Protecting People and the EnvironmentUNITED STATES NUCLEAR REGULATORY COMMISSION
UNITED STATES
NUCLEAR REGULATORY COMMISSION
OFFICE OF NUCLEAR REACTOR REGULATION
WASHINGTON, D.C. 20555
August 11, 1988
Information Notice No. 88-61: CONTROL ROOM HABITABILITY - RECENT REVIEWS
OF OPERATING EXPERIENCE
Addressees:
All holders of operating licenses or construction permits for nuclear power
reactors.
Purpose:
This information notice is being provided to alert addressees to potential
problems resulting from design or analysis deficiencies identified in control
room ventilation systems. It is expected that recipients will review the
information for applicability to their facilities and consider actions, as
appropriate, to avoid similar problems. However, suggestions contained in
this information notice do not constitute NRC requirements; therefore, no
specific action or written response is required.
Description of Circumstances:
The NRC has received several construction deficiency and licensee event
reports pertaining to safety systems that are used to ensure control room
habitability. These reports identified potential safety concerns resulting
from design deficiencies, which were attributed to inadequate analysis and an
inability to justify those conditions that were assumed in previous evalu-
ations of plant design and operation. These reports are summarized below.
Comanche Peak 1 and 2:
On January 15, 1988, the permit holder determined that radiation doses to
control room operators for some postulated radiological accidents could exceed
the limits of General Design Criterion 19 of Appendix A to 10 CFR 50. This
determination was attributed to an inadequate analysis of control room
habitability systems for postulated radiological accidents. Particular ac-
cident scenarios that were incompletely analyzed included a fuel handling
accident and a rupture of a radioactive gaseous waste tank. To correct this
situation, the licensee is developing new calculations, upgrading the existing
control room intake radiation monitors and associated cables to safety-related
Class 1E requirements, and installing two additional safety-related Class 1E
radiation monitors, one in each control room intake.
8808050248
.
IN 88-61
August 11, 1988
Page 2 of 5
Quad Cities:
On November 25, 1987, the licensee's engineering department discovered that
design basis assumptions used in the control room habitability study were
inconsistent with technical specification requirements. The study was con-
ducted to satisfy a requirement of the NRC's TMI Action Plan. The adsorption
efficiencies of the standby gas treatment system and control room HVAC system
were assumed at 99 percent for organic iodide removal. The relevant technical
specifications, however, require only an organic iodide removal efficiency
more than or equal to 90 percent. All the filters meet the technical
specification requirements. Since December 31, 1984, tests of filter
efficiencies indicate that the relevant assumptions of the study were met with
two exceptions. The licensee attributes the cause for this condition to be an
inadequate review of design and analysis during the development of the study.
Vogtle 1 and 2:
On July 2, 1987, plant engineering personnel identified an inadequacy in the
dose analysis for control room operators. On receipt of a safety injection
signal or a control room outside air intake high radiation signal, the control
room heating, ventilating, and air conditioning (HVAC) system is automatically
transferred from the normal system units to the essential control room (ECR)
fan-filter units in the emergency mode of operation. The ECR system consists
of two redundant and physically separated 100 percent capacity fan-filter
units for each side of the control room, associated with the corresponding
reactor unit (four for the combined Unit 1 and Unit 2 control room (Figure
1)). Each of the ECR units belongs to a different safety train, but portions
of the outside air intake ductwork and control room supply and return ductwork
are common to each of the units. On initiation of the ECR system, the
associated motor-operated dampers for the essential units are automatically
opened and those for the normal units are automatically closed.
When both of the ECR fan-filter units are operating, loss of power to one of
these units can reduce the amount of outside air available to pressurize the
control room. This can happen because the dampers losing power fail in the
"as-is" position. At the time of the identification of the problem, no back-
draft dampers were installed. This degrades the ECR HVAC system by
establishing flowpaths through the common ductwork and the ductwork of the
failed unit back to the suction of the operating unit, which could potentially
reduce the outside air flow to the control room and reduce the control room
pressure below the design value. Maintenance of the design control room
pressure is required to minimize unfiltered inleakage.
The licensee concluded that the actual effect on control room pressure could
not be calculated in the absence of test data and decided to modify the system
by installing backdraft dampers, as noted on Figure 1, for the ECR HVAC
systems of both Units 1 and 2.
.
IN 88-61
August 11, 1988
Page 3 of 5
On July 4, 1987, an additional condition was discovered that could have caused
an insufficient control room pressure. All ECR systems share common outside
air supply ductwork. The common air supply ductwork has intakes from the out-
side atmosphere associated with both Units 1 and 2. Redundant isolation
dampers in series are provided for both Unit 1 and Unit 2 duct openings from
the outside atmosphere. During construction, the Unit 2 duct opening had been
isolated by locking the dampers closed. If one of the outside air isolation
dampers for Unit 1 had closed as the assumed single failure of an active
component, no source of outside air would be available to the control room
(shared by both Units) and the required pressure would not be maintained.
Because the licensee had removed chlorine gas sources from the site, the
capability to isolate toxic gases was no longer needed. Thus, the licensee
deactivated and tagged open the outside air isolation dampers.
For all these problems, the licensee concluded that the defective conditions
would have been discovered earlier with an adequate failure modes and effects
analysis.
Beaver Valley 2:
On March 27, 1987, the permit holder determined that timers initiated by con-
tainment isolation phase B signals were not served with uninterruptible power
as required. These timers are designed to actuate banks of compressed air to
supply the control room emergency pressurization system one hour after receipt
of the isolation signal. They were powered from the respective fan control
circuit energized by an emergency ac distribution panel fed from an emergency
motor control center. Loss of offsite power would interrupt power to the
motor control center. If a loss of power occurred after timer initiation, the
timers would reset to the beginning of the timing cycle and would not begin
the cycle until power was restored to the emergency motor control center. The
compressed air would thus not be supplied after one hour. To enhance the
reliability of the safety systems, the licensee revised the circuitry to power
the timers from the Class 1E 125-V dc battery system.
McGuire 1:
On November 5, 1987, the licensee discovered during an 18-month surveillance
test of the control area ventilation and chilled water system that control
room pressure was below the technical specification requirement. The licensee
determined that the cause was leaking seals on seven control room doors. The
doors were designed to seal by seating against sealing strips in the door
frames. Although not visibly deteriorated, the sealing material apparently
had been deformed and compressed over time from normal use of the doors. The
licensee adjusted manual volume dampers to increase total train air flow and
the proportion of outside air flow to the maximum 60 percent allowed in the
.
IN 88-61
August 11, 1988
Page 4 of 5
test procedure. This action was effective in increasing control room pressure
to meet the technical specification requirement. The licensee installed
sealing tape around the seven doors and established a preventive maintenance
program to inspect the seals on all control room doors every 6 months for a
1-year period. The licensee plans to determine an appropriate frequency for
preventive maintenance of the door seals. The surveillance test of the
control area ventilation and chilled water system will also be repeated every
6 months until sufficient information is obtained for determining an
appropriate frequency.
Farley 1:
On June 5, 1987, the licensee discovered that none of four fire dampers in the
control room ventilation system would fully close and latch with or without
air flow because they had not been exercised and/or lubricated or, in one
case, because a latch was damaged. In addition, investigation revealed that
the dampers would not have received an actuation signal from a Firestat set to
sense 160xF in the control room ceiling because of installation errors. The
licensee attributed the problems to three causes: a design deficiency
pertaining to the full closure and latching of the dampers, inadequate testing
of the actuation circuitry, and inadequate preventive maintenance of the
dampers.
Discussion:
In addition to the above, the NRC has recently completed an engineering evalu-
ation, "Design and Operating Deficiencies in Control Room Emergency
Ventilation Systems," AEOD/E802, April 1988 based on recent events that
highlight single failure vulnerabilities in control room emergency ventilation
systems.* The NRC also has recently conducted a survey of control room
habitability systems at 12 operating plants. Numerous discrepancies were
found between the analyzed and actual performance of these systems. For
example, differences exist between design, construction, operation, and/or
testing of these systems and the descriptions and analyses provided in
licensing documents, as for example in the assumptions used in the toxic gas
and radiation dose calculations. In addition, analyses assumptions have not
always been consistent with technical specification requirements. The NRC has
issued several information notices related to this subject (see Attachment 2).
Resolution of Generic Issue 83: Control Room Habitability is also ongoing.
*A copy of the report is available in the NRC Public Document Room, 1717 H
Street, N.W., Washington, D.C. 20555, for inspection and copying.
.
IN 88-61
August 11, 1988
Page 5 of 5
No specific action or written response is required by this information notice.
If you have any questions about this matter, please contact one of the techni-
cal contacts listed below or the Regional Administrator of the appropriate
regional office.
Charles E. Rossi, Director
Division of Operational Events Assessment
Office of Nuclear Reactor Regulation
Technical Contacts: Vern Hodge, NRR
(301) 492-1169
Charles R. Nichols, NRR
(301) 492-0854
Attachments:
1. Figure 1. Schematic Air Flow Path Diagram of Essential Control Room
Heating, Ventilating, and Air Conditioning System (ECR HVAC) at Vogtle.
Assumed Containment Isolation Signal for Unit 1 with Train B Failed.
2. List of Information Notices Related to Control Room Habitability Systems
3. List of Recently Issued NRC Information Notices
.
Attachment 2
IN 88-61
August 11, 1988
Page 1 of 1
LIST OF INFORMATION NOTICES
RELATED TO CONTROL ROOM HABITABILITY SYSTEMS
NO. TITLE DATE
86-76 Problems Noted in Control Room August 28, 1986
Emergency Ventilation Systems
85-89 Potential Loss of Solid-State November 19, 1985
Instrumentation Following Failure
of Control Room Cooling
83-62 Failure of Redundant and Toxic Gas September 26, 1983
Detectors Positioned at Control Room
Ventilation Air Intakes
.
