1) Background
Clogging of containment sump screens by fibrous material such as thermal insulation can lead to total loss of emergency core cooling and containment spray systems. This well-known safety issue is common to all LWR units worldwide, as an important common mode on safety systems, and had not sufficiently been taken into account at design stage of most units currently in operation. This is also true for VVER 440/213 for which debris coming from thermal insulation can clog sump screens (See IAEA-EBP-WWER-03 “Safety issues and their Ranking for WWER 440, Model 213 NPPs”, April 1996, issue number S5).
The containment is equipped with sumps to collect water of primary circuit and of water storage tank after a LOCA in order to recirculate water in the second phase of the accident. The sumps are covered with screens that are intended to prevent debris penetration to safety systems that could impair their performance and water circulation in the core. The thermal insulation used inside containment (that can be damaged by LOCA jet effect) and the total area of the screens above sumps together with dust and dirt that occur in containments form a combination that raises high safety concerns regarding the possibility of maintaining ECCS circulation after a medium or large break LOCA.
Operational experience based on events in Sweden (Barsebck NPP in 1992) and in USA has demonstrated that even a relatively small amount of fibres can efficiently block a large screen area.
On VVER units, mineral wool has been widely used for thermal insulation. Tests at Loviisa NPP (Finland) and at Zaporozhe NPP (Ukraine) have confirmed this risk.
Many entities worldwide performed research and engineering activities dealing with sump screen clogging.

2) Objectives
The present TACIS project U1.01/97A aimed at implementing sump screens with improved design for VVER 440/213 units at Rovno NPP (units 1&2) in order to cope with clogging risk.
This project followed the PHARE project PH2.05/95 which aimed at proving experimentally that the debris retention function of the containment sump filters can be ensured without risk of clogging and without transportation of particles to the reactor core with risk of partial flow blockage.
For this purpose, U1.01/97A had to compare actual insulation and sump filters to the ones of project PH2.05/95 (Dukovany, Paks and Bohunice NPPs, all of type VVER 440/213) in order to limit as far as possible new large analysis and to dedicate the main part of the project to the development and replacement of Rovno NPP sump screens.

3) Project Results–Achievements

Preparatory phase:
From 1997, TS (Technical Specification) was prepared by OSA utility EDF (see general OSA contracts U1.01/96 and U1.01/97) as the technical basis for future call for tenders. After a long process of updating and validation by all counterparts of the project (Rovno NPP, Energoatom, UNRA – Ukrainian NRA, EC, JRC), final version of TS approved by all was issued in May 2000, together with EC (Evaluation Criteria).
Then tender dossier was prepared, and call for tenders was launched in February 2001. A few months later, tender evaluation took place in EDF engineering centre in Lyon, France, and consortium Framatome ANP/Fortum was awarded.
TACIS supply contract between Energoatom and consortium Framatome ANP/Fortum was signed on 22nd January 2002, which is the milestone for the beginning of the supply contract.

Implementation of the supply contract:

Tasks to be performed by the contractor were the following:

• Establishment of a work plan and a quality assurance plan
• Studies for implementation of a basic design of new sump filters (phase 1)
• Studies for implementation of the detailed design of new sump filters (phase 2)
• Manufacturing of new sump filters
• Supply of new sump filters at Rovno NPP
• Qualification of new sump filters
• Support to licensing
• Supervision of on-site installation
• Writing and delivery of design and operational documentation
• Warranty period (still ongoing)

Tasks within consortium were split as follows:

• Framatome ANP: leader of the consortium, general project management, implementation of contractor’s activities dealing with transport, installation, acceptance tests and warranty;
• Fortum: studies for basic design and detailed design, with relevant documentation;
• Local sobcontractor EUM/Inconsatom: manufacturing, transport, support to installation and commissioning, with relevant documentation.

Other partners have been involved in the implementation of the project, outside contractor’s scope:

• Rovno NPP: validation of all steps of the project, installation, commissioning, licensing;
• Energoatom: validation of all steps of the project, licensing;
• KIEP: studies of integration of new filters to Rovno NPP;
• EDF (OSA utility): general follow-up of the project, and general support;
• UNRA + Riskaudit: licensing.

Note: “2+2 approach” has been used during implementation phase of the contract, i.e. support of an Eastern plant by a Western Utility and parallel support of the Eastern NRA by a Western TSO in the same projects. Thus, consortium Riskaudit supported UNRA during this project, and participated to many meeting during this project. This support is funded by
TACIS in a separate contract.

Time schedule of implementation phase:

• 22nd January 2002: supply contract signature;
• project kick-off meeting at Rovno NPP on 31st January 2002;
• prototype validation tests (studies phase 1) performed successfully in January 2003 in Fortum premises;
• approval of detailed TS by UNRA in July 2003;
• signature of manufacturing contract between consortium Framatome ANP/Fortum and EUM/Inconsatom in August 2003;
• factory acceptance tests in February 2004 for sump filters of unit 1, with some reservations from Framatome ANP;
• delivery in April 2004 at Rovno NPP of sump filters for unit 1;
• installation of sump filters for unit 1 in May 2005;
• provisional acceptance of sump filters for unit 1 was issued on 11th May 2004;
• factory acceptance tests in June 2004 for sump filters of unit 2;
• delivery in July 2004 at Rovno NPP of sump filters for unit 2;
• installation of sump filters for unit 2 in July 2005;
• provisional acceptance of sump filters for unit 2 was issued on 20th July 2004;
• final acceptance certificate for services has been issued on 15th July 2004;
• 3-year warranty period is still running, and if no problem happens during warranty period, final acceptance certificates for hardware should be delivered in May 2007 for unit 1 and in July 2007 for unit 2.

The whole supply project U1.01/97A has been implemented in consistency with initial schedule.
Technical description of the supply:
Each unit has 3 sumps (VVER 440/213 is a 3-train design). So, 6 sump screens have been replaced for the whole project.
Each new sump screen is made of several (between 15 and 25) vertical hollow filtering panels. Outside part of each vertical filtering panel is upstream part of the screen where insulation debris is accumulated in case of LOCA, and inside part is downstream. All inside parts of panels are gathered into a flow channel leading to the sump inlet.
All filtering panels are arranged vertically in parallel, and are enclosed in a big grating, preventing damage of filtering panels by ingress of big loose parts inside containment. Each filtering panel mesh is made of single mesh less than 1 mm. Total filtering surface for each sump screen in between 22 and 30 m2.
All new sump screens are made of stainless steel. No I&C was provided. Sump screens are seismically qualified.
Assumed amount of produced debris is very conservative: the most penalising LOCA location is assumed, and all insulation in a radius of 2.85 meter from LOCA place is assumed to be destroyed into debris. In addition, all debris are assumed to be routed to the same sump screen.
With these conservative assumptions, tested prototype (phase 1) showed that maximal clogging ΔP in 11 kPa, which gives a large design margin, as maximal requested value of ΔP is 20 kPa from ECCS pump design.

4) Comments
Preparatory phase of the project took a very long time:
• Validation of TS by all counterparts took more than 2 years;
• Tendering and contracting process took about 1 year.
In comparison, implementation phase has been performed on schedule.

Further information
Further information on the project results could be sought from Energoatom and Rovno NPP.
The Project Final report is available at the JRC-IE archive.