973 revised version specify ,
- 1) Test methods to subject specimens to stress conditions. .
- 11) Criteria for specimen preparation and use.
iii) Criteria for physical condition of specimen after stress tests. iv) Imaching assessment methods. The publication Safety Series No 37 Advisory Material for .the Application of the IAEA Transport Regulations 1973 sets out supplemengary information to assist in deciding how to achieve the requirements of the regulations. his document discusses some general features of the design of sealed sources which should be considered. It also states that ey agreement with the competent authority concerned a capsule designed in accordance with ISO /TC85/SG4/Wgl (Secr 53) 77 and meeting specified minimum ISO class, specifications need not be actually subjected to tests specified in the IAEA regulations. It further advises that by agreement with the competent authority, a volumetric leakage assessment method may be used if tests of a capsule design are not performed with radioactive contents. Tests for ISO 2919 classification,
; he work carried out by ISO /TC85/SC4 culminated in International Standard ISO 2919. 21s document specifies:
- 1) Test procedures to subject specimens to specified stress conditions.
ii) The types of specimen acceptable, and number to be tested. iii) Criteria for determining compliance with the tests. iv) Wat leak tests such as those described in ISO Technical Report 4826 shall be used to determine maintenance of integrity. Relevant details of each of these standards is tabulated below. 4 e
-- - . _ - . . _ - . . . - . - = - . ~ - . - - - . - . ~ _ _ _ - . . . . . - . - _ ~ . . . . - . . _ - . - - - . _ _ _ .-_
t
* .o
- 1. DIVIItOilHDITAL TESTS TYPE OF TEST IAEA ISO It1 PACT The specimen shall fall onto the target from a height The specimen is mounted on a steel anvil of 'ss of 9 m. The target shall be a flat, horizontal at least 10 times that of the hammer. The el surface of such a character that any increase in its hamer which has a flat striking surface 25 mm ,
resistance to displacement or deformation upon impact in diameter with its edge rounded to radius of by the specimen would not significantly increase the ' 3 mm is dropped onto the most vulnerable part damage to the specimen. of the specimen from a height of 1 mett e. (A steel plate is normally used). Class 4 - mass of hamer 2 kg.
- Class 3 - mass of hammer 200 g.
PDtCtISSIOll The specimen shall be placed on a sheet of lead NOTE:-- Safety Series flo. 37 IAEA para. 74B which is supported by a smooth solid surface and states that an ISO test is accepted as meeting struck by the flat face of a steel billet so as to both the IAEA impact and percussion providing: produce an impact equivalent to that resulting from 3.t class 4 at least is met a free fall of 1.4 kg through 1 metre. The flat face of the billet shall be 25 mm in diameter with : 2,[he ISO hamer has a mass 10 times the edges rounded off to a radius of 3 m i O.3 mm. greater than the mass of the capsulc un The lead, of hardness number 3.5 to 4.5 on the content. Vickers scale and not more than 25 mm thick, shall cover an area greater than that covered by the specimen. A fresh surface of lead shall be used for each impact. The billet shall strike the specimen so as to cause maximum damage. I l e
_ m. .. . . _ . _ _ _ _ _ _ . - _ _ _ _ _ l l l l l
=.
TYPE OF TEST IAEA ISO Pill!CrllRE !!o requirement. The specimen mounted on a steel anvil is stnick by a 3 rnm diameter 6 mm long pin of specified hardness which is rigidly fixed to a steel hammer. Five classes of test are available with increasing mass of hammer and pin from I g to 1 kg. A drop height of 1 metre is used.
'hz +
r 1
=
g e # 0
# I
- _ _ _ _ _ _ _ . __ - - - _ _ _ = _ _ _ _ . -- _ ___ - _ _ . _ - - _ _ _ _ - - _ _ _ - _ - - _ _ _ - - _ _ _ . - _ _ _ _ - - - _ - -
~ *
- 2. SPECIMENS IAEA f ISO r
Specimens (solid radioactive material or capsules) to be tested shall be prepared as normally presented for transport. The following defined types of specimens are acceptable. ,
'Ihe radioactive material shall be duplicated as closely as Dummy Sealed Source: Facsimile of a radioactive sealed practicable.
source the capsule of which has the same construc' , and is made with exactly the same materials as the _ of the sealed source that it represents but containing, in place of the radioactive material, a substance p resembling it as closely as practical in physical and [ chemical properties. i Prototype Source: Original of a model of a sealed source which serves as a pattern for the manufacture of all sealed sources identified by the same model designation. Sealed Source: Radioactive source sealed in a capsule .. or having a bonded cover, the capsule or cover being strong enough to prevent contact with and dispersion of f the radioacQve materlat under the conditions of use an.1 wear for whici lzit was designed. Simulated Yource: Facsimile of a radioactive sealeu > source the, capsule of which has the same construction ' and is mada with exactly the same materials as those of the sealed. source that it represents but containing, in j place of the radioactive material, a substance with mechanical, physical and chemicalp'roperties as close as , possible to, those of the radioactive malerial and , containing, radioactive material of tracer <piantity only. [' The tracer is in a form soluble in a solvent which does not attack the capsule and has the maximum activity compatible with its use in a glove box. The following l activity levels are acceptable: I 90Sr + 90y as soluble satt: 50pcl 60 Co as soluble salt: 20pC1 ! _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _ _ - - _ _ _ _ . - _ _ _ _ . _ _ - . _ _ - _ - - _ _ _ _ _ - - _ _ _ _ _ --m__ _- _ u _
IAEA IM The number of specimens required is not stated but para 729 The classification of each sealed source type shall be Safety Series 6 states that a different specimen may be used determined by actual testing of two sources (sealed, for each of the tests. prototype, dummy or simulated) of that type for each test Para 703 deals with ntuber of specimens but does not t tve " * * ' # Y *# * " # " E#* "* 8 ** C dem nstrate that the source would pass the test if the actual ntabers and appears to relate to packages only. test were performed. Different specimens may be used ! each of the tests. d.
' ' ),
4 O e
- p g ,
e
- 3. EVALt!ATION i i IAEA ISO The specimen shall not melt or disperse when subjected to the The specimen shall be examined visually.
heat test. 1 The specimen shall not break or shatter when subjected to the impact, percussion or bending tests. Af ter ear-h :.est, a leaching assessment shall be performed on the specimen by a method no less sensitive than the methods given below. 1 For Indispersible Solid Material: l a) The specimen shall be immersed for 7 days in water at . ambient temperature. The water shall have a pII of 6 - 8 and a maximum conductivity of 10 pmho/cm at 20 C. ;! b) The water with specimen shall than be heated to a i' temperature of 50 1 5 C and maintained at thic ff temperature for 4 hours.
- j; c) The activity of the water shall then be determined. The ISO Standard is not applicable to this form of ![
radioactive material. There is no equivalent leachabiliiy !; d) The specimen shall then be stored for at leant 7 days test. In still air of humidity not less than 90% at 30 C. I! e) The specimen shall then be immersed in water of the same specification as in (a) above and the water with ['r specimen heated to 50 1 5 C and maintained at this , temperature for 4 hours. 'l i f) The activitly of the water shall then be determined.
'f The activities determined in (c) and (f) above shall not '[
exceed 0.05 pC1. _______m,_.._____-_.____.__m _ . - .m_______._________ -_-_ _____ _ . _ _ _ _ _ _ _ _ _ . _ _ - _ __ _ _ _ _ - _ _ _ m m_____
_ _ _ _. _ - . _ _. . _ _. . . . .- --. _ ._. -___.__.m . . _ . . . _ __ _ _._ . _ . . . _ m . i f ISO
~
IAEA
, i For Encapsulated Materials: % e ISO Technical Report 4826 describes a variety of 1eak test methods for sources by both radioactive and -
a) %e specimen shall be immersed in water at ambient non-radioactive methods. It gives a comaarison of leak l temperature. test methods and acceptable thresholds, which are based on those quoted in IAEA Safety Series No 37. b) he water shall have a pH of 6-8 with a maximum conductivity of 10 sho/ca. W e water and specimen i shall be heated to a temperature of 50 + 5*C and mainitained at this temperature for 4 hours. c) We activity of the water shall then be determined. d) W e specimen shall then be stored for at least 7 days , in still air at a temperature not less than 30 C. e) Repeat a).
- f) h e activity of water shall then be determined. ,
% e activities determined in (b) and (e) above shall net et exceed 0.05pci. %,
I When, by agreement with the competent authority concerned, the performance tests of a capsule design are not performed with radioactive contents, the leakage assessment may be made : by a volumetric leakage assessment method. A rate of 10'4 torr - 1/s for solid content and a rate of 10-6 torr - 1/s for liquid and gaseous content would be considered in most ' cases to be equivalent to the activity release of 0.05pci prescribed above. __ t
+ i g e 8 0
- 4. GE2iERAL AllD ADMIrlISTILATIVE ASPECTS IAEA ISO The Safety Series No. 61973 Regulations state:- This International Standard establishes a system of Approval of Special Form Radioactive Material classification of sealed radioactive sources based on performance specifications.
Ifgheactivityoftheradioactivematerialinvolvedexceeds It provides a set of tests by which the manufacturer of 10,, A 1, the design for special form radioactive material ! sealed radioactive sources can evaluate the safety of his sbstl require unilateral approval. An application for products under working conditions, and by which t he user i approval shall include:- of such sources can select types which suit the applicatir n he has in mind, especially where protection against the
- a) A detailed description of the material or, if a release of radioactive material, with consequent exposure fo I
capsule, the contents; particular reference shall ionizing radiations, is concerned. This International be made to both physical and chemical states; Standard may also be of guidance to regulating aut horil.ien. b) a detailed statement of the design of any capsule to~ This International Standard makes no attempt to classify be used, including complete engineering drawings and either the design of sources and their method of schedules of materials and methods of construction to construction or, their calibration in terms of the radiat lon 'i be used; emitted. Radioactive materials inside a nuclear reactor and fuel elements h[e specifically not covered by this 2 c) a statement of the tests which have been done and International Standard.
- their results or evidence based on calculative methods to show that the material is capable of meeting the tests, or other evidence that the special form radioactive material n.eets the requirements of these Regulations.
The competent authority shall establish a certificate stating that the approved design meets the definition of special f orm radioactive material in Section 1, para.135, and shall attribute to the design an identification mark. The certificate shall specify the details of the radioactive material.
IAEA ISO The Safety Series Ho. 37 1973 comment on some, design aspects: This International Standard includes a list, which is not intended to be comprehensive, of typical applications of ' The following factors require consideration during the sealed radioactive sources with a suggested test schedule design of a sealed sources- for each application. These schedules are minimum requirements corresponding to the applications in the a) The possibility of galvanic reaction between broadest sense, does not consider exposure of the seale ' dissimilar materials, particularly metals for source or the source-device to fire, explosion and encapsulation; and corrosion. In the evaluation of sealed sources and source-device combinations, the manufacturer and user have to b) the location of welds, which should be such as to consider the probability of fire, explosion and corrosion h minimize both stress concentrations and the and the possible results. possibility of failure under impact, percussion and bending. Factors which should be considered in determining the need for actual testing ares-This document also points out that it should be ascertained that the capsule, even af ter use in the intended manner, a) Consequences of loss of activity; will still meet the regulatory specification for special form radioactive material (Section I of the Regulations, b) quantity of active material contained in the scaled para. 135). Consideration should be given to external source; l contamination from the material cnd to longer-term effects, , such as diffusion of the radioactive contents through the c) radiotoxicity; capsule walls. #' A d) chemical and physical form of the material and the geometrical , shape; e) environment in which it is used; and f) protection afforded the sealed source or source-device combination. 3
. O e e
- e
- ..--- _ . - _ _ - . . _ _ _ - _ . . _ . - . _ . _ . - . - _ _ . _ - - . _ - - - . _ _ . - - _ - _ - _ _ _ _ _ . - _ _ _ _ - - _ - - . - - - - e-- , - - - ~ w ,,'owa
. m _
s . . TYPE OF TEST IAEA ISO TEttPERATURE The specimen shall be heated in air to a . Class 6: The specimen is subjecteil to -40"C temperature of 800 C and held at that temperature for 20 minutes then heated to 800"C an1 held fot ' for a period of 10 minutes and shall then be allowed to cool. 1 hour and cooled. A further thermal shock te:;l involving heating to 800 C for 15 minutes followed by rapid cooling in water is also applied. i flOTE:- Sa.'ety Series tio. 3 7 I AEA par a. 748 states that this ISO test is accepted as an alternative to the IAEA temperature test. BEllDIllG The test is applicable only to long, slender llo similar test. sources with both a minimum length of 10 cm and a , length to minimum width ratio of not less than 10 The specimen shall be rigidly clamped in a horizontal position so that one half of its length protrudes from the face of the clamp. The orientation of the specimen shall be such that the ' j specimen will suffer maximum damage when its free end is struck by the flat face of a steel g billet. The billet shall strike the specimen so as to produce an impact equivalent to that resulting from a free vertical fall of 1.4 kg , through 1 metre. The flat face of the billet shall be 25 mm in diameter with the edges rounded off to a radius of 3 mm i O.3 mm. i VIBRATIOtl tio requirement. The specimen is subjected to vibration. Three classes of test are specified with increasing severity. The test procedure and test cycles are , specified. t P _ . - . _ . _ _ - . . . . - - _ . _ ....___._.a _ - - - - - _ . . _ _ _ . - . . . _ _ . - _ - . - _ _ - _ _ - - - . _ . - _ . - . _.__.._-_----..---__a _ - - _ - - - - - - - r- - - - - - - - , - -- --
APPENDIX 2 A STUDY OF IMPACT USTkNG ,
- 1. INTRODUCTION
- 2. EXPERIMENTAL
- 3. RESULTS -
3.1 9M drop test
- 3.2 ISO. Class 3, ISO Class 4 and Percu.ssion test 3.3 ISO Class 6 3.4 Other tests
- 4. DISCUSSION
- 5. CONCLUSIONS a
'e LIST OF TABLES TABLE 1 ISO Class impact tests TABLE 2 ISO Class 6 impact tests TABLE 3 Size of leaks ocurring S
4
- 4
'12T CF FIGURES FIGURE 1 CAPSULES USED FOR TESTING FIGURE 2 ORIENTATIONS USEO IN IMPACT TESTING FIGURE 3 9M DROP TEST FIGURE 4 ISO CI.hSS 3 FIGURE 5 ISO CLASS 4 FIGURE 6 PERCUSSION TEST FIGURE 7 1.2Kg IMPACT FIGURE 8 ISO CLASS 6 IMPACT ON END (EMPTY CAPSULE)
FIGURE 9 ISO CI. ASS 6 IMPACT ON SIDE (EMPTY CAPSIIf.E) 4 . FIGURE 10 ISO CLASS .6 IMPACT ON CORNER (EMPTY CAPSULE) FIGURE 11 ISO CLASS 6 IMPACT ON END (STEEL INNER) FIGURE 12 ISO CLASS 6 IMPACT ON SIDE (STEEL INNER) FIGURE 13 ISO CLASS 6 IMPACT ON CORNER (STEEL INNER) FIGURE 14 ISO CLASS 6 IMPACT ON END (ALUMINIUM INNER) FIGURE 15 ISO CLASS 6 IMPACT ON SIDE (ALUMINIUM INNER) FIGURE 16 14Kg IMPACT FIGURE 17 14 Kg IMPACT I E e 9 0 4
- ~ , - -. - - .. - ~ . - - . -- - - . _ - _ . . - . - . _ . . - - . - -. . _ - . .-
- 1. INTRODUCTION The object of this study was to investigate important areas of bpact testing.
ever a hundred specially manufactured capsules were prepared and tested. A ccmparison was made between ISO impact tests and tne nine metre drop and percussion tests as stipulated in Special Form regulations. The capsules were tested empty and with steel or aluminium inserts and the effects of using such inserts cbserved. Different orientations were used in an ! attempt to discern the most vulnerable orientation. The capsules were tested to destruction using the most severe ISO class of impact test ie ISO Class 6 (20kg) . From the results the size of leaks occurring and the mode of failure of the capsules could be determined.
- 2. EXPERIMEJTAI, The most commonly found type of source design is a gimple cylindrical shape.
There are two cor.venient methods of fixing the 11dp by side welding or end welding. '4
- Two types of capsule were used for the tests, one with a side welded lid and the other with an end welded one. Apart from the type of lid the capsules were of similar dimensions. rig 1 shows sketches of the capsules. The side welded capsules weighed 3.59 and the end welded ones 2.6g. Some of the capsules were tested empty others had steel or aluminium inserts. These were cylindrical in shape and filled the entire void space inside the capsule. The steel insert weighed 2.99 and the alum:nium one 1.1g. ':hree types of impact test were investigated.
- 1) 9m drop test The source is dropped dr om a height of ni. e metres on to steel.
i
- 11) Percussion test The source is placed on to lead and a hammer of weight 1.4Kg dropped from a height of one metre on to it.
Tne face of the hammer has a diameter of 2.5cm. iii) ISO impact tests The source is placed on a steel block and struck by a hammer from a height of one metre. The face of the hanner has a dian.eter of 2 5cm. The hammer weights for the ISO Classes are shown in Table 1. TABLE 1 ISO CI. ASS IMPACT TESTS CLASS Wt Kg 4 3 0.2 4 2 5 5 6 20 The capsules were struck in three orientations as illustrated in fig 2. These were on the end on the side and on the corner. 'Ihe capsule was held on its corner by a small piece of plasticine.
. . _ . . _ _ . . _ _ _ . . . . _ . . - _ _ . _ . - _ _ . _ . . _ _ _ _ _ _ . _ . . _ . . _ _ _ . . . _ . . , . -m.
7 After.cn impact t the sources were inspected is. /idually and then
- photographed. They were then leax-tested. The leak test involved a bubble
- test followed by a helium pressurisation test. The helium pressurisation *
; test was performed so that leaks as small as 10 mbar 1/see would be
- detected.
t
- 3. RE3ULTS
~
3.1 9m drop test . J Capsules were tested empty and with steel inserts. There was a negligible , amount of damage caused. Capsules with steel inserts suf fered slightly more damage because of the extra weight. See Fig 3 which shows the damage caused to an empty capsule. 2 3.2
- ISO Class 3, ISO Class 4 and Percussion test !
{ l Side welded empty capsales were tested in three orientations. The tests } were ISO Class 3, ISO Class 4,and the percussion [ A range of weights between ISO Class 3 and ISO Class 4 were teste4 to find the ISO weight j which caused the same damage as the percussion. ISO Class 4 proved more severe than the percussion which proved more severe l l than ISO Class 3. Weights of 0.9 Kg, 1.2 Kg and 1.4 Kg were tested and 1.2 i Kg was found to produce the same damage as the percussion test. This is somewhat orientation dependent since the cushioning effect of the lead is j greater for a greater surface area. Figs 4, 5, 6 and 7 show ISO Class 3, 3 ISO Class 4, percussion and 1.2 Kg tests in one ' orientation, on the end. l Gimilar relationships are shown with other orientations. All capsules f passed the leak test. i 3.2 ! ISO Class 6 (20 Kg) j 2- i ISO Class 6 was a convenient weight to test capsules to destruction. Side l welds and and welds were used in conjunction with steel.and aluminium j inners as well as testing empty. Two types of aluminium inner were used,
- solid aluminium insert of weight 1.1g and a hollow insert of weight 0.7 9.
a Table 2 si=== arises the results. } Table 2 ISO Class 6 Impact tests SPECIMEN END SIDE CORNER
; Side welded empty FAILED FAILED FAILED j End welded empty FAILED FAILED FAILED .
t Side welds + steel inner PASSED PASSED PASSED ' I End welds + steel inner PASSED PASSED PASSED ) i Side welds + solid altaninissa inner PASSED PASSED PASSED
- i. Side welds + hollow altaninium inner .
FAILED FAILED FAILED , 5 I End welds + solid aluminium inner PASSED PASSED PASSED End welds + hollow aluminium inner PASSED PASSED FAILED , j From these results the effects of testing with inners is clear. The tougher the inner material the less likely is the capsule to fail. 1 2 i
I The mode of failure in :nese cases is quite catastrophic particularly for those capsules tested empty. The major:ty of f ailures were obvious, as leaks could clearly be seen. alu:ninium inserts required a vacuum bunble One test or two of those with to confirm tne presence of a lear.. All the capsules which passed the bubble test were helium pressurised and no additional leaks detected. Figs 8, capsules, 9 and 10 show the damage caused to side welded empty
; easily seen. tested in three orientations. The mode of failure is The capsule tested on its end has failed at the weld.
- The capsule tested on its side has folded in at the base and split, the weld has remained intact. The capsule tested on the corner has catastrophically split at the weld.
--gs 11, 12 and 13 show and welded capsules with steel inners, no sailures have occurred.
Figs 14, 15 and 16 show eno welded capsules with hollow aluminium inners, note that the end and side or4pntations passed while the corner failed. 3.4 Other tests A range of weights between ISO Class 5 and ISO Class 6 were tested on empty capsules. At around 11 Kg it was found that the side and corner orientations passed and the end failed. This was true for both side and end welded capsules. The modes of failure were slightly different. j The side welded capsules failed at the weld and
' split around the middle where the capsule folded in. The end welded capsules failed around the middle with the weld remaining intact.
Figs 17 and 18 illustrate this. Further testing indicated that the side welded capsules failed around 10 4 Kg and the end welded capsules at around 10.8 Kg. To investigate the size of leak occurring large numbers of capsules were tested at or below their failure weight. End welded capsules
.were tested in the end orientations using weights around 10 8 Kg.
After the impact test the capsules were bubble tested. Those capsules which passed 'the bubble test had a helium pressurisation test performed on them the conditions of which were such that leaks as small as 10-7 in Table 3. mbar 1/see were detectable. The results are shown Table 3 Size of .'.eaks occurring Weight of Em=mer No of Capsules Bubble Test Helium Pressurisation in Kg Tested Passed
. Failed test on capsules which ~
passed bubble test 11 0 3 0 t 3 - 10 9 5 3 2
,* 10 8 ,
8 5 No leaks detected 3 10 75 i 4 No leaks detected ( 3 10.7 i 5 1 No leaks detected 1 4 i 10 6 { 3 1 No leaks detected
) 3 0 1 1 No leaks detected
- 4. DISCUSSION For the size of sources considered (up to 200g) negligible damage is caused by ,
the nine metre drop. De percussion causes the'same amount of deformation as an ISO impact with a weight of 1.2 Kg. It is difficult to make conclusions on the effects of testing in different orientations since only one design of capsule has been considered. However for this capsule design when the capsules possess an insert the corner is the most - vulnerable orientation. When the capsule is tested empty the end is the most ' vulnerable. The end welded capsules were on the whole tougher than the side welded ones. However interestingly enough the side welded capsules suffered less deformation for a given weight than erad welded capsules. All the leaks found in this study were detectable on the vacutan bubble test. Small leaks le less than 10-5 mbar 1/see were not observed. n ere is no graded spectrtsu of leaks from zero upwards. The snail increments in hanumer ) weight used during the tests would have revealed $1s if St occurred. It is ' also of interest to note the small range of hansner weights, about dog, between i 100% failure and zero failure. Inside this critical region there is a probability distribution between a capsule failing and not. In view of the absence of small leaks a sensitivity of 10-5 abar 1/see appears adequate for leak tests. ;
- 5. CONCI,USIONS Any study of this nature has limitations since it is impossible to test every .
conceivable type of capsule design. However several definite conclusions can be put forward. , 1 i) An ISO impact of 1.2 Kg can be regarded as equivalent in severity to a nine metre drop and percussion test, providing the source has a mass of less than 200g.
- 11) Imak test methods with a sensitivity of 10-5 abar 1/see seem adequate to detect any leaks occurring in capsules.
iii) Prototype sources should be tested with an inner which as closely as possible resembles that of the production source. iv) Analysis of source designs will normally reveal the most sensitive orientation. In some cases this may not be possible and two sources . should be tested in different orientations. 1 o . t l 1 4 l l
_ . . . - . . .__. . - . ~ . . . . ~ . . _ . _ _ _ _ _ . - _ . . . . . . . . . . _.. _.-______-_ __-- _ _ . _ _ . . _ . . _ _ . _ . . _ . E' *s 'e:C* IE0 : 1 f i f Ji - - . l
~L - A
{ 756 w CA D i, e t , i I ( l i
. t v,i e
Y l u
. t
_f - A 7, 5 i
\ mass o Ls ~ ~
t SIDE WELDED ,,, 'I e7
~4 .
\ e. . w se __ :
E E-e.
. T_.
- A !
\ wucae o45 mm ,
FIG 2. ORIENTATIONS USED IN IMPACT TESTING END UIDE O PLASTICINE C < x , o , \. 1
\ 'g \ \ M \ s \ \ \. \\ \ \ \\ \ \\ \ .* \ \\ \ '
i i t . q l t , I i 1
~
l e EB - # \ I A
-Lu l
i l Fic.3 E=pty side welded capsule after 9m drop test. l l i l l l 1 - i . 1 i l l i I 1 , E b( 1 1 4 Fig 4. Empty side welded capsule after 1503 impact test. l i 4 1 J i 4 i j i . . . - _ _ _ _ _ . . _ __ ._ __ ._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
l-I --- - ._ _ _ _ __ l i i l ( l If - l l l Fig.5 E=pty side welded capsule after ISO 4 impact test. 1 i l I l i I
- a. M,f i
b . j d Fig.6 E=pty side welded capsule after percussion test i i
l l E..t.. 1 I l Fig. 7 Empty side welded capsule after iz: pact of 1.2 Kg hemmer I i i i l l l 1 t
. 9*].;_.
1 l r i. , i
;- 'a..
O % - , I Fic. 8 E=pty side welded capsule after ISO 6 impact on end.
- i i
4 1 h
} .i
i 1 i . 4 i j 1 a ( A 1 i i i 1
- .M ,
i - 4 ..
~.
,. ~
b '..$ g l m-{,7 N:py[.e Se) - s I i ,I Fic. 9 D::pty side welded capsule af ter ISO 6 impact on side. } l 1 \ l l $ .l l . I 1 . l 2
- e.
- e s
J
- [o 19 g j .
1 . I i i s ( 4 y Pic. 10 D::pty side welded capsule after ISO 6 on corner. I a 1 4
_ _ . _ . . . . ~ - - - - - . . . . . . . . - . . - - - - . . - . . . . . - . - - . - - -_ .-_. . . _ _ . 4 ! l I . l l i
. 7 i
s' , . , l . t A I i l l Fig. 11 End welded capsule vs.th steel inner after ISO 6 i.mpact on end I i l l l P i
.]
fig
- y
.r '( '
g 3 1 i 3 Fig 12 End welded capsule with steel inner after ISO 6 impact on side. l .I 1
s l* l 4 l
\*
l - j l 5
)
i i 1 l i ' 1 i 1 . J 4 d . t i i l 4 4
^'
t _ _ 4 l Re3c :; - i j j f ) Fic. 13 End welded capsule with steel inners af ter ISO 6 impac on corner l 4 1 1 4 i l 1 ! i 1 i i i
=
i*
.r l .
.,r-y 9 m-.:.
r
< s
}-1 l Avi i j Fic.14 End welded capsule with hollow aluminium inners after j TSO6 impact en end. N 1 ^
4 1 l
. t 1 . -i;#l ,
J.,- v p s
, ,q,
*~~
; -- 3
?
s '4
.}.'h _t -
"[ f '
i Fig. 15 End welded capsule with hollow aluminium inner after ISO 6 impact on side.
*y
- p. - .If ,
)
e 1
*l 6
l.s
- ., .a."
h, . ;- f,
. 2 4 % ..... G 1
Fig. 16 End welded capsule with hollow aluminium inner after Iso 6 impact on corner.
, . - . . - . - . . . . . . , - . - - . - . . - . - . . . - . . . . . . ~ . - . - . . - . . . - . - . - . - - . . . _ . . . - - . . - . - . . - ~ . - i i k= l t-i l - . . . . . _ . . . _ _ _ _ _ _ _ _ _ _ 1 . i i r . _ _ . i f i Fig. 17 E:::pty end welded capsules after impact of 14 Kg hamer on end. i l f 3) 5 i *
~
M' 1 i 1 l Fig. 18 E=pty side welded capsule after impact of 14 Kg har.mer ! on end. i
APPENDIX 3 } THE VACUUM BUBBLE TEST: A THEORETICAL AND EXPERIMENTAL ASSESSMDIT 4
- 1. INTRODUOTION
- 2. THEORETICAL ASSESSMENT 2.1 Ntaabe.r of bubbles produced
- 2.2 Sensitivity of test 2.3 Bubbling rates
- 3. EXPERIMENTAL RESULTS 3.1 Verification of formula .,
3.2 Investigation of void limits and senstivity 3.3 Suitability of liquids 4
- 4. PROCEDURE FOR MAKING TEST
- 5. CONCLUSIONS REFERENCES 4
l 4
_ . _ _ _ _ _ . . _ _ _ _ _ _ _ _ . . _ . . . - _ _..____.--_m _ _ _ _ _ _ - . . . _ _ . _ _ _ _ . . . _ . . ._m . . ._. . . . t LIST OF FIGURES I l . FIGURE 1a) APPARATUS USED TO INVESTIGA~E VARI ATIONS IN VOID SIZE b) FORCES ACTING ON BUBBLE BEFORE RELEASE FIGURE 2 RELATIONSHIP BE'NEEN CAPILLARY DIAME'*ER AND . BUBBLE SIZE FIGURE 3 VARIATIONS IN NUMBER OF BUBBLES PRODUCED WITH PRESSURE
'. FIGURE 4 NUMBER OF BUBBLES PRODUCED VERSUS VOID SIZE (ISO PROPYL ALCOHOL) i FIGURE 5 BUBBLING TIME VERSUS *?OID SIZE (ISO PROPYL ALCOHOL)
FIGURE 6 NUMBER OF BUBBLES PROOUCED VERSUS VOID SIZE (ISO PROPYL ALCOHOL) FIGURE 7 NUMBER OF BUBBLES PRODUCED VERSUS VOID SIZE (DIFFERE!ff LIQUIDS) , 1 l FIGURE 8 BUBBLING TIME VERSUS VOID SIZE (DIFFERENT LIQUIDS) , l FIGURE 9 BUBBLING PATI' ERNS .," , LIST OF TABLES , l TABLE 1 SURFACE TENSION AND LEAK TEST SENSITIVITIES TABLE 2 MINIMIM VOID SIZES TABLE 3 PHYSICAL PROPERTIES TABLE 4 FACTORS AFFECTING SUITABILITY FOR VACUUM BUBBLE LEAK '3: STING e 4 i J e G !
. . _ _ . . _ .-. . - _ _ . _ _ -___ .___.m_____ _ _ _ , _ -_ ._ _ _ _ _ . . . . _ . . . _ _
^ 1 APPENDIX 3
, THE VACUUM BUBBLE *EST 4
A THEORETICAL AND EXPERIMENTAL ASSESSMENT An experimental investigation into the effect of void size was undertaken and from the results the minimum allowable void size can be decided upon. We main theory of the test, including sensitivity, is discussed with reference to previous work. l
- 1. INTRODUCTION The bubble test .s known to cover a wide range of leaks and it is clearly important to know the upper and lower limits between which the test is
- applicable. With large leaks there may be insufficient void inside the source to maintain a stream of bubbles for any reasonable length of time. With small leaks bubbling rates may be too slow and escape deteption. Thare is a region in between where experience has shown that leaks show pp as a clear and unmistakable stream of bubbles. '4 .
The theory underlying the test, including its sensitivity is discussed with reference to previous work. Experimental work into the effect of void size has been undertaken to determine the minimum acceptable void volume. A further aspect of this work has been to investigate and recommend suitable procedures for carrying out the test.
- 2. THEORETICAL ASSESSMENT 2.1 Number of Bubbles b duced A knowledge of the theoretical mechanism is important as apart from improving our understanding'of the test, the influence of other parameters may be more easily appreciated.
The mechanism of the test is believed to be as follows. The capsule is submerged in the test liquid and the pressures above the liquid reduced to around 0.1 bars. Se capsule has a void V filled with air at atmospheric pressure, since this pressure is higher than external pressure air is forced out to form bubbles. The pressure inside the bubt,le P is related b to the external pressure F by Iaplaces equation. II Pb-P=g
. R Where S is the surface tension of the liquid and R is the bubble radius.
As bubbling progresses the pressure inside the void falls and eventually 1 reaches a cut off pressure when bubbling finishes. Se cut off pressure j Pg is given from 1). j J P - P = 2S I T , Where the R here stands for the final bubble radius. R ese equations have been checked experimentally by King (5) and found to be correct. j I
n e size of e bubble depends on the diamete. ,f the leak end a formula relating tha two is wall known. Consider Figure 1b, which shows a bubble just before release. The surface tension acting on the circumference of , the leak is balanced by bouyancy. 2trrs = 4_vr R3 ,3,9, 3) 3 Where r is the leak radius, R is the bubble radius, j is the density of the. liquid and g the acceleration due to gravity. Rearranging 3) and expressing in terms of the bubble diameter D in m, we get 4). " D= 0.85' 's.d.. 4I
- 3.
- Where D is the bubble diameter in a, S is the surface tension in Cl d is the leak diameter in a and j is the density of the liquid in Kg a . ,
Clearly from the results of 2) and 4) we can3 write an expression for the number of bubbles produced N in terms of the~ void sige V (m3 ). We have a known void filled with gas at atmospheric 3ressure. This leaks out in the form of bubbles at a pressure given by 1) until the pressure in the void reaches cut off as given by 2) . Se voltane of gas which escapes as bubbles at pressure P cis given by .5) . Volume of gas escaped = V (Po - P) 5) , P c Where Po is atmospheric pressure, P is the applied pressure. Se volume of ' the bubbles is from 4). Voltase of bubble v3 = 4 v (4.25 x 10"I)3 s.d. 6). 3 j Bence the number of bubbles produced is simply given by 7) ! l N = V(P - P) 7) Pc y 2.2 Sensitivity of Test From equation 2) we can see that when the cut off pressure Pe reaches atmospheric pressure, no bubbling will occur. Pc =P+2S 23 R For a given liquid and applied pressure this depends on the bubble radius ' R, and from 4) this depends on the diameter of the leak. In a normal test using iso-propyl alcohol (S = 21.7 x 10-3 Na~l j = 7 86 x 10 2 ' l' Kg m"3) atmospheric pressure at 1.0 x 10 5 g 1,and the applied pressure 0.2 x 10 5 y,-2, 105= + 0.2 + 10 5 R R = 5.4 x 10-7 m ! l l i .
~~ .-- . . - . - . . . . - . - - . . .-- - ~ . . . . . . ..-. - .
- . i The smallest .apillary for wnicn bubble growth .s_possible will have a !
diameter d equal to 2R, nence d min is 1.1 x 10"' m. l
]
A usef ul model is to approxim .e real leaks to regular cylindrical j capillary leaks. For Laminar gas flow Poiseuilles equation 8) gives the volumetric leak rate for a tube whien is long compared to its diameter. L= ( P,3 - P,) P ,r d4 x 10 ll 8) 64 ] l
'9 L = Leak rate in mbar 1/sec
{ P,3 P2 = Pressures at each end of capillary in bars
, P = average pressure in capillary and equal to (P3 + P,)
2 d = diameter of leak in m 7 = viscosity of ges in Kg m~l sec ~l 1 = length of ' capillary in m This represents laminar flow, for very smal[fleaks mclecular flow may be important when collisions between moleculeg become negligible compared with
; those with the wall.
Using the minimum diameter of the capillary we have already calculated ie 1.1 x 10~'m and typical values for the bubble test. If the capillary leak is to represent a real leak the capillary length should be around 1mm as most sources have a wall thickness of 1mm. Taking the viscosity of air as being 1.71 x 10-5 gg ,-1 ,,e-1, p2 = zero and P3= 1 bar (conventionally leak rates are for one bar pressure dif ferential) . L = (1 - 0) . (1 + 0) . (1.1 x 10-6)4 x 10 11 64 x 1.71 x 10 5 x 10 3 2
= 2 x 10~7 m bar 1/see This is the absolute theoretical maximum which is unlikely to be achieved in practice where real leaks are not smooth and cylindrically shaped.
LANGLEY (2) has suggested that in practice around-10-5 abar 1/see is a reasonable figure to take as the limit. 23 Bubbling Rate in our discussion of sensitivity no account was taken of bubbling rates and this is cler.:ly important as it may be the limiting factor for the sensitivity. Equation 8) is a volumetric leak rate, it takes no account of the resistance of the liquid into which it flows, therefore using it to predict bubbling times represents an approximation. Iangley (21 has checked the equation using glass capillaries and found that reasonable agreement is j found wnen the gas leaks into a liquid. Equation 8) can be used to predict bubbling rates quite simply. We have ; from 8) the vol Je of gas leaked per second at one millibar pressure. j
, Converting this to a volume at the pressure inside the bubble given by 2) and dividing by the volume of the bubble in litres we get 9) .
r
_ - . . . . . , ., - . . . ~ . . . . , - .- _ . . - . - - - . . - - - . - . . - . . -.. - - . - - No of bubblos ..r 9) cecond = Lg x 4 , b *b l . Where Po is one millibar and V bis the volume of the bubble in litres.
- Pb is the pressure in the bubbles given by 2) in millibars.
Substituting values for our leak rate of 10-5 mbar 1/sec. No of bubbles per second = 10-5 x 1.0 x 1 . i 1mtr t:t x 10 9 . > 4 4
= 1.7 '
l l ie Nearly tw bubbles a second which is easily detectable and hence the 4 limit of. sensitivity is not set by bubbling rate. i ! a ! Y . l . I i-1 a .l 1 J
]
i ' l i> 4 k 4 e 4
__ . - - - _ _ _ , _ _ _ . _ . - _ . . . . _ . ~ , - _ _ _ _ _ _ . . _ _ _ . _ - . . - _ . _ . _ . _ . . _ . _ . . i
- 3. EXPERIMENTAL RESULTS '
3.1 Verification of Formula
'f i
The formula - given in 2.1 4) l t D= 0.85's.d.'$ j - has been checked experimentally using dummy capsules, the results together with the calculated values are shown in Figure 2. The agreement is . considered to be reasonable. i
- i 2e formula given in 2.1 7) '
i N = V(P,, - P) P.V e b ' er Mis equation essentially. depends on two par-dr the pressure variation and the volume of the bubble. . The pressure variation has been checked using a dummy capsule of known void. The results are shown in Figue 3 - the points are plotted, together with 3theoretical prediction and show the number of bubbles produced per 10mm of void at a given applied pressure P. As can be seen agreement is good and hence equation 7) should be essentially correct. As a check the number of bubbles produced versus void size was performed on a dunsrf capsule and the results, together with the prediction according to 7) are shown in Figure 4. Again the agreement seems reasonable. 3.2 Investigation of Void Limits and Sensitivity
) 'tb investigate the theory associated with the test and principally to test th= importance of void volume some capsules were specially prepared. hey consisted of a . body of known void and a screw on cap with a hole of known dimension (see Figure 1a)). In this way a seriec of bodies could provide a range3 of voids for a con'stant hole size. Voids in the range 7m3 to 100mm were used with holes of diameter 0.'1m, 0.2mm and 0.4m. Rese were used te investigate the variation in number of bubbles produced with void size and bubbling times versus void size.
Upper Limit of Test The largest leak which can reliably be detected depends upon the internal void of the capsule under test. 3 The currently accepted minimum void volume 100m is given in ISO Technical Report 4826. Se ISO Standard 2919 Sealed Radioactive Sources - Classification specifies that the leak test methods in this report are suitable for assessing leakage of source centents. The Technical Report 4826 however, states tha.t if the vacuum bubble test is used for sources with free volume less than 100mm3 the user must be able to demonstrate the validity of the test. 1 One of the aims of this investigation was to establish that tests using . l reduced void volumes are valid under specified conditions. ! I i
. 1 There is no lin... to the size of hole that can be datacted givsn suf ficient volume, hcwever, prior to testing the capsule would be given a thorough visual examination and this sets a limit to the largest hole that need be .
detected. 'Ihe re is clearly a limit to what the human eye can see and when this has been decided upon, sufficient void must be present to detect holes l of this size. After some tests it was decided that holes down to 0.1mm diameter could be ' easily seen and for this reason 0.1mm, 0.2m and 0.4m holes were chosen for investigation. Tests however have indicated that much smaller volumes
- may be used even with large leaks. ,
Clearly the initial stream of bubbles must be maintained for some period of l time so they are not missed. Figure 5 shows a plot of void size versus bubbling time for holes of 0.2mm ando.4mm diameter. The number of bubbles produced is also a useful indicator and this is shown in Figure 6. A 0.1mm l hole is not shown as it is not appreciably different from the 0.2m hole. j a Reasonable criterion to choose to give the operator maximum chana of detection would be at least ten bubble produand in at. least five seconds. 3 From the graphs we can see this implies a void of at least 10mm for the 3 0 4mm hole and at least Sam for the 0.2m hole. . Lower Limit of Test The limit of. sensitivity of the test has been investigated by researchers in the pst. For example NIEMEYER 3) claims to ha measured leaks down to 4 x 10~ aba'r 1/sec, HAFF, NIEMEYER and ROBINSON I4 claim to have ' measured 3 x 10-6 abar 1/sec. Great care must be taken not to interpret I these figures at face value. It is normal to approximate real leaks in I capsules to a capillary leak, the magnitude of the leak depending on its length and diameter. For radiation sources most capsules have a wall i thickness of 1m and hence any capillary leak used in analogy must be approximately 1m long. The results quoted above were obtained using drawn glass capillary leaks much longer than 1m and thus are not really applicable. TAYLOR (1) and LANGLEY (2) have argued on theoretical grounds that the limit of sensitivity should be around 10-5 mbar 1/sec. A full discussion
, of the reasoning involved has been presented in Section 2. Clearly due to the nature of real leaks it is impossible to give an exact fi and until more experimental evidence has been accrued the figure of 10" mbar 1/sec seems acceptable.
3.3 Suitability of Licuids , There are many liquids which may be used in the vacuum bubble leak test and , their differing properties effect the performance that can be obtained in the test. In current standards glycol, water or " oil" are mentioned (BS 5288 1976, ANSI N542 1977). The latter specifies the oil as having a mass density of 890 kg/m 3 and a viscosity of 25 cst at 20 C and 90 cst
, at 50 C. In addition to the above mentioned liquids many papers on leak testing mention iso-propyl alcohol as does one early standard ANSI N5.4 l 1968. However little discussion has been found in the literature as to rhy the various liquids have been chosen.
l
.- . - . . . . - . - . - - . - . - . _ - . ~ - . .. . - . . - ..-.. .. . _ . . .
"h e experin. ental program:ne was devised to consider;
- Test sensitivity
- Minimum void size General convenience Test Sensitivity i
Calculations of the maximum sensitivity of the vacuum bubbie test have been discussed under 2.2 and 3.2. The sensitivity L depends on surface tension S in the following way: 4 LAS The surface tension of relevant liquids and the maximum sensitivities which can be calculated from them are shown in Table 1. Although due to the general dirtiness of real leaks such figures will not be reached in prac ice they clearly illustrate the advantagg. of a lower surface tension. Table 1 Surf ace Tension and Leak Teb Sensitivities Surface Tension (Nm~l) Sensitivity (mbar 1/sec) l ('Iheoretical) Water 73.1 x 10~3 Glycol 2 x 10-5 47.7 x 10~3 5 x 104
; Iso-propyl 21. 7 x 10 ~3 2 x 10'7 alcohol Minimum Void Size As de. scribed earlier, for the efficient detection of large leaks a certain minimum internal void is required. The criteria used for deciding upon a safe minimum void were as follows. Holes down to 0.2mm in diameter can be detected by eye, so a 0.2mm hole is the largest we have to detect instrumentally. To give the operator a good chance of seeing bubbles in a leak test at least 10 bubbles should be produced in around 5 seconds. The number of bubbles produced and the bubbling time depends on the volume of the void and this therefore fixes the minimum void volume.
The bubbling times and number of bubbles produced from a 0.2mm diameter leak for the three liquids tested are shown in Figures 7 and 8. 'Ihe minimum satisfactory void volumes can be read from the graphs, and is shown in Table 2. Table 2 Minimum void size 4 Minimum Satisfaetory void volume (mm3) Water 30-40 Glycol 10 Iso-propyl alcohol 5-10 7-
. _ _ _ . ._ . _ _ _ . __ _ . _ _ . .m . _ _ _ - _ _ . - -. .__ . ._ _. _ _
i Geeral Conysnience Under cis heading we will include all other factors which affect the performance of the liquids in the test, and their general convenience in use. For a g'.ven siz,e of leak the three liquids will show different bubbling patterr s. Due to their higher surf ace tensions water and glycol tend to Iso- - produr.e larger bubbles. 'Ihere are few of them and they move slowly. These , prop',1 alcohol produces a stream of small bubbles which move fast. dif ferences can be recognised in Figure 9. Af ter a source has been tested it must be dried. with its low boiling point (Table 3) iso-propyl alcohol is very volatile, so that it can easily Water is second best in this respect and be removed from the source. 1 glycol is the worst, particularly as it also has a high viscosity and is j very sticky and difficult to remove from threads and welds. If a bubble leak test is inconclusive the source is generally warmed gently before j ratesting. To drive any liquid out of the leak.-glycol being the most viscous requires most warming. , '8 . One main disadvantage of iso-propyl alcohol is .that it is a fire risk, being highly flammable. Vapour pressure is an important factor, as if it is high the liquid will boil as the pressure is reduced and many unwanted bubbles will be formed. Iso-propyl alcohol has the highest vapour pressure, but even so boiling It is does not occur until the pressure has fallen to about 0.02 bars. particularly susceptable to extraneous bubbling as dissolved air is released as the pressure is reduced this can be avoided by outgassing beforehand, but it is more troublesome with iso-propyl alcohol than with ,
)
the other liquids. Table 3 Physical Properties. Water Glycol Iso-propyl alcohol 198 82 3 Foiling point, *C 100 0 2.24 (15 Kinematic Viscosity (CP) 1.02 (20 C) 22 (20 C) 0.02 0.001 0.05 l vapour pressure at 20 CC l bars
- i l
. l l
l J l 1
; i I.
i
~8 -
4 l
. I 3 . Summary
-j- + }*' The three liquids. listed are the most commonly used and d.iscussed in connection S with vacuum bubble leak testing, but this should not lead one to neg!.ect the f
?
j. possibility of using other liquids if they. can be shown to have suitttble properties. f
+ Table 4 summarises the f actors which have been discussed .
E I i Table 4 Factors affectine suitability for vacuum bubble leak testing i'
. i
- l 1
, Water Glycol 1so-propyl alcohol '
i - sensitivity (abar 1/sec) 2 x 10-5 5 x 10-6 2 x 10~7' (Theoretical) i l { Minimum Void volume (mm3 ) 30-40 10 5-10 l n. I e
- t. Mak Indication Few bubbles, '4 Large bubbles Small fast-
{ but continu- which move moving bubbles j ! i ing for a very slowly j longer time [ i i
- . Cleaning Source after Easy to dry Difficult to Very easy to dry
'I4 sting dry l
j Flammability No No Yes Extraneous bubbling Not Not Can be a problem significant significant if not outgassed properly 1 ? a a -a e j e l 1 I
-9 l l
.._.___.m.___ _ . _ _ _ , . _ . . _ . . , _ _ _ _ . _ _ , _ _ _ _ _ _ . . _ . . . _ - . . . _ _ _ _ . . . . _ , . . - . . __ . . _ . . . _
t e . L 3
- 4. PRCCEDURE FOR US. . THE VACUUM BUBBLE TEST
! Ecuipment ! i j A glass vessel with a vacuum sealing lid and connection to an external vacuum ' . pump i.5 required. 4 , The liquid used may be iso-propyl alcohol, water or glycol. 1 . Test Procedure ' j
- r
$ 1. He source is, examined visually, using magnification aids where necessary, ' under good lighting conditions. The results of this examination are recorded. *
- 2. Ensure there is sufficient liquid in the vessel to cover the source to be i j -
tested to a depth of at least 50mm. l
- 3. Assemble the vessel and reduce the pressure abov$' the liquid to 15-25 k Pm f (0.15-0 25 bars) and hold for one minute to lowr the, air content of the ,
l fluid. Return the chamber to atmospheric pressure'. [ L l 4. Ensure the source is as clean as possible before placing it into the vessel ! I for testing. 4 5. Reduce the pressure to 15-25 k Pa and watch carefully for a stream of bubbles emanating from the source. Observe closely for at least. two , minutes.
- 6. If no'such bubbles are seen the source is considered to be leak free.
1 Notes I
- 1. He liquid is changed and the vessels cleaned thoroughly at regular ,
j intervals. l 1 i 2. If the source is to be retested issnediately it should first be heated
. gently on a hot plate for 2 minutes at a temperature of 100*C to drive out the liquid from the source. i
, 3. He source is always placed in the liquid in such an orientation that the most likely area to leak (eg the window) is clearly visible. Test sensitivity l VOID SENSITIVITY mbar 1/sec Iso-propyl alcohol 10am 'b 10-5 . Glycol 10am 3 10-5 Water 3 10-4 40 sun a 9 b 4 l 1 I
- 5. CONCI,USIONS The vacuum bubble test is a sw.ple, cheap and reliable way of testing for leak tigntness. Provided the test is used in conjunction with a visual examination
- it will detect leaks from the very largest down to around 10-5 mbar 1/sec.
j The minimum internal 3 void present to ensure, detection of the larger leaks must be between 10-40mm depending on the liquid used. The vacuum bubble test i ~' method can be accepted by users, regulatory and other authorities providing j , acceptable procedures such as those given are followed. 4 I a
~4 I
f 3 i i i o i 3 ,I O el' l
+e 9
e _. . -_ . _ . . .. . - . . . . . - . - - - . . . . . . - . ~ . _ , . . . . _ ~ -. -. .-. .. - . 1 l . l References-
- 1. TA YLO R, C.
- Leakage of water into source capsules stored under water.
Amersham International IPU report F2/18 1965. I 2. LANGLEY R. ; l Leak detection by the bubble method. Amersham international IPU technical memorandum 71/16.
. 1
- 3. NIEMEYER, R. -
Leak testing encapsulated radioactive souress. l ORNL-4529, 1972. .
- 4. HAFF, K., NIE1MEYER, R., ROBINSON, R.
Radioisotope source safety testing. . ORNL-4092, 1967.
- 5. KING, C. f tacuum leak testing radioactive source capsuqr. .
ORN1-3664, 1965. , 0 f f I t O f O 6 1 l l l 1 l 1 i
4 ll *- o J 1 1, l - 4
;\\\\\
!l, f_ '
.i a -_ _ _ _.__ _
~
.! i v
.,~ is , s s s -s s 's Vs \
1 4 i VOID i HOLE REPRESENTING LEAK i a a j Fig.1. a) APPARATUS USED 'IV INVES"'IGATE VARIATIONS IN VOID AND BOLE SIZE .l j 1 1 1 1 4 i , l 1 4 3 l 3- n Bouyancy = 7 "TT R j.g.
- l f
Surface Tensions Q\ \'\ \\\
\
s s s xxs N .,
\\-
\ \*
s, -
\\g\
g
., x Fig 1. b)
FCRCES ACING ON BUBBLE BEFDRE REI. EASE
. _ _ _ . - . . . _ . _ . . ~ _ _ . . .m_. _. . . _ . . ..._.m.___.._____.___. _ _ _ _ . . _ _ _ _ _ _ _ . _ _ . . . _ _ _ _ . _ - _ _ ...
BUBS:.E OIAMETER MM O M N N W bl 2 - . s . . . .a i & C L- O LA O m 1 I I I I ! F 1 ,x 9 .. ' 1
- .~
<O ,
= ,
a > f M e .C ,
's =
4 n M i 4 o .o . 4
=. .
> (n N
=
M * . T- ' 1
- te v
l
- o.
M *
.g
- O -
#+
O I i M 1 .O . . i a O - i 1 m tf in i
< 4 '
! O M i 4
< v 4
4 2 .e . 1 M Ln l E ,
- o . .
W l a .C g . W t* 4 tg ? 4 (n :lr m p @ l to y o te .o "@
*J *
< s $ b a e4 9
= M
! M i 2 Q Q
^
O i *4
- O
- 4
' Y ! M ta . x ed 1 ). o E a 1 1 M l 1 m 4 t 3_ W I i a 4 i ,e 1 , i 4
l >'4ji l,}8 i , ! '.' 1 3 1 6 3( 0 1 0 2 S 0 5 0 5
~
u F i 5 q 0 3 V T A R 1 I 0 . A 0 T I 9 9 O t 1 I t i 1 t . 5 J U 0 \y. M B E T s,, R h ,i e O o s* F 2 r 9 B 0 . e . 0 t U i B c B a L l E S
\
P 2 A R 5 O 0 '* D U _ C E
\ _
D _ W 3 I 0 _ Ti 0 . _ l _ A P O -
. P b -
L s I E 3 5 e .
. r D 0 .
v P e R . N d E S S N U R E 4 0 0
. % ~ - 4 5 .
0
- 9 P 5 .
R 0 . E 0 S S s U R
.E 5
5 n 0 u n I/ gl N
~
t
2, / j 4 1ii4; iii I I , ' : <1 ii4 j;]) jI,3 i^' l I 4 tI j2l4
+lfl.i,I.
F 1 2 3 a 5 6 7 3 , g i 0 0 0 o 0 0 0 0 4 N i _ _U . M O .0 B b E - s 2 . R 1 e 0 p ) rm O vm e F dd B i U Ta . B h . B eh L 2 oo . E 0 } r l S . ee . t P i R c O a . D f f l U . C - E 3 D 0 I V E R S _ U S ;h _ - p V 4 , . O 0 . I D - S _ I Z
,E 5 0 .
T i l E O . R E T I C 6 A 0 L A N D . O To b B 7 _. h e s .0 S 0 , . E oe4 . R . r r V e vu E _ t ea m D i d c d ( a i l I S 8 .
.a -
O 0 h P o R l O e _ P _ Y _ L _. A 9 ,. L 0 C _ o l l _ O . - L
)
V _. - O 1 g I 0 D 0 _ . n u n . 3 '
> 4 t
s=.m w .=. =*,i-
. e s v m. g
._ .e ..a 1 i
e e,, emo. w W M l l e 1 1 i 9 g w w \\ N l 4 O .O A x N
@ S
, ~
i * - m s a m c s L*4 tT N - g s O CT C .
- t a.- -=
m n O 3 t 4 [6 w I 1 N \ m W d i
* \ i
> s u \
- fD o \ -
l O i O-e
< \
s O M \ I
~ m i
- C h
< O .
O a r r a y ' 1 I
>~
N .O.
.G g tn .
O' 1 O 0 , ' 1 a w M m N
=
D O.
; O
, a l I
.O h
3 a .a ' O . . l O r w I O , O l 9 I. @ O
'e l
O - w O O
.c Y I , i .g ,.
l .ksJ l e ' 1 1 9
_ _laJA- d. h e h.a.au.a--mam..& -8 h ,i AtM s .AA 4.s e 4.h e m.h .W.mM=J--es.4%
++
'3++--4-#.a,f E-.41# SeJD- e a ,14 MeEW E WaM 4 h4h M4 A 64.43_.I_. s4.-Jl- J. e_ ama e. .L.JA ,.; Mm. g A5g4 ugd O h O h h a G&W Mg g e
a j W d W i O o C C C C C C l
, , a
(
- I t
- w. . . *..
t N
~
.O
- . N l
c n = l C i i 4 * ===9=== 3:1 ! I ! C, n 'r
+ g
- i
\ .
a r \ e ,
* , i
, M
- t @
Q " * [ Z O
=
O
- N '
l .O o -*-- t- ' w m Of LsJ '
- c O -
c. O O **
- 7 9 n ee ,
i g l$ , r 1 g O ,. F m 5 4 m m * - i Gn O b . . r De o O f O C -
- O A b @
t i lll: $ C, O m U1
- w m O- 9 1 e
- 8 m ,
s a C w 0 n ~ O 2.
% i M (A m '. ,
l O
- O T N .
*< d W e }
a w O g =J Q ' w p f O= 9 O w 0 I w @O,
- e a
4 w O " O
' e i
l
< i I.
? r 9P J
, ,, .- , ,- , ,. ~., . . , , -. ,. . , - ,,
.mem ,m. --
5 e , f E 5 $ E g g x y -
\, ===
l l N b I M
~ \
h ,' \'s I >- O i F !a
- =
I 8?
=m W
N O
. \
hh5 l #
- P m l T !
O ' C u . C O
- e .
m .- e u . (!) O 4 O ts
- P O tn o
- O
- o N r*
m b & l e 2 O O g ai M i
% 1 o l i
l g as , *N
",,n .
.f.< .,,,-
8; ,. :*, s, m r. a O
- e .
O Q N m W O O a , 1 1 1 I
.i I l l I l
Y
4 j 41} aii4)i 2lf1 )i 14i, ti1 < 4y >-lt.^iil$ 4 , .T:4
<l<.i 4 ii ia11 i ] 4i E.i (, n,. , d 4l,
- 4 1!
m' bnf.y
,,,;7o e 1 ii!.{ .
.. 1 8 1 1
,. 2 6 2 4 0
a F i g . 8 1 0 - [ fr U U B
/ ,
B _. L Z . N G
.2 e/
T 0 . _. I . H S , . V
- E R
S - U 3 . S 0 . V O I
~ D ,2
,. S I 4 , F '. Z - 0 E 0 .- . 2 , a u 5 - ._ s 0 -
~
. D .
~ I - .-
A - - I l O - L 6 , o E 0 - e _ m. _ m. m e 7 , _. m 0
, I S
O 8 . 0 P R
. O P
Y
- G L
A _ 9 W L Y
- L 0 A - - C T - C V _ E - O - O l . O _ R - L i O '
I _ L D _
- / _ .
s u n 3 0 , 1 0
f f f i water ric; . 9 . S t:. colin; .tterns for *se-prcpy; alconci 1
- j. Shows a capsule with a hole of C.2 c= d:.a .eter. '"his is being vacuu: bubble tested in water ( cp photograph) and shown ;
)
j ] iso-prepyl alechol (bott c photograph) , i i i i e . 4 I' -
. - g~ ... a
.~
- .f- ['* Q:J,-
*- 4 h , , .
' f[' .43ff" l {j p lh
- d*g c '*A . j y
'4 g i lf6 4 '#. +1 : , , . . _
h l.1k,b. $ 5 h, K3Y;'?.M [ k'NM4, r?- D$3. I $1- kc,S
,e . -p .
k i.,m .
~
6,. 4 ., .,, S .
. . ' N. .
h I
;r &., .3W ~vy;tq. m. .i .bt ,.4. ;f'[n'd
# ,.. i ,e .
.s'. + .S.c.'
n i n
! '? i f'&., -l* - .h&$$f ' j. . ; _
7.. .
$y.... ,2. , ,.y~[~i
';..k [ ,I. ^y ;
k &.pD .. rn'..'
- 1
^ 'k.
['.' ~
.,s. 7 'si~. g.
e,. j ht:~
. 3'}: .
- .; v.%fyc_ '~ .
Q.y&;Y&.~ .
;+.. , jj;&y
x*-
~
;. . l p.y ; r; . . a
- ~ &, )
l ..).' _ _ :.2.ls . 'j Q'. - l4f c.;.Q..e L fR}A.
- t f . 4 ...gcw, ll ,'.' .;; ~-%;N-;
- . ' 2 , , ;. -f. ..
lfy.. lr.. j ,Ll . yy7 *f . ?L'. ~~- f^'g;;, T
%'k - >
1
.., '*N v..
* .?.
; ;f. cgo:w;-c
~-. b: ;h , &y ,...'.3; [...'m..
s :t w -;.. - f ..~:~,3 _ s .:.":;y . ~..'. :A . y. i 1 I l> 4
" 1 i
g ** 6gc :".Q%.. .g. ,.
- [a h.D+-
j u c;.:
' o x. e; . n fa m .
< . 7, - ~ .-
- -4 { '* *f N . ef.;., f','.
I .N . i' g(8 if. . ,,c,,'*
. . !* * }.
bii . S']1 lmP. . .. $ y
.n.< s .. ,. . Po. :p.c,- s*;t A j p;y ;
,-.. ] 'f.%'s . y' y t *'{,3 :{:.
3.6. .
'5:~
. %,r4. M...k1"*u . .' f. ,f',( '$g N % !E..,a -
I.*** # ggg7
"'m; g. p. ', fEs.
.* a # /
*.p ,,
...,.. 4' "
l$":['*',?
,;f : -
r
- 8 o ' .,'@0,.6-
' 4'
.,. c f .
\3:,if.r&o we y. V~ "
k A. b' -- Jii=""Q,.
.: 6.3...ti -
] ' s {
, ;s2 rs
- r. . . T-
- s. . . .*
*/.
u . 3 +.. .4 M^h.h-i f,* ,*
%, ' **le .
[Jg,N. * . 3: Y7'i.h"*%.Q' -. ,O E-F*. 4* 6'*L- < J /.- Q i % k, 5%*. . s__ : ' m.s . e
- s. s..E. U.
a w .1.t a. w,. e
., O W,.. k.U..,N. , $.Wh.%.,'.i.1 V.M v2 w.... ~-a:A ~ w. - w' sf o .
,. ; ws" t
l 1 e i
.-.- - . . - . .. ..- . -. . - . - - . - . - - . . = . . - . . . _ . . .... --- ~.-- .- - - -.
i APPENDIX 4 . 1 THE PRESSURISED BUBBLE TEST l ! 1. INTRODUCTION
- 2. THEORETICAL ASSESSMENT .
4
- 3. EXPERIMENTAL RESULTS .
- 4. PROCEDURE FOR MAUNG THE TEST I
- 5. CONCLUSIONS LIS'I OF TABLES TABLE 1 PRESSURISING CONDITIONS f 4 _:
G
= = 4 e e
G 9 6
.. -- - . -- ~. --~ - .. .. .
APPENDIX 4 THE PRESSURISEO BUEBLE TEST
- 1. INTRODUCION The pressurised bubble test consists of placing the source under test in a high pressure of helium, removing it from pressurisation, and immediately afterwards carrying out a normal vacuum bubble leak test.
. The pressurising process increases the gas pressure inside the source. This may greatly increase the sensitivity of the test, compared to that of the normal vacuum bubble test.
- 2. THEOREI'ICAL ASSESSMENT In section 2.2 Appendix 3, the sensitivity of the vaguum bubble test was calculated. The calculation is easily modified to, Include an internal pressure of greater than one atmosphere. The sensitivity of the te'st depends on the internal pressure, and in principle by using long pressurising periods we can arrange for this to be as high as desired. We shall choose a sensitivity of 10~7 mbar 1/sec and calculate the conditions which this requires. The reason for this choice is that leaks smaller than 10~7 mbar 1/sec are unlikely to
, exist in practice, see Appendix 2. I J A leak of this size, as discussed further in Appendix 6, 'The Helium Pressurisation h st', is in the molecular flow region. For molecular flow through a long capillary we have: 6 Q = (P 3- y 2) 8_ 3 [ 1 wM x 10 2m (3) j Where Q is the leak rate in abar 1/sec, P1 and P2 are the pressures at each i end of the capillary in bars, r and 1 are the length and radius of the capillary in metres, T is the temperature in degrees Kelvin, and m is the molecular mass of air in kg. i For a pressure difference of one bar and a capillary length of 1mm the radius required for a leak of 10~7 mbar 1/sec is found by substituting in equation (1): r= 10-7 x 3 x 10-3 x 10-6 x 2 x 28 x 10~3 1
/2 1
/3 8
"" /
$ = 4.6 x 10~ m 4 l Prom section 2.2 Appendix 3, we have equation (2) from which we can calculate i the internal pressure required: a o I P = 25 + P (2) 4 3 Where P, is the cut-off pressure below which no bubble formation occurs, in Nm-2, S is the surface tension in Nm~I, R is the bubble radius in m, and P o is the pressure above the liquid in td 2, l _1_ t J
i l Substituting the values for iso-propyl alcohol 5 = 21.7 x 10"'tC I and the radius 1 I calculated frcun equation (1), and taking at:nospneric pressure as 1x 10 5 h-2, ,
]
l we haves 1 P C
= 2 x 21.7 x 10~3 -7
+ 0.2 x 105 ,
)
4.6 x 10 ; i l = 1.13 bars , We can now calculate the conditions required te produce this pressure inside a ~ source capsule. For a capsule of void volune V = 100mm3 and a leak L mbar 1/sec
- the internal pressure is given by equation (3) as discussed in Appendix 6.
I P=Po [1 - exp ( - 2.65 1 t3)] exp ( - 2.65 L t2) (3) l V V J l - 1 Where P is the final internal pressure, P iso the pressure oS helium used ("pressurising pressure"), t3 is the time under prefsure, and t2 is the time between the end of prenaising and the time of tige test, all tsmes being in
- ~ -
seconds .- u -: r. ' Assume that the vacuum bubble test will be performed within half an hour of ; pressurisation, so t2 equals 1800 seconds. Most sources have voids less than ) 50 nun2 , so this upper limit is taken as the void volume V. If sources with larger void volunes require testing ge pressurising period t 2 must be increased in proportion, ie for 100mm void volume the time must be doubled. ; We require the total pressure, air and helitan, inside the capsule to be 1.13 bars. During the pressurising process some of the air inside the capsule at atmospheric pressure will leak out. However, for the small leaks considered here this leakage is less than 1s. We therefore require a partial pressure oi helium cf 0.13 bars. We can now choose a. convenient pressurising pressure and l j solve equation (3) for t . 7he results are shown in Table 1 f ar three g j pressures: 1 TABLE 1 Pressurising conditions l i Sensitivity of 10~7 mbar 1/sec ' void voltanes up to 50mm 3 Test liquid = Iso-propyl alcohol i
, 1 ; Pressurising Pressure (bars) Pressurising time t., (minutes) l i 6 70
- 12 35 ,
34 15 , So f ar we have used helium as the test gas, but there is no reason why other gases, sich as nitrogen, uhould not be used. The calculation for pressurising time will need to be modified, however, In equation (3), the factor 2.65 is replaced by Jm 3/m2, where m3 is the weight of air (assumed to be 289). l , substitution for helium, which has a m.w. of 4g, yields the original factor 2.65. Substitution for nitrogen (m.w. 28) yields a faction 1.0. Hence if nitrogen or air is used as test gas we expect to use longer pressurising periods. I l 2-
_ _ _ . . _ _ _ _ _ _ _ _ - _ . _ _ . . . _ _ . _ . . - _ _ _ . . _ _ _ _ _ . . _ . .~. _ - ___ i The vacuum bubble test is performed using iso-propyl alcohol but there is no i reason why other suitable liquids such as glycol should not be used. Iso-propyl From ' alcenol is recommended because of its low surf ace ' tension, see ref (3). equation (3) this can be seen as a distinct advantage, because it requires a l ' lower internal pressure. If other test liquids are used the calculation of the j pressurising period is easily modiiied. As other commonly used liqu..ct have a . 4 higher surface tension than iso-propyl alcohol the resulting pressurising *M es will be longer.
- 3. D:PERIMENTAL l
To support the above theory tests were performed using artificial leaks. Short l , lengths of thick walled copper tubing (I.D. = 1. Osm, O. D. = 1.6m) were cut and j ; the end 2-3mm squashed in a smooth-jawed vice. The other end of each tube was sealed usina a crimping tool. About a dozen such leaks were prepared. They
- r l
l were calibrated using a helium mass spectrometer and found to be in the region 10-6 mbar 1/sec'. The leaks were vacuum bubble tested in iso-propyl alcohol l and no bubbles were observed. hy were then presgurised in helium for 15 ] minutes at a pressure of 34 bars. On removal framr the pressure chamber the leaks were bubble tested (again using iso-propyllicohol) .' A clear stream of i j bubbles was observed in all cases. 4 i 1 1 4 i i - . l k ) i t I I . 1 i b* . 4 J i t
. _3 I
N l
. I 4
l
- 4. PROCEDURE FOR MAKING TEST .
j EQUIPMENT As for vacuum bubble test plus pressure chamber. ! TEST PROCEDURE - . 4
- 1. 'Ihe source is placed in the presrure vessel and pressurised with helium ,
at one of the following convenient pressures for the specified time. , Pressure / bars Time / mins . e 6 70 j 12 35 ', 34 15 - .
- 2. Within half an hour of removing the capsule fh the pressurisation chamber a normal vacuan bubble test is perfdimed as described in Section 4 Appendix '3.~ ,
I
)
TEST SENSITIVITY . The' test shall be regarded as having a sensitivity of 10-7 abar 1/se'. c 1 i
)
i I 9
~
l i s-e e
- 5. Conclusions The pressurised bubble test is a convenient and reliable way cf testing sources to sensitivities as hign as 10 mbar 1/sec. Its main advantage over other tests, suen as the helium pressurasation test, is the wide range of leak s;:es it will detect, ie it has the range of the ncrmal vacuum bubble test ( f rom leaks of about 0. 2mm 7,lameter that can be detected visually to 10~5 mbar 1/sec) it will also deter.t acall leaks of 10~7 mbar lesec. The apparatus required for the test is simple and cheap. The test not only indicates a leak is present but it also shows what part of the source is lear.ing.
I I a 4 . l l l l i i l l l l 1 i 4 6 APPENDIX 5 LIQUID NITROGEN BUEBLE TEST i
- 1. INTRODUCTION '
- 2. THEORETICAL ASSESSMENT ,
- 3. EXPERIMENTAL RESULTS 3.1 Minimum void volume 3.2 Sensitivity
- 4. DISCUSSION
- 5. PROCEDURE FOR USING THE TEST
- 6. CONCLTISIONS ~4 'O
l LIST OF FIGURES FIGURE 1 BUBBLING TIME VERSUS VOID SIZE FIGURE 2 NIMBER OF BUBBLES RELEASED VERSUS VOID SIZE ! l LISTS OF TABLES TABLE 1 COMPARISON BETWEEN LIQUID NITROGEN TEST AND THE VACUUM BUBBLE TEST e 6 o S e
1 l APPENDIX 5 THE LICUID NITRO"iEN BUBBI.E TEST h 1. INTRODUC'" ION l The source to be tested is completely immersed in liquid nitrogen (ANSI N542 l 1977 specifies five minutes) . The source is then transferred to the test liquid (normally methanol). The presence of a leak is indicated by a visible stream of l bubbles. l ,' Liquid nitrogen has a low viscosity and this should give the test a high ! sensitivity. The expansion ratio of liquid nitrogen to nitrogen gas is very j high and this should prove useful when testing sources of small internal void. $ 2. THECRETICAL ASSESSMENT i The mechanism of the test is as follows. The sourge contains a void filled with l air. As the source is cooled in the liquid nitrqpen the air cools and contracts s and liquid nitrogen is drawn into the void. . When the source is.placed in methanol at room temperature the nitrogen vaporises and escapes as a visible i i stream of bubbles. i l l' The large expansion ratio of liquid nitrogen to nitrogen gas makes the test ideal for use with small void volunes. his is easily calculated as follows: . volume of one mole of liquid nitrogen = m = 28 x 10
~3 = 3.46 x 10-5,3 2 ~
I 3 8.08 x 10 where j is the density in kg m-3 and m is the molecular mass in kg. j . i l W e volume of one mole of nitrogen gas at 77 k is
= 77 x 0.0224 = 6.3 x 10-3m3 i 4 273
! .'. expansion ration s 6.3 x 10~3 = 180 3.46 x 10-5 1 l For the largest leaks we expect to detect by this method the volume of the 3 3 . bubbles produced may be around 1mm . Hence for Insn of liquid nitrogen drawn into a leak around 180 bubbles will be produced. This implies that the test could be' used with extremely small voids. However, we cannot guarantee {! i that a given void will become filled with liquid nitrogen. Even taking this , into consideration void volumes as small as imm3 appear to be quite adequate. I i The sensitivity of the liquid nitrogen bubble test is believed to be very high. j , An estimate of it can be made fairly simply. The differential pressure to move a liquid down a capillar'y is given by equation (1): ! 1 l
- 4 P = _2 S (1) i r e .
- whereap is the differential pressure in Nm"2, S is the surface tension in j ten ~1 and r is the radius of the capillary in m.
i . The differential pressure is easily found since the air inside the source cools to liquid nitrogen temperature (77 K) and the pressure falls accordingly. From the ideal gas laws we have i l P= 77 x 1.0 x 105 = 0.281 x 10 5 y,-2 l '73 1 l < l I
..~ - - - - - .---.~ _- _--__ . _ - . . _ . . - . - - - ~ . - . . . - . . . The pressure above i liquid nitrogan is still et one at:nosphere pressure ' (1.0 x 10 5 Nm-2), hence the differential pressure is given by: P = (1 - 0.28) x 105 = 0.718 x 10 5 Nm ~2 substitute into equation (1) together with the surface tension of liquid nitrogen 10.5 x 10~3 Nm" at 70 K) to find the minimum radius of capillary I the liquid nitrogen will travel up: r min = 2x 10.5 x_10~3 = 2.9 x 10~7m 0.718 x 10" . i
. l Substitute into the equation for molecular flow along a capillary (2). Thir is ;
discussed further in Appendix 6. Se Helium Pressurisation Test'. l 6 Q = (P3-P) 2 8- #-
/w RT' x 10 3 1 / 2m Where Q is the leak rate in m bar 1/sec, Pg and P2 a e the pressures of each end of the capillary in bars, r and 1 are the lengn and radius of the capillary, and a is"the molecular mass of a'ir in kg.'
For a capillary 1 ann long (average wall thickness of a source) we have Q = 8, (2.9 x_10~7)3 x 10 6 [wx8.314x300 3 10 ' 2 x 28 x 10 d
= 2.4 x 10-8 abar 1/sec This represents a theoretical maximum which is unlikely to be achieved in practice.
However, the sensitivity we would expect will still be in the region of 10~7 mbar 1/sec. 21s represents a very high sensitivity in comparison with the sensitivity of the vacuum bubble test.
- 3. EXPERIMENTAL 3.1 M4 n 4 == void volume As was discussed in appendix 3 The vacuum Bubble hat, the combination of a large leak and small internal void poses aroblems, as few bubbles are produced in a short space of time and may yyily be overlooked. This imposes a restriction of the smallest void we may use. For the vacuum bubble test the minimum void was determined in the following way. Holes down to 0.2nen can easily be seen by eye, which is therefore the largest hole we need to detect instrtamentally. To give the operator a good chance
- of detecting a leak at least ten bubbles are needed in around 5 seconds.
These criteria coupled with a 0.2 nun diameter hole imply from the i experimental work that, in the ordinary vacuum bubble test, the void should be at least 10 nun , 3 l The gsuph in fig 1 shows bubbling time versus void size for a 0.4 nun
- diameter hole, for the liquid nitrogen bubble test (methanol as test i liquid) and the vacuum bubble test (Glycol). Fig 2 shows the number cf '
bubbles produced against' void size for the two tests. As can be seen the - nitrogen test is far more efficient at detecting large leaks, and even in a { i defect as large as a 0.4 nun hole an internal void as small as 1-2 nun 3 ,,,,, ; amply sufficient. , l e
, - _ _ _. _ . . _ __ _. _ .. _. _ ._.- _ _ _ _ -. _ .__. _ _ _. _ _ _ _ _ .. ~ _.-- _ _ ._ _ .m_ _,
i I
- . i
- 3. 2 ' sensitivity l A simple check on the sensitivity of tne test was made in the following way. Short lengths of thick walled copper tubing (I.D. = 1.0mm, l O.D. = 1.6mm) were cut off and the end 2-3m squashed in a smoothjawed )
vice. The other end of each tube was closed and sealed using.a special i ~ crimping tool. About a dozen such leaks were prepared. They were I calibrated using a helium tuned mass spectrometer and found to be in the I ] region 10-6 mbar 1/see to 10~7 mbar 1/sec. A liquid nitrogen test was I
- performed and all the leaks showed a clear and unmistakable stream of j
- bubbles. From these results we can conclude that the sensitivity of this 1
) , test is at least 10-7 abar 1/sec. ' j 4. DISCUSSION The main advantages of the liquid nitrogen test are its high sensitivity and I tha:, it can be used_ reliably with very small void volumes. A comparison is made between the liquid nitrogen test and the vacuum bubple test in table 1.
.7 TABLE 1 Comparison between liquid nitrogen testiand the, vacuum bubble test Licuid nitrogen test vacuum bubble test Test liquid methanol iso-propyl Glycol water alcohol ;
Practical 10-7 10-5 10-5 10~ ' ) Sensitivity abar 1/see Minimum void 1-2 5-10 10 30-40 volume : required mm 3 i C1hrity of test Bubbles small fast Large Few unmistakable moving bubbles bubbles j bubbles which continuing i move for a slowly. longer
. Liquid time j must be i outgassed properly otherwise-extraneous bubbles
; produced General Relatively Convenient Convenient Very
. convenience inconvenient but but convenient as a constant flammable difficult
' supply of to dry liquid nitrogen is
. recuired
m . ___ ..m_ . .._._.-- __. _ _____..._. _. _ ..__ _ _ _ __ . . . . . _ . ~ . . . - _ . - h t a
- 5. PROCEDURE FOR USING THE TEST '
} EQUIPMENT . i
- Dewar vessel containing liquid nitrogen. l
! Large boiling tube containing methanol. l
- TEST PROCEDURE -
3 l
- 1. Se source is examined visually, using magnification aids where necessary 1
{ and under good lighting conditions. The results of the examination are i recorded. e
- 2. Ensure the source is clean and dry.
1 ! 3. Place source in liquid nitrogen and leave for five minutes.
- 4. We source is removed from the nitrogen and immediately placed into the tube containing methanol. Ensuresufficient14quidispresent to cover
, source to a depth of 50mm. 1 j 5. He source shall be left in the methanol for two minutes. The presence i of a leak is indicated by a stream of bubbles. i NOTES
- 1. Ensure the liquid nitrogen is clean and free from ice.
- 1 1
- 2. Should a source require re-testing it shall be warmed to room temperature j beforehand. i '
i
; 3. The source is always placed in the ligsid in such an orientation that the i j most likely area to leak (eg window).is clearly visible.
d ! 1 a TEST SENSITIVITY The absence of bubbles indicates a leak rate of less than 10~7 mbar 1/sec. { Void volumes of at least 1-2mm3 are required. i a . j I c . , 4 S e
- 6. CONCLUSIONS I 1
i The liquid nitrogen butele test is superior in all respects to tne vacuum bumble test. It is more efficient at detecting large leaks, and internal void volumes l of only 1-2mm3 are sufficient. It has a very high sensitivity, 10-7 mbar ! 1/sec. Leaks show up as a clear and un=istakable stream of bubbles and there j are no problems with extraneo s bubbling as with the vacuum bubble test. Care j however must be taken to enstre that the sour e is dry, otherwise ice may form l and block the hole. Its main disadvantage is that a liquid with a low melting point has to be used, such as methanol (m.p. = -97.5 C), and these are invariably inflammable. Glycol, f or exar.tple, which melts at -13. 5 C, cannot be used because if freezes when an object at liquid nitrogen temperature is e placed in it.
)
i l l
.e I
~4 4
S 9 1 l 5_
^ . .
Fig.1. Bubbling tism varcus void size, e 80 leak diameter 0.4 nun leak length = 1.0 nun 70 ; Liquid nitrogen test .- 60 - _ 4
~
p-0 *: a 50 y^ 'p 18 # o . . a ii /
- 8 40 . .
? !i
.j __ 30 a m. Vacuum bubble test (Glycol) 10
/
/ ~~ *
- 10. 20 30 40 50 60 70 80 90 1DO 1i0 120 VOID SIZE, nun 3
. .- _ ..m_m .._m_____. . _ _ _ _ _ , _ _ _ _ _ _ . . _ _ _ . _
5 t i ,
- APPENDIX 6 i
THE HELIUM PRESSURISATION TEST . t (
- 1. INTRODUC"!ON
- 2. THEORY i
2.1 Leaks and leakage rates 2.2 Flow regimes , i 2.3 Predicting standard leak rates from helium mass spectrometer , I measurements i 4 2.4 Imak blockage 4
- 3. EXPERIMENT 11 3.1 Introduction y l - 3.2 Preparation of artificial leaks ,,
3.3 Effects of transfer time . 3 3.4 Surface helium levels j 4. DISCUSSION l J 5. PROCEDURE FOR MAKING TEST i f 6. CONCLUSIONS
' APPENDIX A Derivation of Equation 3 1
j RErERENCES i i! ' 1 i 4 4 i ? 1 i
- t i
l b 1 i i 2 4 4
-, ,v- 4 y , - - - , , -
f'ig . 2. tiumber of bubble released versus void size. leak diameter 0.4mm leak length = 1.0 wn 1 10 4
' Liquid nitrogen test 100 'iq t
60
.r.
60 10
;io Vacuum bubble test (Glucol)
, ' ^ ~ ~ -.. . . - - - ----->
d0 b0 60 70 80 90 100 110 120 130 10 20 3d 3 VOID SIZE, mei t I 4
1
- 1. INTRODUCTIO!i In the behum pressurisation test the capsule under test is placed in a' s. table enameer and a high pressure of hehum is applied. This is termed the 'bor,ing pressure'. Af ter a period of t.me, ter:,ed the ' bombing time ' , the - capsule is ,
removed and placed in a helium-tuned mass spectrometer. The time delay between removal and making a measurement is termed the ' transfer time'. If the capsule has a leak, then during the bombing- time some helium will pass through the leak and into the capsule. On removal from the chamber this helium leaks out again , and this leakage will be detected by the mass spectrometer. i
- l
, During Special Form testing it is envisaged that the leak testing of a capsule will proceed in three stages. Firstly, a thorough visual examination is made to I detect large leaks. Secondly, a vacuum bubble test is done to detect leaks of intermediate size ; and finally a helium pressurisation test is done to detect any very small leaks.
Clearly, if such a testing programme is to be effeep ve there must be adequate overlap between the three stages. The overlap bet #een visual and vacuum bubble has been discussed in Appendix 3 in which it is ,slown tha1i the vacuum . bubble test can have a sensitivity of 10-5 mbar 1/sec. The helium pressurisation test should therefore be designed to detect leak rates up to at least 10~4 mbar 1/sec, to provide a reasonable overlap. The problem with detecting such relatively large leaks is that during the transfer time the helium leaks out so rapidly that there may be none left to detect by the time the measurement is made. In theory the sensitivity of the helium pressurisation test is very high, but practical problems limit this. The principal difficulty is surface helium, le helium which during the bombing time has become absorbed onto the surface of the capsule. When the capsule is removed from the pressurisation chamber this helium diffuses away from the surface and the spectrometer indicates a spacious leak. The capsules presented for helium testing have already been vacuum bubble tested so thers is a possibility that any leaks present will become clogged by the test liquid. As the helium pressurisation test incurs using a high pressure of helitan this may serve to unblock such clogged leaks. All these factors must be carefully taken into account and a procedure for the , test designed accordingly. e S en 1
- 2. THEORY .
2.1 Leaks and leakace rates The terms ' leak' and ' leakage rate' are sometimes used loosely, and this can lead to conf usion between them, and to a lack of understanding of the mechanism of the test. The leakage rate' is simply the reading we may record or. our mass spectrometer. It is expressed as the quantity of gas detecteo in unit time. mis quantity can conveniently be expressed as a volume of gas at . stated pressure and temperature. In this report we use the unit mbar 1/sec, at a temperature of 293 K. When quoting leakage rates the test gas used must be stated, the most commonly used one being helium or air. If a capsule pos'4 esses a leak, we quantify this in terms of the rate- at which gas passes through it under specific condit. ions. The conditions sped fled refer to dry air at 25 C, with a preyttre difference across the lea 2 of one bar. 21s is perhaps best referred to as the standard leak rate. Thus, for ex eple, a capsule could have a standard leak which we quantify as 10-6 mbar 1/sec under the above condition, but during a typical helium pressurisation .est the reading on the spectrometer might indicate a leakage of say 10~7 mbar 1/sec. For a given standard leak the leakage rate observed wi.1 depend on the pressure inside the capsule at the time of the test. his nill depend on the bombing pressure, bcabing time and transfer time. In the next two sections we relate gas flow rates to pressure differen:es, and then to the condition applying during typical tests. 2.2 Flow regimes i As stated in thre previous section the leakage rate measured with a given leak will deper.d on the test gas used and the pressure difference across , the leak. 21A variation with pressure and test gas will depend on the magnitude of *.Jae leak, since two distinct flow regimes may occur, molecular flow and lam uar flow. Molecular " low occurs when the cross section of the leak is significantly less than the mean free path of the gas molecules. Collisions between molecules are then rare compared to those with the wall. c, , A*. A- Laminar flow occurs when the diameter of the leak is greater than or of the , T,y order of the mean free path. Collisions between molecules become more
'"-, frequent, and the viscosity of the gas beccmes an important factor.
[L , spqq . Equations which quantify the flow rate for stated conditions for pure Q molecular or pure laminar flow are well known. These depend on the
' ph" geometry of the leak considered. We shall consider a smooth long b ; cylindrical capillary whose length is at least fifty times its diameter.
gf;differenceacrosstheleakandisgivenby: Fur molecular flow the flow rate Q is direc 3 6 g3)
, Q = (p3-p2) 8 r Y RT x 10 if : 3 1 2m l'e d r am e
}*
1 For laminar flow the fitw rate is dependent on tne dif f erence of the squares of-the pressure:
*
- 11 0 (2) '
Q= v r (p'-P[) x 10 8}l Where Q = leak. rate in abar 1/se r = capillary radius in m .
. 1 = capillary length in m
~l
'7 = viscosity in kg m~l scr P,$ P 2
= pressure at each and ::t capillary in bars T = temperature in K m = molecular mass in kg i R = gas constant in J mol~I K~I A point worth making about the equations is thap suppose a capsule contains helium at 2 bars leaking into the atmosphere. .Che pressure at the other
. I end would for all intents and purposes be zed since the partial pressure of helitan in the atmosphere is very small.
In practice real leaks may exhibit mixed laminar and molecular flow. Howl and Mann (1) state that for capillary leaks the product of leakage rate and capillary length must be less than 10~0 mbar 1/see nun for pure molecular flow. For pure laminar flow this product must be greater than 10-3 mbar 1/see mm. Real leaks vary greatly in cross section along these lengths so these limits are not always strictly applicabis, but nevertheless they do give some indication of where the two regimes lie. From equations (1) and (2) we can see the effects of using dif feret . test gases. For laminar leaks t?,e rlow depends on viscosity. For molecular leaks it depends on molecu:.ar mass. F.r a comprehensive discussion on the relationship between leak rates for different gases and conditions see Amesz(2) , In the helium pressurisation test we use helium as the test gas, whereas leak rates are usually quoted with air as the test gas. Table 1 shows the conversion factors between the two for molecular and laminar flow. Air has a viscosity of 1.71 x 10-5 kg m~l sec ~l and an assumed molecular ~l mass of 28g. Helium has a viscosity of 1.86 x 10-5 kg m~l sec and a molecular mass of 4g. The air flow rate is taken as unity and the helium flow rate calculated relative to this. TABLE 1 Relative flow rates for air and helium, one bar pressure ' different'across leak flow regimes j d
. Test gas Molecular Laminar Air 1.00 1.00 Helium 2.65 1.07 2.3 Predicting leak rates from helium mass spectrometer measurements In the previous section we gave equations for calculating the leakage rate f rom a knowledge of the pressure dif ference across the leak. We now wish to calculate what leakage rate will be observed for a given leak under the i
1 test conditions. Using equations (1) and (2) we find the pressure inside . the capsule at the end of the bombing time. Tnis pressure falls during the transfer time, and we cal culate the final pressure inside, wnen tne leak l rate measurement is taken, only the final results will be quoted here. I For a full derivation see the paper of Howl and MannIII. l l We assume the capsule under test has an internal void volume V , cm3 and a . ) leak rate L mbar 1/sec. The capsule is subjected to a bombing pressure P j bars for a bombing time t3 seccnds. Transfer time t2 seconds elapses I before a measurement of the leak rate R mbar 1/sec is made. ) For molecular flow leaks the following equation of the leak rate (3) is i easily derived. It has been much quoted in the literature, but it must be i borne in mind that it applies to molecular flow leaks only. To illustrate , the method used a derivation is given in Appendix A. l I R = LP x 2.65 [1 - exp (-2.65 L,t))] exp (-2.65 4 t2) (3) V .g v The f actor 2.65 appears because we are using helium a~s test gas whereas the standard leak rate L is for air as test gas. The physical significance of the terms in brackets is easily seen. The first bracket represents the fraction of the bombing pressure P reached inside the capsule after time t g. The second exponential represents the fraction of this remaining after time t 2 f being exposed to air. i l Howl and Mar.nIII quote the following formula, for a leak with laminar l flows
- 1 + S exp (4)
R = P'-Po L -2Lt.,[1~2 V' P 1 - S exp - 2Lt.,i
/
where S = P-Po, and P = 1 bar o P+Po The above expression assumes that the pressure inside the void reaches the bombing pressure P before the end of the bombing time. This is a reasonable assumption as laminar flow leaks have high leak rates. The expression breaks down when the pressure of helian inside the void becomes low. . 2.4 Leak blockage As was discussed in the introduction, it is envisaged that a helium - pressurisation test would be preceeded by a vacuum bubble test. However, it is well,known that liquids may block small leaks, and if the liquid has low volatility this blockage may persist. It is of interest to consider whether the high bombing pressure used in the helium pressurisation test (about 6 bars) may be enough'to force the liquid out of the leak. _4
. __ -_ _ _ _ _ _ _ . . - __ > - _._ . _ _ _ _ . .__ _ .. __m.-_.- . _
The proolem can be approacned tneoretically by assuming that leaks in j pra: tice correr:and to small capillaries laut long this being a typical wall thar..r.ers of e i be given here: m.all radiation source. Only an outline of the theory will
- BurrowsI3I.
- or a more coc:plete discussion see for example a paper by
, Let
- us assume that af ter the vacuum bubble test any leak cresent becomes totally filled with liquid. There are two " actors to consider, 1)
- what is the minimum pressure required to move the liquid.
l ii) how long does the leak take to clear. i The pressure required to move the liquid depends on its surface tension and is given by P = 45 cos d d ) l ^' l 1 2 Where P is the pressure in C , S s the suDhkce tension in Nm~l, d is the diameter of the leak in m, and f6is the' contact angle which is dependent on surface tension. For simplicity we will assume the worst case, complete wetting, for which p = 0. 1 The rate at which liquid is removed from the capillary is
- 4 V = 4P . "IT . r (6)
! 8 t L
]
i 3 j Where V is the volume in m , t is the time in seconds, P is the pressure
; difference m,
across the leak in Nm~2, L i, the length of the capillary in liquid in kgradius r is the m~l sec of~Ithe capillary in m, and g is the viscosity of the 1 Clearly from equation (6) the time taken to clear a capillary is given by t = B_L2 ,) (7) i 4 Pr' 3 l 1 The leak is, as previously stated, assumed to be about 1mm in length, and
- its diameter is related to leak rate according to the equations stated in section 2.2.
If we use glycol as the test liquid in the vacuum bubble test tension = 47.7 x 10~3 Cl and the viscosity = 2.4 x 10~4 kg m~the surface 4 l S~l The pressure normally used in the test is 7 bars. The results for leaks of
- 10-5 and 10~7 mbar 1/sec are shown below in Table 2. We assume that the former shows pure laminar flow and the latter pure molecular flow.
TABLE 2 Blockage of leaks Leak rate Radius of (mbar 1/sec) Pressure to Time to clear leak (m) move liquid leak (sec) in leak (bar) 10-5 1.5 x 10-6 0.6 120 10~7 6 x 10~ 1.6 760 l
- From these calculations we conclude that lea). clogging by liquids will not
!,i seriously affect the helsum pressurisation te st. . i 3. EXPERIMENTAL t 3.1 Introduction g The main aim of ' the experimental work was to demonstrate that - there is a - suf ficiently secure overlap between the ranges achieved by the vacutan ] ! bubble test and the helium pressurisation test. The vacuum bubble test is 1 known to have a limiting sensitivity of around 10 0 mbar 1/sec, and as
- l mentioned previously we need the test to be able to detect standard leaks of up to at least 10~4 mbar 1/sec., This range of. leak rates shows a l mixed laminar and molecular flows and so is difficult to treat I
j theoretically. Even if we assume pure laminar flow equation (4) cannot be j used since it f ails when the pressure inside the source becomes low. For i these reasons the problem was approached experimentally. Y ! When a large leak is present most of the hel.iqis will leak out during the transfer time and nothing will be detected by the helium mass spectrcaneter. ! i To detect large laaks we must use short transfer times and arrange for large internal voids to be present. The vacuum bubble test requires at l least 10mm3 of void.to be present so it would be convenient to use voids i of about this size in the helisan pressurisation test. Some artificial 3 known leaks were used in conjunction with 10mm of internal void volume and the maximum transfer time after which the leak was still detectable above background was determined. This way the largest conveniently 3 [ detectable leak for a 10mm void voltmie was found. The result is 3
' pessimistic since 10 nun is the maallest void voltme we expect to use in !* practice. With larger voids the detection of even larger leaks would become practicable by this method.
t 3.2 Preparation of inrtificial leaks e Short lengths (80:en) of thick welled copper tubing (0.D. = 1.6sm, I.D. = 1.0 man) were annealed at 800 C for several hours and allowed to cool. A 2-3amn length of the and of each tube was crushed in a small smooth-jawed vice. The leakage rate through this crushed and was calibrated using a helium tuned mass spectrometer by connecting one side to a helium cylinder at 0.7 bars over pressure and the other end directly to the spectrometer. The reading obtained was converted into a standard leak rate, for air at
. one bar pressure difference, by asstaning pure laminar flow and using equation (2). This proved a reliable method of producing leaks in the
{ region 10- abar 1/see to 10-6 aber 1/sec. The leaks were tested as . soon as possible after being made (1-2 days) as changes in calibration - occur over a period of time. The open3 ends of the tubes were cut of to
'. leave an internal void volume of 10aun , the open and being sealed using a crimping tool. The seal was leak tight to better than 10~8 abar 1/sec.
I
- 3. 3 , Effects of transfer time The leaks were bombed under two sets of conditions, 34 bars for half an hour and 7 bars for half an hour. Readings of leakage rate were then taken on the spectrometer at regular intervals, to find the maximum transfer time permissible if the leakage is to be detectable above background. In a practical case this depends on the extent to which the outer surface of the source is contaminated with adsorbed helitant this was investigated (see
. l
\
next section) and 5 x 10-7 c. car 1/see was decided upon as beira reliably ; detectable above background. l The results of the tests with variable transfer time are shown in Figure 1. { They show the time elapsed for the measured helium leakage rate to fal.1 to ' 5 x 10~7 mbar 1/see, d for the two bombing conditions being the internal ^ void volume 10mm in all cases. As an example of an individual , measurement in this series, figure 2 shows the measured helium leakage rate ! recorded for a leak of magnitude 10-4 mbar 1/sec after bombing at 7 bars for half an hour. A
. 3.4 Surface helium levels 1
On (<10-8clean stainless mbar 1/sec persteelem2surfaces
). the levels of surface helium are small i
During 0 Special For1n testing however a temperature test is employed (800 C for 10 minuter) and the resulting surf ace may be very dirty and consequently produce high levels of surface helium. ^* ' '4 4 A piece of stainless steel bar was heated at 8000C foi several hours and the resulting surface was extremely black and sooty. The bar was subjected i to two sets of test conditions mentioned above namely 34 bars for half an hour and 7 bars for half an hour. The surface helium levels for the two 1 test conditions were recorded as a function of time. The results are shown in Figure 2 and represent a surface area of 10cn 2 . This area is considerably larger than most sources. l In the previous see l taken to be 5 x 10 gion the baseline for background surface helium was ! justifiable. mbar 1/see and from these results this seems Note that levels of surface helium are essentially pressure dependent
- extend 2d bombing periods will not affect the surface helium levels.
l 4 9 j I 7
I i
\
i
- 4. DISCUSSION '
l Armed with the foregoing theoretical and experimental evidence a definite procedure for the test can be established. One final question remains however, ' what is the smallest standard leak we wish to detect? The question of sensitivity of leak tests was discussed in Appendix 2 and a figure of 10-5 mbar 1/sec arrived at. To give some margin of overlap the test conditions will be devised so that standard leaks of at least 10-6 mbar 1/see will be detected. In principal the sensitivity of the test can be as high as desired. ' Consideration of Figure 1 shows that a measurement should be made within 25 . minutes of removing the capsule from the pressure chamber. This guarantees ] detecting standard leaks as large as 10~4 mbar 1/sec. A convenient pressure to choose is 7 bars as some sources may be damaged if higher pressures are used. The surface helium levels for this pressure are , around 4 x 10~7 mbar 1/sec after five minutes. An oytgassing period of five minutes should be allowed. The figure of 4 x 10~7 pass limit. Any leakage above this is regardeda aM,aAar leak anything 1/see is taken below this as the is surface helitan or leakage due to standard leaks less than 10-6 mbar 1/sec.
~
The surface helium levels quoted are the highest imaginable as they were measured on a very dirty surface of area 10cm2 . Most sources have areas less ! than 10cm2 and will probably be not as dirty. There is therefore no l possibility of confusing a leak with surface helium. l Very small standard leaks may have a leakage rate with surface helium. The smallest standard leak we wish to detect is 10-6 mbar 1/sec and we must arrange for the conditions of the test to make such leaks detectable. The difficulty with very small leaks is when they are coupled with large voids since more helium has to leak in to raise the internal pressure. Most sources have internal void volumes of less than 200mm3 . We can use equation 3 to predict the leakage rate free a source of void voltane 200 nun 3 and a standard leak of In-6 mbar 1/sec. The conditions used will be 7 bars for half an hour. The asurement of leakage will be taken after 25 minutes. O = 2.65L P [1- exp (-2.65L tj)] exp (-2.65L t2) I3I V V
= 2.65L x 10-6 x7 (1-exp (-2.65 x 10-6 x 25 x 60)] exp (-2.65 x 104 x 25.60) 0.2 0.2
= 4.2 x 10 ~7 mbar 1/sec .
This is above our background cut off level of 4 x 10~7 abar 1/sec and so ~ standard leaks of less than 10-6 mbar 1/sec would be detected. Figure 4 - illustrates the situation graphically. Large leaks are a problem when coupled . with small voids. Small lecks are a problem when coupled with large voids.
?
. 5. PROCEDURE FOR MAKING THE TEST EQUIPMENT Pressurising onamber Helium tuned mass spectrometer PROCEDURE 1.
.' Place source in chamber and pressurise with helium for half an hour at a pressure of 7 bars.
2. Remove from pressure chamber and leave for five minutes. l 3. Within 25 minutes place capsule in spectrometer and measure leakage rate. l 4. y ; A leakage rate of less than 4 x 10~7 gfiar 1/sec shall be regarded as surface helium. Leakage rates greater-than this are regarded as leaks an i render capsule a failure. NOTES
- 1. Source should be clean and dry.
2. The 3above procedure is applicable for sources of void volumes of between 10mm -200mm3 and surface areas less than 10c:2 2, TEST SENSITIVITY A leakage of less than 4 x 10~7 mbar 1/sec shall indicate that no standard ' leaks of larger than 10-6 mbar 1/see are present. S e l a
_ . . ~ . . . t
- 6. CONCLUSIONS The, foregoing procedure will confidently detect leaks from 10~4 mbar 1/see to 10~D mbar 1/see with any void volume between 10-200nnn 3 . The pressure of 7 !
bars used ts high enough to unblock any leaks clogged with liquid used in the vacuum bubble test. The test is very versatile, since larger void volumes can be [ 1 tested and higher sensitivities achieved by using longer pressurising periods. [ Sources with smaller void volumes can be tested providing a shorter transfer
- time is used.
i L l
- I
*=
k I r 9 l l l l i
- I s
j i l
~10-
REFERENCES
- 1. HCW!, D.A., MA!TN, C.A.
'The bacx pressurasing technique of leak testing' . !
! Vacuum, 15, 7, 1965. l l r 1
- 2. AMIS", J.
. ' Conversion of leak flow rates for various fluids and different pressure
, Conditions'. ,
Eur atom, Eur 2982 E, 1966. i
)
', 3. BURROWS, G.
l ' Flow through e.nd blockage of capillary leaks' I l Trans. Instn. Engrs . _39, 1961. '
.?
\
l i 1
\
l l
.I
.l 1
I i 1 I 4 l l l I 6 i 1 i l l l ! - l 1
. . _ . . _ _. . - .. . . . . _ . m,. _ _ _ _ _ . _ _ . _ . _ _ . _ _ _ ~ _ . _ _ . _ . - _. _ . _ _ . . _ _ . - . . _ . . _ . _
4 ! i ' APPENDIX A 1
. The derviation of equation 3. Relating measured leak rates to standard ones for i j molecular flow conditions.
t l A capsule of internal void volume V cm3 has a leak L mbar 1/sec. The capsule j is pressurised at P bars for t 3 seconds. A time t2 seconds elpases after ! removing from pressurisation and a leakage rate R mbar 1/sec is recorded. i . since we are assuming molecular flow the outflow of gas from the capsule Q is j , given from equation 1. ) .1 j i ! Q =-2 65 L (po - p) - 9) ; f Where Po is the external pressure and p the internal. The f actor 2.65 arises l because we are using helium as test gas, see section 2.2. The outflow of gas in ; j mbar 1/sec is simply the rate of change of volume at a pressure of one abar. j l \ Ar '
- Q= dV .- -
- 10) l E p s.
~4 Where P, = 1 mbar, clearly Q is related to the rate of change of pressure inside the capsule. From the ideal gas laws we haver PV = const P
s dv i + Vo [dPj =0 - 11) [dt ( lP ( dt ) V Where Vo is the void volume in litres. Substituting 10) and 11) in 9). p= 2.65 L P, (Po - p) - 12) dt Vo Cianging units the ration Ps/Vo can be expressed as I /V if Ps is in bars instead of mbars and V is in em3 instead of litres. We can now resolve into two stages the bombing period and the transfer time. STAGE 1 l During the bombing time the pressure outside the capsule P bars remains constant i for a period t3 seconds. Frczn equation 13 we have; dp = -2.65 L (P - p) dt V i The pressure p rises from zero to p3 Integrating yields. t in P-p3 = -2.65 Lt 3 e .P V P3=P [1 - exp (-2.65 L/V t3)) - 13) STAGE 2 During the transfer time of t2 seconds the pressure inside the capsule falls from P3 to P 2. From equation 12. dp = -2.65 L p E V
Integrating P; = P3 exp (-2.65 L t,) - 14) V When the final measurement comes to be taken the reading on tne spectrometer R abar 1,'sec is given by 15. R = 2.65 L p; - 15) substituting from 14 and 14. R = 2. 65 PL [ 1 - - exp (-2. 65 Lg t ) ] exp (-2.65 L t2) V V This is the final expression we required.
.f.
'4 0
0
. l J
} o l no Fig.1. Leak size against time to decay to background level of 5x 10 - 7 mbar 1/sec. bombing conditions of 34 bars for half an hour and 7 bars for half an hour la l s .t i a
/ *o h.
F 7 34 bars for half an hour 10 [__
~
ur -
/
7 bara fos hal f an hout l _ _ i jo
/
-{ l
_. ~ r - 10
,/ '
10-1 _ _ _ . . _ _ . . _ _ . . . , . _ . . _ . . . . 10-2 10-3
, j 10-4 10-5 10-6 Leak size, miar 1/sec.
1 8
i1. ' n. j
?
! d' 1$$t 1
. .f .i .H .
u 8 O O O
- 0 x n x x -
1 1 1 1 1 0 0 0 0 ~ 0_ ~ -
- ',- 7 6 7
b - 1 ,
- 8. l F
i g 2 1 . 2 l' , t P AS
- ru er 1 , af 6 a oc fe sh ue rl fi
. au 20 m. cm e
sl e 1 v 0e l cs m 2 g _ 4 s f Bu on mc bt ii 2, no g n 8 1 3 o 4f _ bt ai rm 3, se 2 af no dr j 7 s' bo ao 3 rt . T 6 sy i ' - m f ,e os
,& rt .
a m hi n i an n l l u 4 *- fe t 0 - s e , as s n - s _ ht _
- oe ue _
4 .r l . _ 4 s _ u _ r _ f a _ c - e s 4-8 M.-
Decay of 10 ' mbar 1/sec leak. Vold volene 10ica . Fig.3.
.g Dombing conditions 7 bars for half an hour -t>
5 d
.- o i .( y
-i, i: pi
- c.
*<10 .
, - l-
-s ll)
! OO l-
.,- -- . _ _ _ _ _ _ __. __ , _ _ ,, _y I t2 9 12 15 18 21 21 27 30 33 36
. , . Time, minuter
- e ,
4 @ e 9 9
- ~
= s c .
o -
-3 Fig.4. DETECTABLE COMBINATIONS OF LEAK SIZE AND VOID VOLUME 10 (SCHEMATIC)
\
'N
\s Large leak. Small void (10- mbar 1/sec, 10 ama )
\
10~
'< i .
t* o
.o m
.- 5
- 5. 10 -
e I} U d 19 5 2
'1
- _ c, 10 x\
-6 3
. Small Leak, Large Void (10 mbar 1/sec, 200 mun )
Background level 4 x 10 , mbar 1/sec. , s 10~ _ _ _ _ _ _ _ 10 15 20 25 30 5-Time after removing from chamber, mins.
.. . . . ~ _ . _ . . . . . - . -
m-__-._____._.__.-_-__.._mm __ _ _ _ _ _ _ __._ _ - _ _ _ _ _ -_
o l APPENDIX 7 k 1 s l RAOIOAC*IVE METHOCS OF LEAK TESTING PROTOTYPE RADIATION SOURCES
- 1. INTRODUCTION
- 2. EXPERIMENTAL
- 3. RESULTS e 4. DISCUSSION f
- 5. CONCLUSIONS
.f
~4 .
I l l e e 4 d 1
i
- 1. INTRODUCTION l
, During the prototpye testing of radiation sources several forms of prototype source may be employed, the most concton ones being:
i) A simulated source containing an inactive insert but with a small quantity of soluble radioactive tracer material. Nomally 103C1 of Caesium-137. 9 3
, 11) he use of a fully active production source.
e Two types of test would be employed on the above le a wipe test and/or an immersion test. Wipe tests This test as a means of measuring leakage depends on the transfer of active material through a leakage path to the external surface of the source. Providing the source was free from surface contamination prior to any environmental test any subsequent wipe will show gether any leakage has occurred. The wipe test is not normally used as a leak test and it is not , mentioned in Special Form regulations. Its main use is as a quick method for l checking contamination on production batches. Immersion Tests In Safety Series No 6 paras 736 and 737 immersion test procedures are specified for " solid indispersible naterial and encapsulated material. In the fomer case the test determines the amount of activity transferred to water after the material is subjected to seven day storage periods under water and in humid conditions. It is thus an assessment of leachability. In the case of encapsulated material the source is placed in water at 50 C for four hours. The activity of the water shall not exceed 0.05%C1. The source is then stored for seven days in still air at 30 C. The imersion is repeated. In this case the active material has to find its way through a leakage path in the capsule to the water in which it is immersed. The seven day store period allows sufficient time for the material to ' creep' out. Certain other radioactive leak tests are used for example special emanation tests as used for radon sources, these will not be discussed.
- 2. EXPERIMENTAL 4
To assess the various leak testing methods some artificially prepared leaks were required and clearly these needed to be reproducible so accurate comparisons could be made. Taking a standard Amersham International X11 capsule (cylindrical in shape of diameter 10mm and height 5mm) and merely fitting the lid without welding produced a reliable leak. The leak shows up clearly on the vacuum bubble test. Several leaks were calibrated using a mass spectrometer and found to be 7 x 10 ~3 mbar 1/sec. 9 Some Americium-241 ceramics were available in two activities, 100 mci and 2mC1.
'Ihe ceramics were immersed in water at 500 C for four hours. The 100 mci ceramics had activity's of around 100nci leached from them. The 2 mci had activity's of around 10nC1 leached from them. The ceramics were then placed into the X11 capsules, four were made up, two of each activity. These were l
l
I immersed at 50 C for four hours each day for a week. A savan day store was performed whereby the capsules were stored for 7 days at 30 C and then re- . immersed. Two capsules were dispensed with 10pCi of soluble radioactive caesium chloride.
~
The material was dispensed onto the bottom of the capsule. The capsules were immersed at 50*C for four hours. .
- 3. RESULTS The activity of the samples was determined using a liquid scintillation counter ,
- whose sensitivity was around 0.1nci 2 mci Ceramic DAY ACTIVITY OBSERVED nC1 SAMPLE 1 SAMPLE 2 1 <0.1 16 <0.1
~
2 <0.1 <0.'1 . 3 <0 1 <0.1 f 4 <0.1 <0.1 1 5 <0.1 <0.1 J ! 12 <0 1 <0.1 1
- 100.nci Ceramic 1
' DAY ACTIv m OBSERVED nCi SAMPLE 1 r SAMPLE 2 e 1 <0.1 <0.1 1 i f 2 (0.1 <0 1 i ' 3 <0.1 <0.1 f <0.1 l 4 <0.1
- l 5 <0.1 <0.1
' 12 <0.1 <0.1 i e 4 10 Ci Caesium-137 DAY ACTIVITY OBSERVEDwC SAMPLE 1 SAMPLE 2 i 1 <0.1 <0.1 2 <0.1 12 , ] ~ 3 0.8 4800
- 4 1.5 %
2 i ! 4
- 4. DISCUSSION One important distinction to draw before a discussion of the results is made is that we are concerned with the leak testing of sealed radiation sources. We are not concerned with solid radioactive material for which the leachability tests are the only method of testing.
e e
. . _. . . . .-. . - . . ~ -. - - . - . . - .
e The evidsnca prese
.d is somewhat limited but it is .Acar that the Special Form leachability tests on fully active sources is unsuitable for detecting leakage ' paths in sealed capsules containing a non leachable solid. The leaks present were large ones, they were easily detectable on the bubble test. This is the sizeeasily.
and of leak that any leak test used must be capable of detecting both reliably The results of the soluble radioactive tracer study underlines the unpredictable
, way in which activity may be released.
for ' creep' out. Some time was required for the activity period. This strengthens the case for using the seven day s' tore
-e j
{
- 5. Conclusions i i
l l The use of v::1umetric leak tests seems preferable in all respects over radioactive methods for sealed radiation sources. Volumetric leak tests are ! i simple, quier, and reliable and of sufficient sensitivity that there seems no ! possible reason for not using them. t W 1 ."
~4 .
l l
)
l l e e l D i I i l l- t l6 l l 1IST OF FIGURES FIGURE 1 Leak sat against tme to decay to background FIGURE 2 Surf ace nelium
-4 levels as a f unction of tme FIGURE 3 Decay of 10 moar 1/see leak as a function of time FIGURE 4 Detectable combinations of leak sizes and vold volumes LIST OF TABLES TABLE 1 Relative flow rates for air and helium
. TABLE 2 Blockage of leaks r,
4 O 8 e 1 e O e l l 4}}