ML20035A086
| ML20035A086 | |
| Person / Time | |
|---|---|
| Site: | Framatome ANP Richland |
| Issue date: | 03/10/1993 |
| From: | Scarano R, Wenslawski F NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V) |
| To: | |
| Shared Package | |
| ML20035A085 | List: |
| References | |
| 70-1257-93-02, 70-1257-93-2, NUDOCS 9303240047 | |
| Download: ML20035A086 (36) | |
Text
.
U.S. NUCLEAR REGULATORY COMMISSION REGION V Report No. 70-1257/93-02 Docket No. 70-1257 License No. SNM-1227 l
Licensee:
Siemens Power Corporation 2101 Horn Rapids Road Richland, Washington 99352-0130 Facility Name: Siemens Power Corporation b
Inspection at: Richland, Washington Inspection Conducted:
February 9-12, 1993 Team Members:
C. A. Hooker, Fuel Facilities Inspector,' Region V M. Klasky, Nuclear Criticality Safety Specialist, HMSS 3!9!P3 Team Leader:
Alv-t:t F. A. Wenslawski, Deputy Director Dat'e 'Sfgned Division of Radiation afety and Safeguards Approved by:
M 7/[#//1 AossA.Scarano, Director Date Signed Division of Radiation Safety and Safeguards Areas Inspected:
A special announced Augmented Inspection Team (AIT) inspection was performed to review the circumstances and determine the cause of the February 7,1993, event involving the inadvertent accumulation of 124 kg of low enriched uranium oxide powder inside of a lexan enclosure.
Inspection procedure 93800 and the AIT's charter (provided as Appendix A) were addressed.
Conclusions and Findinas
.The AIT concluded that the spill of UO, powder was caused by an equipment failure and the disabling of a limit switch (Sections 2.0 and 7.0).
Contributori causes were (1) failure to follow work control procedures when installing the limit switches, thereby resulting in the bypass of administrative controls intended to assure that equipment modifications are accomplished with proper training and procedure changes (Section 5.0), and (2).
9303240047 930310 PDR ADOCK 07001257 C
~
poor communications within the chemical operations organization (Section 7.0).
Although not directly related to the event, a criticality control concern was identified involving an inadequate evaluation of the potential for moderating liquid intrusion into moderation control areas (Section 8.0).
This issue resulted in the issuance of a confirmatory action letter. The licensee's response to the february 7th event was determined to be prompt and thorough
+
(Sections 3.0 and 9.0).
Open Items:
I One followup item involving the need for the NRC to review the licensee's engineering evaluation of the potential for intrusion of moderating liquids into the Line 2 and Line 3 uranium powder preparation systems was identified (Section 8.3).
f l
1
t TABLE OF CONTENTS 1.0 Introduction - Formulation and Initiation of the AIT Page 1-1.1 Background,and Formation of AIT j
1.2 Persons Contacted 2.0 Descrintion of Event Page 1
-l 2.1 System and Equipment Description 2.2 General Description of Event 2.3 Sequence of Events 2.4 Initial Operating Conditions 2.5 Chronological Sequence of Events j
3.0 Licensee's initial Response - Conclusions Page 7.
3.1 Operators / Supervisors 3.2 Criticality Safety Component j
i 3.3 ' Management 4.0 Precursors-Page 8
[
5.0 Control o'f Maintenance / Modifications Page 8
?
5.1 Background for Installation of the Limit Switches.
5.2 Procedural Requirements 5.3 Circumstances involving Limit Switch Installation 5.4 Conclusions 6.0 Licensee's Corrective Action Trackino Page 14
[
l 6.1 Description 6.2 Conclusions 7.0 Evaluation of Human Performance Page'16
-[
7.1 Training / Communications 7.2 Working Conditions f
I
?
P t
L 2
7.3 Conclusions 8.0 Nuclear Criticality Safety Page 19 8.1 Event l
8.2 Moderation Control in General 8.3 Conclusions 9.0 Licensee's Investication Page 21 9.1 Investigation - Root Causes 9.2 Other Contributory Causes:
9'. 3 Licensee's Recommendations 9.4 Lessons Learned 9.5 Restart Actions 9.6 Followup Actions 9.7 Causal Factors Task Group 9.8 Conclusions j
.10.0 NRC - Summary Conclusions Page 25 i
11.0 Open Public Exit Interview Page 26 j
ATTACHMENTS:
i APPENDIX A - Augmented Inspection Team Charter APPENDIX B - Persons Contacted l
APPENDIX C - General Schematic of System i
APPENDIX D - Confirmatory Action Letter t-i i
1 LIST OF ACRONYMS ADO Ammonium Diuranate AIT Augmented Inspection Team ATP Acceptance Test Procedure BWR Boiling Water Reactor CAL Confirmatory Action Letter j
CSA Criticality Safety Analysis l
CSCAR Criticality Safety Corrective Action Report l
CSS Criticality Safety Specialist ECN Engineering Change tiotice f
i FS feed System j
GSCO General Supervisor, Chemical Operations l
IIB Incident Investigation Board i
HP0 Manager, Plant Operations NRC Huclear Regulatory Commission
[
PPS Powder Preparation System j
i PWR Pressurized Water Reactor i
SOP Standard Operating Procedures SPC Siemens Power Corporation
~
- Temporary Document Revision 00, Uranium Dioxide U,0 Moisture Free Oxidized Scrap fuel Pellets f
WO Work Order l
NOTE:
All times are given in Pacific Standard Time (PST) j
\\
-i f
i
^
4 1.0 Introduction - Formulation and Initiation of the AIT I
1.1 Backaround and Formation of AIT On Sunday, February 7,1993, at 12:15 pm PST, following a substantial f-e' degradation of a criticality control parameter, Siemens Power Corporation notified the NRC Operations Officer in accordance with NRC Bulletin 91-01 reporting requirements.
The Line 3 uranium powder preparation system had been operating at the normal uranium feed flow when, at about 8:15 am, an operator discovered -
an accumulation of a large quantity of low enriched (4.60 wt.% U-235) l uranium oxide (UD,) powder on the floor of a ventilated enclosure that i
encompassed a 00, powder feed hopper. The operator observed that a.
dislodged transfer tube from the feed hopper had caused the accumulation-of the powder and that a limit switch, installed to automatically shut off the power to the system in the event of such an occurrence, may have been taped to prevent its function. A subsegrent inspection of the nearby Line 2 system identified that the limit switch was also taped.
l Line 2 was already shut down for enrichment clean out and Lines 1 and 3 were shut down pending further investigation of the matter.
On February 8,1993, after regional and cnsite inspection staff
[
briefings and consultation with senior NRC Headquarters management, the Region V Regional Administrator directed the dispatch of an AIT. The AIT's charter is provided as Appendix A.
One of the AIT members was onsite conducting a routine inspection and l
the others arrived at the facility at about 7:00 am, on February 9,-
1993. To initiate this special inspection, an entrance interview was held with licensee management on the morning of February 9,1993, followed by a tour of the areas of the facility affected by the event.
This tour and several subsequent tours allowed the team to view the equipment involved in the event.
1.2 Persons Contacted Persons contacted are denoted in Appendix B.
2.0 Description of Event 2.1 System and Eauipment Description The licensee processes low enriched uranium (equal to or less than 5.0 wt.% U-235) for the production of PWR and BWR reactor fuel.
These i
processes are located in the U0, Building of the licensee's facility.
following chemical conversion (two process lines, Line 1 and Line 2) ammonium diuranate (ADU) is heated in a calciner (propane fired rotary kiln) and reduced to uranium dioxide (U0,) powder.
The Line I and 2 l
chemical. conversion lines a e separated within the confines of the U0, Building. The Line I calciner supplies U0, powder to its respective
{
powder preparation system (PPS) and is in a separate room from Line 2.
The Line 2 calciner supplies powder to two PPSs (Line 2 and Line 3 PPS)-
f 2
which are collocated within the same operating area of the facility.
The UD, powder is transferred from the calciners to favorable geometry slab hoppers for intermediate storage. The stored powder is sampled from several sample ports for moisture analysis. The discharge valve from the slab hoppers is padlocked closed and only unlocked by the Shift i
Supervisors, following his/her verification that the moisture sample results are within the prescribed safety limits.
The UD, powder is transferred fro.a the slab hoppers to 21 cubic foot (unfavorable geometry) blenders. A specified amount of verified moisture free oxidized scrap fuel pellets (U 0.) is also added and 3
blended with the UD, powder. The normal blend consists of about 1500 kg i
of low enriched uranium powder.
Powder is transferred from the blenders to an unfavorable geometry PPS l
where it is treated by a hammer-mill, roll compactor, granulator, and transferred to unfavorable geometry 45-gallon poisoned (neutron absorber i
boron inserts) storage drums for ultimate fuel pellet fabrication.
Criticality safety in the blenders and PPS is assured by moderation controls (production of dry powder from the calciner and verification of the moisture content of the powder in the slab happers before the material is transferred to the blenders). The unfavorable geometry blending and PPS operations are also zoned as moderation controlled areas.
f All of the blenders are vertically mounted conical shaped Nauta Mixers with a small V-shaped section at the bottom for discharging the powder.
Powder from the blenders is gravity fed to a hopper equipped with a screw feed system (FS) at the bottom of the hopper. The FS screw feeds the powder through a nylon tube (about 2.5 inches inside diameter),
l secured to the FS by a short hose coupling and two hose clamps, to a vacuum transfer system located about 3 inches off the floor on Lines 1-2 and about 18 inches off the floor on Line 3 (see Appendix C).
The FS is housed in a HEPA filtered air exhausted lexan enclosure. The vacuum transfer system transfers the powder from the FS through hoses and stainless steel piping to an elevated level PPS.
L 2.2 General Description of Event t
At approximately 8:15 am on February 7,1993, the licensee experienced an accumulation of a large quantity of 4.6 wt.% U-235 00, powder on the floor of the ventilated enclosure that encompassed a 00, powder FS on the Line 3 PPS. The operator immediately shut down the Line 3 FS and i
PPS and notified the lead operator who in-turn notified the shift supervisor of the event. The licensee had initially estimated that about 50 kg of UD, powder had been discharged into the lexan enclosure.
+
After the UO, powder was removed and weighed, the licensee determined that 124 kg of powder had been discharged into the enclosure.
9 e
i l
l 3
I i
2.3 Seouence of Events j
The sequence of events was developed by the All from observations by an AIT member who was onsite before the AIT was formulated, interviews with cognizant licensee representatives, and review of selected licensee generated documents.
l 2.4 Initial Ooeratina Conditioni At about 3:55 am Sunday on February 7, 1993, the Line 3 PPS was restarted after a re-blend of UD, powder and continued to run until 7:00 am when the system was shut down for shift change. The blender initially contained about 1500 kg of UO, powder and about 680 kg.had been processed through the PPS.
Following a screen check of the Line 3 PPS by the shift powder prep operator, the system was restarted at about 7:20 am and he noted that there was a good powder flow through the clear plastic _ vacuum hose from the FS. The nearby Line 2 system had been shut down for an enrichment clean out from a previous blend of material, and the area was posted as an airborne radioactive materials area (typical during enrichment clean outs). The operators were waiting for verification of air sampling results to determine if the area could be released from respiratory protection requirements, before resuming normal operations on Line 2.
The Line 1 PPS was operating at a normal
[
rate.
2.5 Chronolooical Seauence of Events Date Time (PST)
Descriotion of Events 2/7 8:15 am The powder prep operator noticed no flow of U0, powder in the transfer hose from the FS to the PPS. The operator opened a door on the lexan enclosure around the FS and his investigation revealed that-the feed hopper was empty and the transfer tube from the hopper was dislodged.
Also, the operator noticed the accumulation of 00, powder in the bottom of the lexan enclosure and that the feed screw for the FS was still turning.
The operator immediately shut down the system and notified the lead operator who had just arrived at the location. The lead operator notified the shift supervisor who immediately responded to the scene.
2/7 8:35 am After his inspection of the scene and observing-that the limit switch for the FS may have been jammed by tape, the shift supervisor notified the Criticality Safety Specialist (CSS) at home by telephone of his observations. The CSS informed the shift supervisor to not disturb the
0 t
4 I
spilled powder until he could inspect the situation and that he would be onsite. shortly.
The onsite NRC inspector arrived at the shift-supervisor's office during this telephone conversation. He then proceeded to the Line 3 PPS to inspect the scene.
2/7 8:40 am The shift supervisor _ notified the General 4
Supervisor, Chemical Operations (GSCO) at home of the event. The GSCO ordered the shutdown of all PPSs. Also, the GSCO informed the shift supervisor that he would notify the Manager, Plant Operations of the event.
2/7 8:50 am The NRC inspector inspected the scene of the Line 3 event. The condition of the inside of the enclosure was not readily visible due to powder fines coating the inside wall of the lexan enclosure. Through an opened side door of' the enclosure, the inspector observed the powder accumulation on the bottom of the enclosure and tape on the limit switch.
The tape on the switch appeared to have been torn during the dislodging of the transfer tube and was not completely surrounding the actuating arm of the switch, but it was conjectured that the tape, whether secured around the actuating arm of the limit switch or-not, had jammed the switch which prevented its function when the transfer tube was initially dislodged.
The inspector questioned the operators on the status of limit switches on the Line 1 and 2 PPSs. Since the Line 2 area was posted as an airborne controlled area (air samples taken but counting had not been completed), the inspector visited the Line I blending room and saw no tape on the limit switch.
When the inspector returned to the Line 2 PPS area the air sampling results had not been completed. Uranium dust on the inside walls of the lexan enclosure prevented viewing the status of the limit switch on the FS for this system.
At the request of the inspector, an operator-attired with a respirator, opened the side door of the enclosure. The inspector noted tape on the limit switch but because of its location, the extent of the tape could not be determined-
t i
5 at this tvne.
2/7 8:55 am The shift supervisor took pictures of the UD, powder accumulation in the Line 3 lexan enclosurc and the tape on the limit switch.
2/7 9:10 am The CSS arrived onsite to evaluate the event and assist in its mitigation. Since the area had been released from respiratory protection controls, a closer examination through the door of the Line 2 limit switch was possible. The l
tape on the raitch appeared to encompass the actuating arm of the limit switch.
To further check the status of the limit switch, an operator removed the transfer tube from the FS. At this point the tape was noted to be secured around the actuating arm of the limit switch. Testing (no powder in the system) of the feed screw proved that the limit switch was taped in a manner to defeat its purpose.
2/7 9:30 am Under instructions provided by the CSS, t
Operations made preparations to remove and i
determine the amount of UO, powder that had accumulated in the Line 3 lexan enclosure. The initial estimate was about 50 kg of UO, powder.
The shift supervisor ordered the restart of the Line 1 PPS since no tape was found on the limit switch of this system.
The inspector requested, of the shift supervisor, that pictures be taken of the limit switch on the Line 2 FS. The shift supervisor relayed this request to one of the operating crew.
Subsequently, it was reported to the shift supervisor that the tape had been removed from the limit switches, therefore no pictures were taken.
l 2/7 9:50 am With the concurrence of the GSCO and the CSS, I
the operating shift began to remove the spilled UO, powder from the Line 3 FS enclosure using the central vacuum system and its associated slab collection tank.
As normally done, 00, powder is dumped from this small collection slab tank into 5-gallon buckets.
Each 5-gallon container was being t
5
t I
transferred to a central weighing station to quantify the spilled powder.
2/7 10:00 am The shift supervisor initiated a walk-down of the chemical conversion and PPSs to verify the status of all known limit switches.
No adverse conditions were identified with other installed limit switches.
2/7 11:15 am After the GSCO was notified of plant conditions and the results of the walk-down of known limit switches by the shift supervisor, he ordered the shutdown of the Line 1 PPS and to keep all three lines down until further notice.
Because of a misunderstanding during the 8:40 am L
telephone conversation between the shift supervisor and GSCO, after the limit switch on the Line 1 FS was verified not to be taped, the shift supervisor had restarted the Line 1 PPS which was contrary to the GSCO's belief of the instructions he provided to the shift supervisor.
2/7 11:48 am The CSS notified the NRC Operations Officer that he had an event to report under NRC Bulletin 91-01, and would be providing a written report by facsimile shortly.
2/7 12:15 pm The CSS made their 4-hour actification by facsimile which was also read by the CSS via telephone to the NRC Operations Officer.
2/7 12:45 pm The cleanup of the Line 3 spilled powder was completed with 124 kg of 4.60 wt% enriched UO, powder recovered.
j Note: The nominal feed rate from the blender is about 300 kg of U0, powder per hour.
Therefore, it was approximated that the feed tube had been dislodged for about 25 minutes before the operator identified the l
spill.
4 2/7 Early Afternoon A management team consisting of the Plant lianager, Plant Operations fianager, lianufacturing Engineering lianager, and the Safety Security and l
Licensing lianager arrived on site to review the event and appoint a root cause analysis task team to investigate the incident. The licensee also formed a separate causal factors task 7
i
l
~
7 e
group.
The root cause investigation team initiat'ed with the GSCO interviewing all powder preparation l
operators, lead operators, and supervisors of all four operating shifts.
Off shift crews were being interviewed by telephone.
j 3.0 Licensee's Initial Response - Conclusions 3.1 Operators / Supervisors Based on the description in Section 2.5 above, the AIT concluded that the initial response by the operators was prompt and in accordance with the licensee's procedure for response to off normal events. The affected system was immediately shutdown and notifications to l
supervision, the criticality safety component, and management were prompt. However, it was noted that the shift supervisor restarted Line i
1 PPS at 9:30 am.
During the 8:40 am telephone conversation between the GSCO and the shift supervisor, the GSCO had instructed the shutdown of all PPSs.
r Section 3.0, " Authority for Resuming Operation After Abnormal Event," of the licensee's new " Abnormal Event Reporting" temporary internal reporting procedure (issued by memorandum from the Manager, Plant-Operations to All Plant Operations Supervisors / Potential Candidates, dated October 21, 1992) required that for certain abnormal events the shift supervisor must shut down the affected (emphasis added) operations' immediately and does not (emphasis added) have the authority to continue restart of affected operations without the concurrence of the General Supervisor of the respective area and the Manager, Plant Operations.
Section 3.0 defined the following events that require approval by the General Supervisor and Manager, Plant Operations for continued operations:
"])
Spills of greater than 15 kg of uranium compounds," and j
"4)
Criticality safety violations which are potentially-i reportable under the NRC Bulletin 91:01 (four-hour reporting requirement)."
Based on interviews with the shift supervisor and the GSCO, the AIT l
determined that the shift supervisor had misunderstood the GSCO's instructions, and he had concluded it was acceptable to restart Line 1 PPS because it was not an affected system in that it was physically separated (in another room), and no tape was found on the Line 1 limit switch. The misunderstanding between the shift supervisor and the GSCO highlights the need for clear communications, especially in the midst of an off-normal response situation.
l
i 8
)
3.2 Criticality Safety Componer1 The CSS arrived onsite approximately 35 minutes after being notified of l
the event. This individual inmediately. inspected the conditions at the Line 3 FS's enclosure to determine.the safety significance of the accumulation of U0, powder.
i The CSS determined that the accumulation of the UD, powder was within the confines of the moderation controlled enclosure; however, the release of a suspected more than a minimum critical mass of-4.6 wt.% UO, powder outside the confines of the process system at the floor level indicated a substantial degradation of a criticality control parameter.
The CSS also determined that due to other controls in place and the condition (dry) of the UD, powder, a criticality accident was not-possible which also negated any potential for the licensee to implement their Emergency Plan.
l The CSS's initial actions were prompt and deemed appropriate.
3.3 Manaaement The GSCO, Plant Operations Manager, Manufacturing Engineering Manager, Plant Manager, and Safety Security and Licensing Manager responded to i
the facility to review the event and appoint a root cause analysis task team to investigate the incident.
Plant management's initial actions were prompt and deemed appropriate.
4.0 precursors Based on interviews of cognizant licensee staff, the AIT determined that some operating shifts had experienced problems with the limit switches shutting off the FS's feed screw due to vibration and movement of the feed tube. The limit switch problems were sporadic with some operators experiencing no problems and others experiencing shut-offs up to six times per shift. On one occasion, a lead operator had observed an operator using a cardboard shim to prevent the limit switch from shutting off the FS's feed-screw.
Such a shim would not had defeated the operation of the switch in the event of a feed tube dislodge. Some operators had observed tape on the switches prior to the event j
(discussed further in Section 6.1). - There was no knowledge of the feed tube dislodging prior to the February 7,1993, event.
l i
5.0 Control of Maintenance / Modifications The-AIT members determined that the tape on the limit switch of the Line 3 FS prevented the shut-off of the feed screw when the discharge tube became dislodged. As described earlier, the Line 2 limit switch was also found to be disabled with tape. The history of these limit switches was investigated. After the background for the installation of the limit switches was understood, the AIT members reviewed the license's program.for controlling maintenance and plant modifications 9
9 t
for non-safety and safety related systems and/or components. The AIT-evaluated the effectiveness of the licensee's procedures and examined l
i the licensee's adherence to these procedures as they related to this event.
t 5.1 Backaround for Installation of the limit Switches On August 27, 1992, a non-NRC reportable event occurred in the Line 2 conversion area that involved over-batching of a 5-gallon U0, powder j
bucket while dumping powder from the central vacuum system's collection-t tank. A Criticality Safety Corrective Action Report (CSCAR) generated as a result of this event identified as a corrective action to " review l
i all related equipment to determine if this potential for overfilling containers exist elsewhere and then determine appropriate corrective actions." In response, Operations developed a draft list entitled l
" Items for Consideration of Modification to Enhance Control of fissile l
Material in the Conversion Area." This list identified nine locations for action, including the installation of the limit switches. The specific statement relative to the limit switches stated, "Line 1, 2, and 3 blender accurate feeders - install limit switches at the end of l
i the nylon feed tubes interlocked to shut off the accurate feeders if the l
tube is pushed out of the feeder. This problem is a rare occurrence."
[
Interviews by the AIT revealed no other past events of the tubes being completely pushed out of the feeder, although some outward ~ movement was i
j The current feed screw design in the Line 3 system was common.
installed in early 1988.
During an undocumented meeting between the GSCO, CSS, and Manager Mechanical / Chemical Engineering, the action items were reviewed and a 1
more formalized list was prepared for implementation of the identified actions. This document selectively categorized each action item for s
each location as "Long-term Engineering Fix, Long-term Action, Short-term Actions, and/or Short-term Adm. Control." A statement at the j
beginning of the list characterized the items as " locations where i
operators don't have a reliable means to assure compliance with criticality safety specification limits."
l Location No. 9, "Line 1, 2, and 3 Blender Accurate Feeders," listed one action categorized as a long-term engineering fix. This action called 3
for the installation of limit switches at the end of the nylon feed l
tubes interlocked to shut off the screw feeder on the FSs if the tube l
)
were to push out of the feeder. With the exception of location flo. 9, all of the corrective actions for the other eight locations listed the r
person responsible for implementation of each item. The estimated time j
to install the limit switches was about one week.
l According to the CSS, the recommendation for installing the limit switches on the FS was an added enhancement to the system and not a control to prevent a violation of a criticality safety limit. The CSS l
stated that since the limit switches were only installed as an 1
enhancement, and not a control, there would be no requirement to change 4
-l a
I
~
i 10 t
the criticality safety analysis. The CSS stated that the addition of the limit switches was to preclude operators from becoming conditioned l
to thinking that spills of U0, powder was acceptable. He further stated that the criticality control for the other eight locations was based on i
mass; therefore, the corrective actions for these areas were to prevent j
exceeding the specified mass limit. The controlling parameter for the unfavorable geometry blenders, including the lexan enclosures for the FS, and the unfavorable geometry PPS was moderation and not mass.
On September 4, 1992, a Work Order (WO) No. B05408 was drafted and j
approved by the responsible electrical engineer (signed for the department manager) with the installation of the limit switches to start t
on September 11, 1992. A copy was provided.to Safety, Security and i
Licensing.
However, as ' standard practice, no copy was provided to Operations. The description of the work activities stated:
j i
Install limit switch on the discharge of the blender accurate feeders on all 4 blenders. This limit switch will shut down the accurate feeder motor if the feed tube is pushed off the feeder.
j (This is a Criticality Concern)"
Note:
The 4th blender is a new large lot blender that is under development and has been shut down since July j
1992.
i During September 11-14, 1992, the limit switches were installed per the WO instructions. Other than the powder prep operator (s) observing the j
installation of the limit switches, no other Operations personnel were l
made cognizant of the work that took place on the day of installation.
}
5.2 Procedural Reouirements General:
The licensee has a two tiered approach for accomplishing repairs and modifications to equipment. Work Orders (W0s) are used for repair activities and are implemented and accomplished with a minimum of administrative effort.
Engineering Change Notices (ECNs) are used for i
equipment modifications and design changes.
ECNs have tighter j
administrative controls and the process is designed to assure, among j
other matters, that design drawings are modified, operating procedures are changed to reflect the design change and operators are trained on the changes. The following provides a description of the licensee's I
requirements for W0s and ECNs.
i Work Orders:
The licensee's WO process allows that any individual can generate a WO for repairs or modifications. The author of the WO provides a l
description of the work to be performed and forwards it to the responsible engineering organization. The responsible engineer reviews i
the W0 and determines if the work can be performed under the scope of a i
)
?
I
11
(
WO or other process.
Section 7.3, " Work Orders," Chapter 3.0, " Nuclear Criticality Safety Standards," of Siemens Safety Manual EMF-30, Revision 10, requires that W0s shall be reviewed before execution by the author, and approved by a supervisor or manager in Plant' Engineering, or his delegate, or the author's manager to assure that the ordered work meets one or more of the following criteria:
"a)
The work is a replacement in-kind of previously approved j
equipment.
b)
Fissile material equipment / operations are not involved.
7 c)
The work is authorized by an Approved ECN."
l Section 7.3 also requires that copies of all W0s issued shall be routed to Safety, Security and Licensing.
Section 2.0, " Scope," of procedure EMF-858 No.1.21, " Work Order i
Instructions," Rev. 7 of Siemens Manufacturing Engineering Procedures and Practices states in pa o:
The instructions in this Procedure apply to personnel requesting Work of Plant Engineering or the Machine Shop. Work orders are for plant repairs and changes that do not alter the l
process, environmental protection, criticality safety, radiation protection, or industrial safety."
This Procedure is intended to cover W0s associated with SPC Operations and Engineering divisions. 'It does not deal with
" Breakdown Work Orders," or work orders which are written in order-to schedule approved ECNs."
1 Section 3.1, " Preparation," on pages 3 and 4 of procedure EMF-858 No.
1.21 state in part:
1 Page 3 "Next, the author, in conjunction with'the Responsible Engineer, must evaluate the current-field condition of the equipment and work area to be affected and compare it with that currently shown on plant drawings.
If current field r
conditions are different from those shown on the drawings, the drawings should be redlined before work is started.
If the WO involves a change to the current condition of plant / equipment, the Responsible Engineer shall either markup the existing drawings, or have a new sketch / drawing prepared-by either himself or herself or Design and Drafting...."
[
t
+
i 12 l
Page 4 "The Respensible Engineer must assure that the scope of work defined by this WO is appropriate for being completed under-a WO, or should be revised / upgraded to an ECN (see j
definition of ECN in ME Procedure 1.13)....
If it is determined that the job should require an ECN, the WO will
[
be eliminated and reissued as an ECH...."
Engineering Change Notices:
t l
Section 2.1 the " Scope" of procedure EMF-858 No. 1.13, " Engineering Change Notice (ECN)," requires that this procedure be used for additions or deletions of, or modifications to, facilities, services and equipment I
when work meets one or more of the following criteria:
j "Affects the basic principles of operation of the manufacturing
[
process.
Directly involves fissile material.'
l Affects criticality or radiological safety.
Section 3.3.2, " Manager, Plant Engineering," of EMF-858 No.1.13 states in part:
"The Manager, Plant Engineering, in concert with the User / Operator l
Manager shall determine if an ATP [ Acceptance Test _ Procedure] is required."
Sections 3.3.3 of this procedure requires the concurrence of the Safety-Supervisor and Section 3.3.4 requires the approval by the Operator Manager, after reaching an agreement with Manager, Plant Engineering as j
to the need of an Acceptance Test Procedure (ATP).
[
Section 6.3, under " Acceptance of Work Completed," of EMF-858 No.1.13 l
requires that if an ATP is not required, acceptance of work is accomplished via a functional check or test Section 2.0 " Scope," of EMF-858 No. 1.14, Acceptance Test Procedure,"
requires that when work is ccmpleted by an ECN it is accepted through l
either a functional Check or-alp. Among the many reviews, checks and/or tests required on the forms' attached to this procedure before equipment can be accepted are:
t Equipment Specifications and/or as built drawings to which the equipment was purchased or fabricated - Copy to Master Equipment File.
l, t
[
f
4 i
I 13 l
Criticality Safety Specification (s) approved.
l Equipment size and location and interconnections verified to agree with the criticality evaluation assumptions, j
Standard Operating Procedures draft copy submitted to Operations l
Operator training accomplished.
t I
Since all work performed under an ECN does not require an ATP, which i
assures operating procedures and operator training have been completed before modifications are accepted for operation, the AIT discussed the ECN approval process with the Manager, Plant Operations (MPO) relative i
to the above criteria. The MP0 stated that although an ATP may not be l
required for all work under an ECN, he does not approve the ECN until Operations performs the required functional test appropriate to the modification. Although there was no formal method for assuring procedures are changed and operators are trained for modifications made i
without an ATP, he provides the respective area supervisor with a copy l
of the signed ECN. The area supervisor assures that operating l
procedures are changed as appropriate to reflect modifications and that j
operators receive the necessary training as to the purpose and the e
l operating controls / limits that apply to any system modifications. The MPO also stated that he was not comfortable with the WO process because, any WO that was not generated by Operations does not get Operations j
approval, and such work may be performed without the knowledge of Operations. The MP0 was confident that when the ECN process is used, appropriate procedure changes and operator training will occur, even if 4
an ATP is not performed.
5.3 Circumstances Involvina t.imit Switch Installation l
The AIT discussed the WO process, concurrences and. approvals with engineering management, Operations, and the engineer who initiated WO B05408.
The responsible engineer who initiated W0 B05408 told the AIT that he viewed the scope of the work as a minor modification that did not constitute a physical change in the process,. and that similar modifications (installation of limit switches) were normally done via l
the WO process without initiating an ECN. The engineer stated that no design change drawings were modified to reflect the installation of the limit switches. This individual also viewed the WO process as a faster-way of getting the job done.
In retrospect, he acknowledged that the installation of the limit switches on the FSs should have been performed l
under the more formal ECN process.
This individual also stated that, in general, the WO process may be used too frequently for such work.
l I
Both, the responsible engineer and his ranager stated that' the installation of the limit switches was considered an interim fix,
t 14 because engineering was in the process of designing an improved feed system. The new design would alleviate problems normally encountered with the current feed system.
/
' The AIT expressed a concern to licensee management that the installation
]
of the limit switches did not occur in accordance with established procedures, which clearly indicated that an ECN was required. The licensee agreed with this conclusion.
5.4 Conclusions Based on the discussions delineated in Sections 5.1-5.3 above, the AIT made the following conclusions:
1)
The licensee's procedure for installing design modifications was not followed.
Indications were that the limit switch situation t
was not an isolated case.
It appeared that W0s may be used to expedite work instead of using the required ECNs.
i 2)
The WO process bypassed administrative controls that would have:
(1) made Operations aware the switches were being installed, (2) prompted applicable procedure modifications, and (3) prompted operator training to be administered.
i 3)
The informality of the WO process may result in configuration r
control problems. Although the W0 would require drawing updates, W0s are frequently viewed as a quick fix, and as in this case, i
drawings may not always be changed.
6.0 Licensee's Corrective Action Trackina 6.1 Description As part of the inspection, the team reviewed the licensee's system of tracking corrective actions that arise from generic nuclear criticality safety concerns. Deviations from criticality safety specifications identified through routine audits'by either the CSS or Operations and/or those identified.through process upsets, or equipment variations are submitted to the responsible manager for correction through the use of a Criticality Safety Corrective Action Report (CSCAR). The procedures for the initiation and administration of CSCAR's are described in Section 9.0, Chapter 3 of the licensee's Safety Manual.
Depending on the nature' and severity of any infraction, the shift j
supervisor, after consultation with the CSS, approves immediate actions.
The responsible manager, in writing, using the CSCAR, is responsible for delineating the (1) cause of the infraction, (2) immediate corrective -
l actions taken to correct the problem, and (3) longer term corrective actions planned to prevent recurrence. The CSCAR delineates who took-the immediate corrective actions and the person responsible for the long term actions. Operations then forwards the CSCAR to the CSS for approval. The CSS reviews the actions prescribed by Operations and E
i 7
i 15 signs his approval if the actions appear appropriate. The long term corrective actions are tracked through completion by a database maintained by the Criticality Specialist. A monthly summary of all open items on the database is provided to the responsible managers as well as l
the' plant manager.
Regarding the tracking of long term corrective actions related to the CSCAR for the August 27, 1992 event, involving the over batching of a 5 gallon bucket (described in Section 5.1 above), the AIT made the following observations fron, a review of the CSCAR:
1)
The CSCAR stated, " Review cll related equipment to determine if this potential for overfilling containers exist elsewhere and then determine appropriate corrective actions."
2)
The CSCAR also described that meetings were held on August 27 and 28, 1992, and a meeting was scheduled for September 1,1992, to complete the review, and that Safety, Security and Licensing would be issuing the minutes of the meetings, which would indicate persons assigned to action items.
3)
The CSCAR was signed by the Manager, Plant Operations on August 27, 1992, and forwarded to the CSS.
4)
The CSCAR had a handwritten note on it from the CSS to the Manager, Safety, Security and Licensing that stated that he, the CSS, did not approve the CSCAR because it did not have definite and trackable actions. The CSS also delineated that had he i
received the meeting minutes, he would have reviewed the corrective actions and if they were appropriate, he would have t
attached them and signed the CSCAR.
lhe AIT noted that the informal list of corrective actions for nine locations, with NO. 9 being the installation of the limit switches for j
the FS (described in Section 5.1) was never provided to the CSS. Based-on interviews with cognizant licensee personnel, the AIT concluded the l
licensee failed to follow through with the administrative aspects associated with this CSCAR. Although the CSS knew through the W0
~
process that corrective actions were being implemented, there was no formal establishment of tracking the corrective actions for the nine locations related to this CSCAR.
Cognizant licensee representatives agreed with the AIT's conclusion and stated that although their formalization of tracking such items on a master plant tracking commitment system has not been completed, had the-
- i August 27th event occurred more recently, the tracking problem would not have occurred. The licensee representatives also stated that when the August 27th event occurred, they were heavily involved with the August 8,1992, event that resulted in subsequent NRC escalated enforcement actions which caused them to divert resources and the August 27th corrective actions were not effectively administered.
i i
t i
I 16 I
6.2 Conclusions The AIT had the following conclusion relative to corrective action l
tracking:
1)
The action to install the limit switches (along with the other actions originating from the August 27th event) was not being
[
effectively tracked. The failure to track the action did not appear to contribute to the February 7th event.
i 7.0 Evaluation of Human Performance r
7.1 Trainino/ Communications j
According to the GSCO and based on the review of licensee records, changes to procedures either from modifications or operational changes are initially handled through Temporary Document Revisions (TDRs). TDRs i
are used until such time the effected operating procedure (s) are revised and approved. When a TDR is generated, each shift supervisor provides training to the operators relative to the change. The operators sign a training form acknowledging their attendance to the training provided.
The attendance form notes the TDR number covered in the training session. The AIT noted that although the training form delineated the TDR covered, the training forms lacked specifics as to the depth of training provided and/or specific topics covered. Records of training Varied significantly between shift supervisors.
According to the GSCO, the general practice for training related to routine procedure updates that reflect current operations and/or to i
reflect changes from generated TDRs, is less formal. Routine procedural updates can include minor administrative changes as long as they have no l
effect on process er criticality safety, which are covered by the formal training method before a procedure can be revised. The usual process to alert operators of general procedural updates involves the GSCO i
informing the shift supervisors of the updates, by writing a note on the update cover sheet that the new procedure has been filed (a mini library l
is available to all operators) and request.ing the updates be reviewed by the staff.
The shift supervisor subsequently instructs the operators to review the updated procedure. Changes made in procedures are marked with an asterisk to reflect the addition.
The AIT noted that a routine update of Standard Operating Procedure, h
EMF-22, No. P66,813, " Preparation of UO, Powder as Press feed," Revision l
No. 5, dated September 24, 1992, inserted item 3.6 under Section 3.0, i
" Radiological, Industrial and Fire Safety," which stated.
" Interlocks are not to be bypassed during operation of the powder preparation process."
l Item 3.5 of Section 3.0 stated:
l "Become familiar with location of electrical disconnects for each
{
17 piece of equipment".
I The AIT noted that Attachment No.1, " Electrical Interlock Verification," to Procedure No. P66,813, Revision No. 3, dated June 17, 1992, through Revision No. 9, provided a list of eight components that had interlocks and a description of functions associated with the various interlocks. The limit switches that were installed on the FS from the blenders were not described on Attachment No.1.
The AIT noted from records that training had been provided on TDRs related to procedure P66.813 in October and November of 1992.
Records of training did not indicate whether the statement "* Interlocks are not to be bypassed during operation of the powder preparation process" was specifically covered. However, the AIT concluded, based on interviews, that operators were generally aware of the requirement-to not bypass interlocks.
Other than the on-shift operators who saw the limit switches being installed during the week of September 11-14, 1992, there was no general recognition within Operations that the switches were installed. There had been no log entry made of the installation and no apparent formal communications between shifts regarding the switches.
From feedback through the operators,.two shift supervisors knew for an undetermined time period that the limit switches had been installed. Another shift supervisor became aware of their installation about two weeks before the event when a problem occurred with the Line 2 feed tube becoming partially dislodged and the FS was shut off when the limit switch performed its function.
One shift supervisor was not aware that the limit switches had been installed. The shift supervisors that were aware of the limit switches had no knowledge of tape-being used to defeat their function. The GSCO who initially participated in making the recommendation for the installation of the limit switches, was not aware that they had been installed.
It appeared that most of the powder preparation operators eventually became aware of the limit switches.
Four operators had seen tape on the limit switches within two weeks of the event. These operators didn't perceive the switches to be interlocks. When the shift supervisor observed the limit switch during the problem in Line 2-(noted above),
the switch was not taped.
None of the operators interviewed by the AIT or those interviewed by the licensee's investigation team provided information as who had taped the switches.
Some of the operators assumed that the limit switches were for preventing the powder from spilling into the hood if the feed tube became dislodged from the FS. Several of the operators assumed that the limit switches were only to stop the feed flow during powder plugging or equipment problems. There appeared to be no clear understanding among operators whether the switches served as an operational convenience, a safety device, or whether they were considered an interlock.
t i
18 l
7.2 Workina Conditions l
Both chemical conversion lines and associated powder preparation lines i
are normally operated 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> per day and 7 days per week by 4 rotating operating crews.
Each operating crew works a 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> shift with shift l
changes at 7:00 am and 7:00 pm.
Each operating crew consists of 12-15 l
operators who are assigned to various operations and a shift supervisor.
I l
Each shift has an assigned lead operator who oversees both chemical and powder preparation operations. One operator is assigned to the Line 1 l
PPS and two operators provide coverage for the Line 2 and 3 PPSs. Time l
off between shift rotations ranges from one day to three and six days.
Shift turnovers normally last about 15 minutes. When an operating crew returns from a six days off schedule, they are required to review l
operator and shift supervisor logs for that period. Operator logs did i
not routinely discuss minor maintenance / modifications such as that for j
the installation of the limit switches or similar work.
l According to the operators interviewed, the feed from the blenders j
(Lines 2 and 3) to the PPS would occasionally stop due to the actuation 1
of the limit switch. The operators stated that the limit switch had l
been activated from either vibration in the system or the FS's feed tube becoming partially dislodged (not completely separated from the feed hopper) from the connecting boot. Depending on the consistency of the uranium powder being blended, occasionally the vacuum box at the end of l
the FS feed tube would plug which required cleaning and disturbance of the limit switch.
On one occurrence a lead operator observed a powder prep operator using a cardboard shim to keep the limit switch on Line 2 from shutting off the feed flow from the blender. The lead operator i
i stopped this practice and called for maintenance to make the necessary adjustment on the limit switch. As stated above, some operators had I
observed tape on the switches prior to the February 7th event.
l The AIT noted that the Line 1 FS's limit switch was more visible and 2
readily accessed from its small lexan enclosure, and that the clamps on l
the boot connecting the feed tube to the hopper could be tightened easily.
It appeared that the Line 1 FS did not have similar problems j
i (activation of the limit switches) as those identified with the Line 2 and 3 FSs.
l Regarding the Line 2 and 3 FSs, the visibility through the large lexan enclosures of these systems was poor due to uranium dust that collects i
on inside walls.
Access to tighten the clamps on the feed tubes' connecting boot to the FS was not easily accomplished.
Specifically, a i
long reach was required, at an awkward position, to access the hose clamps.
The general work area is congested and the PPS encompasses two elevations.
Respiratory protection devices are frequently required i
during a working shift for equipment clean outs, for enrichment changes, and other similar tasks.
Portable lamps are placed behind selected i
clear plastic sections of the vacuum transfer system as an aid for the operators to observe movement of the UD, powder.
l 1
\\
i 19 Based on interviews and discussions, it was apparent to the AIT that many of the operators were aware that the limit switches were not performing as expected (continued force shutdowns), yet supervision and management were not aware of the extent of the problem.
The licensee's explanation for this was' that the problem had not reached a " Threshold of Pain" for the operators such that they would report it to supervision.
Instead operators chose to deal with the problem following their own inclinations.
It appeared to the AIT that management's expectations for reporting operational problems needed to be better i
delineated.
It also appeared that shift supervisors were not aggressive in staying abreast of on-shift issues. As a general statement, communications within and between shift crews was lacking.
7.3 Conclusions i
Based on the above findings, the AIT made the following conclusions:
1)
No training specific to the limit switches occurred resulting in a 1
poor understanding by operators of the purpose of the switches.
l 2)
Communications within and between shifts regarding the limit j
switches was lacking.
Shift supervisors were not aggressive in identif.fing and correcting problems associated with the limit switches.
i 3)
The limit switches were most likely taped by an operator (s) who was frustrated by frequent job interruption and this person (s)
)
likely did not recognize the significance of the action.
8.0 Nuclear Criticality Safety 8.1 Event As part of the inspection, the team assessed the nuclear criticality safety implications of-the February 8, 1992, event. The team concluded that the release of 124 kg of U0, powder into the lexan enclosure did
-j not by itself constitute an immediate nuclear criticality safety threat, l
given that the powder had previously been verified to contain less than i
1.0 wt% water.
(Uranium oxide at 5.0%' enrichment containing less than i
1.0 wt% water cannot be made critical.) However, nuclear criticality i
safety is very sensitive to the quantity of moderating material interspersed in the powder.
For instance, 120 kg of uranium powder at 5.0 wt.% U-235 homogeneously' mixed with 30 kg of water in the most j
reactive geometry (a sphere) and fully reflected (one foot of water surrounding the sphere) may become critical (K,,, = 1.0).
j B.2 Moderation Control in General u
In addition to' satisfying the double contingency principle by assuring i
that the UD, powder entering the unfavorable geometry powder prep processes contains less than 1.0 wt% water, it is imperative to ensure I
that operations are conducted in areas that are protected against the l
~_-
l 20 l
possible intrusion of moderating materials.
The licensee has taken precautions to negate the possible intrusion of I
moderating liquids into the powder prep system by denoting the U0, i
powder prep area as " moderation control area." The licensee's j
procedures restrict lines of moderating liquids from being present above i
moderation control areas.
Also, the room that contains the blending and PPSs is posted as a water exclusion area for fire fighting purposes.
The City Fire Department is cognizant of the licensee's water exclusion l
l areas which are clearly indicated in the licensee's Pre-fire Plan (reference NRC Bulletin 91-01 Report No. 23510 and Licensee's report dated June 24, 1992).
?
j Although the Line 2 and 3 blenders and PPSs are located in marked j
moderation control areas, the team observed sources of moderating liquids (safety showers, demineralized water lines, steam lines, and chemical supply lines) in close proximity to these moderation control areas. This observation questions the validity of the licensee's assertion that moderating materials could not enter the moderation control areas and thus pose a nuclear criticality concern in the event of a spill such as occurred on February 7th. Also, the vacuum transfer j
system inside of these enclosures could route intruded water to the l
downstream unfavorable geometry PPS. The air inlet to the vacuum system i
3 for the Line 3 FS is about 18 inches above the floor and the Line 2 l
vacuum inlet is about 3 inches above the floor. At the time of this inspection, the licensee was unable to provide any technical basis for the assertion that moderating materials could not intrude into lexan
[
enclosures that house the FSs at the bottom of the blenders.
It was the i
licensee's position that a moderator intrusion into the lexan enclosure j
(other than fire fightinq related) was a non-credible accident. No l
sources of moderating li quids were observed in the Line I blending room.
l The licensee had provided a control to preclude the possible accumulation of U0, powder in the lexan enclosure in the event the powder feed tube to the PPS dislodged. The licensee asserted that the presence of the control (limit switch) was not a necessary nuclear
(
criticality control because the moisture content of the powder had satisfied the double contingency principle, and the process was l
contained in a moderation control area.
In so far as the accumulation l
)
of UO, powder in the lexan enclosure was not perceived to be 'a nuclear criticality threat, the criticality safety analysis (CSA) did not I
contain an evaluation that provided the technical bases for assuring l
that the nearby liquid moderating systems did not pose a criticality safety concern.
i The licensee's CSAs did not evaluate U0, powder spills / accumulations i
inside of the lexan enclosures. The AIT noted that the licensee's j
criticality safety analysis for the UD, PPS was deficient in that it did not justify any of the assumptions in the analysis that dealt with the i
possible intrusion of water into these systems and did not present a j
description of what accident conditions they were guarding against.
j
21 i
In response to the AIT's concern regarding moderation control, the licensee acknowledged that there was no documented evaluation to discount the adjacent liquid moderating sources as a criticality safety l
However, the licensee stated that such an evaluation was concern.
considered to be within the scope of its Criticality Safety Update Program provided to the NRC by letter dated December 30, 1992.
{
On February 12, 1993, NRC Region V issued a Confirmatory Action Letter (CAL) tu document the licensee's agreement to perform an engineering evaluation of the potential intrusion of moderating liquids into the I
licensee's Line 2 and 3 00, blending and PPSs, and the controls necessary to prevent a criticality. The CAL also delineated the licensee's agreement to use alternate methods to transfer U0, powder in these systems and other compensatory operational conditions if the licensee planned to use the FS. The CAL is included as Appendix D.
8.3 Conclusions i
The AIT concluded that the licensee had not performed an adequate i
evaluation to assure that there was no potential for intrusion of moderating liquids into the Line 2 and 3 blending and PPSs. The licensee's engineering report relative to the potential intrusion of moderating liquids into the Line 2 and 3 blending and PPSs will be i
reviewed in a future NRC inspection and is considered as an inspector followup item (70-1257/93-02-01).
I 9.0 Licensee's Investication i
In the afternoon of the Sunday February 7,1993 event, plant management j
3 arrived onsite to review the event. A root cause analysis task team was immediately appointed to investigate the incident.
{
9.1 Investication - Root Causes On February 11, 1993, the licensee issued its Incident Investigation Board (IIB) report of the incident identifying the following root causes i
under the following categories:
4 1)
Management System A procedural requirement to not disable interlocks was violated.
i A procedure requiring the use of ECNs to modify equipment that directly involves fissile material was not followed.
The monitoring of the WO program for incorrect usage was
(
f less than adequate.
r f
22 2)
Communications No formal method was available to notify Operations that a i
modification-to process equipment was performed by a W0.
Shift turnover was less than adequate in informing followup shifts of the equipment modification by W0.
1 3)
Trainino I
Instruction in following the'S0P requirement to not disable interlocks was less than adequate.
4)
Procedures The existence of the powder feeder interlocks and their operation were not identified in -a procedure.
5) f_ouinment Reliability
?
The design / installation of the interlock switch was less j
than adequate because frequent and spurious trips occurred-inviting the switch to be disabled.
't 3
9.2 Other Contributory Causes:
4 r
Other causes may have possibly contributed indirectly to the incident:
i 1)
The " maintenance required" section of the powder prep logs (filled out by the powder prep technician) was being used to report process test.results, thereby making it difficult to use the section to report maintenance problems or if used, not making the-item visible enough to be easily recognized.
t 2)
The 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> schedule worked by the Conversion area shifts can-cause communication problems between the shifts and other support personnel.
9.3 Licensee's Recommendations The licensee's investigation team made the following recommended corrective actions lto prevent recurrence of this incident:
1)
Create a systemized Work Station Training and Operator Qualification Guide to be used to assure that employees are thoroughly trained in-the requirements for safe plant operation.
i 2)
Instigate a periodic review of ECN and work order procedures to re-emphasize the criteria used for determining ECN versus WO usage.
i i
t
23 3)
Re-emphasize the review of W0s to detect items that should be on ECNs.
Eliminate equipment modifications done by W0s and thus provide a 4) formal method (ECN procedure) to inform Operations of all modifications to process equipment.
i 5)
Improve shift turnaround communications relative to equipment modifications.
Re-emphasize to all Manufacturing personnel the 6) importance/ requirements for not disabling any interlock or limit l
switches for any purpose.
Assure that all pertinent interlocks are listed in the respective i
7) operating procedures.
The design of the powder feeder discharge tube needs to be 8) reviewed to assure that the design prevents the dropping of U0, powder into the bottom of the Blender Discharge Hoods.
9.4 Lessons Learned The IIB Report identified the following " lessons learned".
1)
ECNs provide a better format for identifying to Operations that equipment modifications are being done. This allows Operations to revise applicable SOPS and conduct the required training before l
accepting the equipment modifications.
It also allows the-Safety Organization to concur by sign-off, that the equipment modifications meet all safety requirements (Criticality,
=i Radiological, Industrial).
A systemized Work Station Training and Operator Qualification 2)
Guide is needed to assure that employees are thoroughly trained in the requirements for safe plant operation.
Engineering needs to provide more " comprehensive long term 3) engineered fixes" rather than " quick fixes" for production problems. Usually quick fixes are not satisfactory and can lead to more serious problems.
9.5 Restart Actions l
The IIB Report identified the following actions to be taken prior to the.
restart of the powder feeders on the PPS.
]
Modify the limit switches so that incidental movement of feeder 1) discharge tube does not activate the interlock.
Plant Engineering to complete by 2/11/93.
?
Functionally test all three interlocks on powder hoppers for 2) f I
i
h 24 l
proper operation. Plant Engineering to complete by 2/11/93.
3)
Conduct an independent walk through of the Conversion area to assure that all limit switches and interlocks are functional.
Plant Engineering to complete by 2/10/93.'
4)
Revise the applicable SOPS to include all pertinent interlocks in i
the powder prep areas. Plant Operations to complete by 2/11/93.
l 5)
Conduct an incident critique and retrain all Chemical Operations i
technicians on the applicable S0Ps which describe the function of interlocks or limit switches and state they are not to be disabled for any purpose.
Plant Operations to complete before each crew is allowed to start their shifts.
6)
Review all outstanding W0s to assure that equipment modifications per ECN Procedure EMF-8581.13 are being done on a ECN.
Plant i
Engineering to complete by 2/10/93.
7)
Specify the use of the ECN procedure for modifications; reserve W0s for repair or replacement work. Manufacturing Engineering to t
complete by 2/08/93.
i 9.6 followup Actions The following followup actions were identified by the licensee:
1)
A Work Station Training and Operator Qualification Guide is being prepared by Plant Operations. Plant Operations to implement by 3/15/93.
l 2)
Provide training / audit of ECN and work order procedures (EMF-858 1.13 and 1.21) to re-emphasize the criteria used for determining ECN versus WO usage. Manufacturing Engineering to complete
{
initial training by 3/15/93.
3)
Re-emphasize to all Manufacturing personnel the importance/
I i
requirement for not disabling any interlock or limit switch for any purpose.
Plant Manager to complete by 3/01/93.
f 4)
Revise " Plant Operations Rules" EMF-22 1.1.1 (P66,318), " Master -
Safety Rules", EMF-30 Chapter One (P65,501), and " Criticality l
Safety Guide Rules", EMF-30 Chapter Three Appendix 9, to reflect i
the statement that " interlocks are not to.be disabled for any i
reason".
Plant Operations and. Safety, Security, and Licensing to complete by 2/28/93.
t 5)
Assure that all pertinent interlocks are described in respective -
Plant Operations SOPS. Plant Engineering and Plant Operations.to I
complete by 3/15/93.
6)
Review alternate designs for the powder feeder discharge tube f
l l
r l
I
~
25 system for possible improvements or simplification to assure that the design prevents the dropping of U0, powder into the bottom of the Blender Discharge Hoods.
Plant Engineering to complete by 4/15/93.
7)
Revise the Powder Prep Logs to provide a separate section to record all required information. This will allow equipment problems to be highlighted in the " maintenance required" section.
Plant Operations to complete by 2/18/93.
8)
Form an independent task force to review the operation and l
condition of other interlock / limit switches in the plant.
Plant Engineering to complete by 2/10/93.
9)
Review the 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> shift implementation.
Plant Operations to complete by 3/30/93.
9.7 Causal Factors Task Group
(
During the entrance meeting, the licensee stated that a causal factors group was formed in addition to the IIB. This group was to evaluate
.j generic implications of the event and recommend actions. The group had not issued its findings as of the conclusion of this inspection.
j 9.8 Conclusions
,f Licensee management's response to the event was prompt and thorough.
Findings of the IIB were consistent with those of the AIT. Actions i
being set into motion appear appropriate.
10.0 NRC - Summary Conclusions The AIT reached the following overall conclusions relative to the j
February 7th event:
1)
The spill of powder was caused by the dislodge of the feed tube from the hopper and subsequent failure of the limit switch to turn l
off power to the feed screw. The limit switch had been defeated by an unknown individual taping the. switch to prevent its i
actuation; an action prohibited by procedure, j
2)
Taping the switch was most likely an act of frustration by an-I operator in an imprudent attempt to improve system. operation.
3)
No training had been provided to operators specific to the limit-switches. Generic training weaknesses in Operations have been previously_ recognized by HRC and the licensee and are reflected in the licensee's IIB report.
4)
Lateral and vertical communications within Operations relative to'
]
the switches was poor.
.j l
q
i l
26 5)
Work control was weak. The switches were installed using a WO rather than an ECN. This action was not in accordance with the licensee's procedures and may be a prevalent, unacceptable practice at the facility to expedite work.
The licensee is 7
developing an action plan to correct this weakness.
6)
Use of ti e WO to install the switch bypassed administrative controls that would have provided for operator training and procedure modifications. This is a significant contributor to the cause of the event.
i 7)
Installation of the switches was not reflected in any licensee design or vendor drawings thereby raising questions on the quality of configuration control.
8)
The event in itself did not present a criticality safety hazard.
However, the licensee had not performed an adequate evaluation to support its assertion that the intrusion of moderating liquids (other than from fire fighting) into the affected areas was a non-credible accident.
9)
The licensee's response to the event was prompt and thorough.
t 11.0 Open Public Exit Interview The inspection scope and findings were summarized with the individuals denoted in Appendix B, on February 12, 1993, during an open meeting in Richland, Washington.
The AIT leader presented the findings and conclusions identified in this report. The AIT leader emphasized that the NRC viewed the intentional defeating of a safety device such as limit switches or interlocks as a very serious matter, and that there was a need for some critical self-examination by the licensee relative to the taping of the limit switches.
The AIT's observations were acknowledged by the licensee. The Plant Manager noted that the NRC and IIB independently arrived at similar conclusions.
I 4
l l
h
r i
APPENDIX A AUGMENTED INSPECTION TEAM CHARTER SIEMENS POWER CORP. DEGRADATION OF CRITICALITY CONTROL j
ON FEBRUARY 7. 1993 t
The Augmented Inspection Team (AIT) is to perform an inspection to' determine the causes, conditions, and circumstances relevant to a February 7,1993, j
event at Siemens Power Corporation, which involves failure of safety-related equipment (criticality control) and involves questions pertaining to licensee operational and managerial performance.
In particular, the inspection should i
accomplish the following:
i 1.
Develop a complete description of the event, develop a detailed sequence of events that occurred during the failure, and identify all-equipment j
failures and human errors that occurred during the event and during event recovery.
2.
Determine the specific circumstances and events which led up to the failure of the limit switch on the hopper at the discharge io the blender on line 3 powder preparation and the discharge of 123 kilograms of 00, powder to a unfavorable geometry container. Determine, i
additionally, the relationship of these circumstances to any previous i
similar events.
Interview licensee personnel as appropriate.
l 3.
Determine the adequacy of the licensee's program / method of identification, implementation and tracking of corrective actions' (including safety-related upgrades to existing processes).
l 4.
Verify and evaluate the licensee *s immediate actions following this I
event. Operations and management effectiveness are to be evaluated.
l 5.
Identify and evaluate the procedures used by the licensee which address installation and maintenance of safety-related controls. Determine the i
effectiveness of, and adherence to, these procedures as they. relate to l
the current event.
J 6.
Evaluate the human factors aspects associated with this event, such as,
?
personnel training and briefings associated with the installation of safety-related controls.
7.
Evaluate the circumstances surrounding the review, approval, and coordination of the corrective actions which led to the installation of the limit switch including a review of the criticality safety analysis,
[
criticality safety specifications and procedures associated with the related equipment.
8.
Within the time constraints of the AIT, evaluate the effectiveness of
}'
the licensee's root cause investigation of this event, including:
Thoroughness and adequacy of the licensee's evaluation and r
a.
corrective actions.
i b.
Adequacy of the licensee's determination of any equipment design l
f
6
\\
2 I
t i
or interface problems.
9.
Provide a Preliminary Notification upon initiation of the inspection and an update at the conclusion of the inspection, j
l 10.
Prepare a special inspection report documenting the results of the above l
activities within 30 days of the start of the inspection.
r I
t i
d k
I i
w
?
h a
-i
[
e A
4
}
t b
1 i
l M
3 1--
~
P APPENDIX B Persons Contacted i
Siemens Power Corporation
- B. N. Femreite, Plant Manager
- M. K. Valentine, Manager, Manufacturing Engineering
- R. E. Vaughan, Manager, Safety, Security and Licensing
- R. L. Feuerbacher, Manager, Plant Operations
- W. A. Baker, Manager, Public Relations L. J. Maas, Manager, Regulatory Compliance T. R. Blair, Manger, Electrical / Instruments Engineering W. G. Keith, Manager, Mechanical / Chemical Engineering i
C. D. Manning, Criticality Safety Specialist r
J. B. Edgar, Staff Engineer, Licensing J. H. Phillips, General Supervisor, Chemical Operations T. C. Probasco, Safety Supervisor T. C. Luzzo, Electrical Engineer M. A. Moberg, Shift Supervisor, Chemical Operations L. G. Stephens, Shift Supervisor, Chemical Operations J. J. Payne, Shift Supervisor, Chemical Operations S. D. Haug, Acting Shift Supervisor, Chemical Operations J. J. Korenkiewicz, Powder Preparation Operator R. C. Arnn, Process Operator J. W. Ayers, Powder Preparation Operator B. L. Ball, Control Room Lead Operator J. L. Newton, Control Room Lead Operator S. V. Bryson, Powder Preparation Operator l
+
S. L. Colby, Control Room Lead Operator
- Denotes those attending the open exit interview on February 12, 1993.
In addition to the individuals noted above, the AIT members met and held-i discussions with other members of the licensee's staff.
i 1
j i
APPENDIX C General Line 3 Powder System FMER rit.T g g C=
FatyrA
,n.
=
\\4 k+
Slab Hopper Slab Hopper N /-
4 f#
e L
S t
)
?
Powder Preparation System a
e
+
+
T Blender c
3 l
v
\\/
~
-~
--~__
Transfer Box g
'l Feed Tube f
l Lexan Enclosure h
Boot I
l Hose Clamps y
i Dump-
[
u to E
Dump to i i
'G j)
Feed Hopper /
Drums I
Drums A
e/
5 0 ON l-
'gFeed tt g ' "I Screw NOTE:
Drawing is not to scale l'
Limit AirIntake)
Switch k
Floor Level l
.l l
~
APPENDIX D
/pue%
UMTED STATES NUCLEAR REGULATORY COMMISSION
^
[..,I REGION V
'\\
/
1450 MARIA MNE WAUJUT CREEK,CAUFORNIA 94s96-53G8 g
Docket:
70-1257 License: SNM-1227 CAL:
5-93-011 Siemens Power Corporation 2101 Horn Rapids Road P. O Box 130 Richland, Washington 99352-0130 Attention: Mr. B. N. Femreite, Plant Manager
SUBJECT:
CONFIRMATORY ACTION LETTER I
' Based on the findings of our Augmented Inspection Team during the week of February 8,1993, the NRC has developed a specific concern related to criticality controls for your uranium oxide blending and powder preparation system. Our concern involves the potential for moderator intrusion into the area of your powder preparation lines 2 and 3, a potential that we feel has not received sufficient technical evaluation. We believe this matter deserves your immediate attention, in view of (1) the configuration of the vacuum suction line, (2) the recently demonstrated potential for inadvertent powder spills due to line disconnects, and (3) the presence of water lines in relatively close proximity to the equipment of concern.
Pursuant to a telephone conversation on February 11, 1993, between yourself, F. Wenslawski (Augmented Inspection Team Leader), and Mr. R. Scarano (Director, Division of Radiation Safety and Safeguards), we understand that you have taken (or will take) the following actions:
1.
Siemens Power Corporation will perform an engineering evaluation of the potential intrusion of moderating liquids into the uranium oxide blending and powder preparation system, and define the specific controls necessary to prevent criticality.
2.
Pending completion of the above review, and the satisfactory resolution of NRC concerns:
Siemens will use an alternate method to manually vacuum transfer a.
blended powder (via 45-gallon borated drums) to the powder preparation system, and will tag the " accurate" feeder system out of service.
b.
If Siemens elects to use the " accurate" feeder system, Siemens will dedicate, on a continuous basis, a qualified operator at the feeder station to observe operation of the feeder and to be watchful for the intrusion of moderating liquids. This operator will be trained in his/her duties and responsibilities relative to this assignment.
t c.
Actions 2.a and 2.b, above, are applicable to Conversion Line 2,
~
I
i 2
which includes Powder Preparation Lines 2 and 3.
These actions will be in effect as of your receipt of this letter.
Pursuant to Section 182 of the Atomic Energy Act, 42 U.S.C. 2232, and 10 CFR 2.204, you are required to:
1)
Notify me immediately if your understanding differs from that set forth above, 2)
Notify me if for any reason you cannot complete the actions within the specified schedule and advise me in writing of your modified schedule in advance of the change, and 4
3)
Notify me in writing when you have completed the actions addressed in this Confirmatory Action Letter.
Issuance of this Confirmatory Action Letter does not preclude issuance of an order formalizing the above commitments or requiring other actions on the part of the licensee. Nor does it preclude the NRC from taking enforcement action for violations of NRC requirements that may have prompted the issuance of this letter. In addition, failure to take the actions addressed in this Confirmatory Action Letter may result in enforcement action.
The responses directed by this letter are not subject to the clearance
+
procedures of the Office of Management and Budget as required by the Paperwork Reduction Act of 1980, Pub. L. No. 96-5)).
In accordance with 10 CFR 2.790 of the NRC's ' Rules of Practice," a copy of this letter and its enclosures will be placed in the NRC Public Document Room.
Sincerely, T
I* Regional Adminfstrator I
cc:
State of Washington i
I i
-