ML19345B200
| ML19345B200 | |
| Person / Time | |
|---|---|
| Site: | Summer  |
| Issue date: | 11/10/1980 |
| From: | Tedesco R Office of Nuclear Reactor Regulation |
| To: | Nichols T SOUTH CAROLINA ELECTRIC & GAS CO. |
| References | |
| NUDOCS 8011260326 | |
| Download: ML19345B200 (19) | |
Text
._ _
%((
w Kloy n
/
UNITED STATES
! \\ > "s,(
NUCLEAR REGULATORY COMMISSION g.' N
...t WASHINGTON, D. C. 20555 o
y,;/~ /
NOV 10 G80 5
o y
5 lO Docket No. 50-395 gj 0E!
?
Eb
~
Mr. T. C. Nichols, Jr.
Efd Vice President and Group Executive 3
Nuclear Operations 2;
o South Carolina Electric and Gas Company Post Office Box 764 Columbia, South Carolina 29218
Dear Mr. Nichols:
SUBJECT:
HUMAN FACTORS ENGINEERING CONTROL ROOM DESIGN REVIEW / AUDIT - SUMMER During the week of August 25-29, 1980, a human factors engineering design review / audit was conducted by the NRC staff at the Virgil C. Summer Nuclear Station, Unit 1.
The review was conducted by representatives of the Human Factors Engineering Branch who were assisted by our consultant, Bio-technology, Inc.
Enclosed is our assessment of the control room design for your facility.
Included are a number of deficiencies which we have identified in the control room design. We will, require that you respond in writing with a program for correcting each of these deficienciet.
Your response should be provided as soon as possible in order to permit us to include our assessment. in the forthcoming Safety Evaluation Report.
If you require any clarification, please contact the staff's assigned project l
manager.
Sincerely, 1
Robert L. Tedesco Assistant Director for Licensing Division of Licensing
Enclosure:
As stated cc w/ enclosure:
See next page B9H260 32-
- = _.
_=.
)
a NOV I 01980 1
Mr. T. C. Nichols, Jr.
I Vice President & Group Executive gt Nuclear Operations South Carolina Electric & Gas Company P. O. Box 764 Columbia, South Carolina 29281 cc: Mr. William A. Williams, Jr.
Vice President i
South Carolina Public Service Authority 223 North Live Oak Drive f
Moncks Corner, South Carolina 29461 i
J. B. Knotts,iberman_q.
Jr., Es Debevoise_& L J
1200 17th Street, N. W.
Washington, D. C._
20036...___
i Mr. Mark B. Whitaker, Jr.
Group Manager - Nuclear Engineering & Licensing South Carolina Electric & Gas Company P. O. Box 764 i
Columbia, South Carolina 29218 Mr. Brett Allen Bursey
~
Route 1, Box 93C Little Mountain, Soutn Carolina 29076 Resident Inspector / Summer NPS c/o U. S. NRC Route 1 Box 64 Jenkinsville,_ South Carolina 2906S O
9
?
l
~
l l
g...
O.
e,
^,-
t 4
e I
MT4AN FACTORS ENGINEERNIG CONTROL ROOM DESIGN REVIEW / AUDIT V. C. SUMMER Durkng the week of August 25-29, a human' fact 6rs engineering design review / audit of the V. C. Summer control room was conducted. The review was performed by the Human Factors Engineering Branch, Division of Human Factors Safety, with assistance from 4
Harold E. Price of Biotechnology, Inc., our human factors' consultant.
At the time of the review the control room at Summer was approximately 3 to 4 months away from canpletion. Many of the systems and subsystems were either not operational, or not installed thus limiting the teams capability to assess the man-machine interface.
The following sections of this report summarize the staff's observations of the control roon design and layout, and of the control room operators interactions with that enviro ment. Where possibib, observed deficiencies were given a subjective ratin;3 based on the potential for that deficiency to result in an operator error during perfonnance of a critical activity. These ratings are divided into three categories:
1.
Major concern for potential operator errors that could affect plant safety.
2.
Moderate concern for potential operator errors that could affect plant safety.
3.
Minor concern for operator errors that should not affect plant safety.
- 1. 0.
GENERAL OVERVIEW a.
Labeling and other aids for recognition and identification such as demarcation and color coding is inadequate at present., Excessive time l
and error potential exist in searching or and identifying systems, subsystems, groups or components.
b.
Most controls and associated displays are in the same general area, but layout orientation is not consistent. Series of controls (or displays) are not arranged in sterotypical patterns and identification of displays associated with controls is not clear.
i i
c.
Controls were not easy to recognize and identify (particularly CMCs); and, i,
when a series of controls were grouped the orientation was not consistent (i.e., vertical or horizontal, or alpha-numeric order). Clusters of controls were not separated by spacing, demarcation,, color or other recognition aids.
~
d.
Some displays (vertical meter scales) are difficult to read, to interpret values, and require mental conversion. Some meters fail at mid scale:
e.
Poor control room layout (from a visual task and perception standpoint). A structural column is located in the middle of the control
- room, i
f.
General layout of bench consoles, computer console, and future foreraan's desk is good. Traffic patterns are satisfactory and passageway clearences are adequate. Visual access from the principal operators position (near rod f
. -.,... _ ~
_y e...
AAgm DM
'E h
m ea A)UAL r
3 control station) to all bench consoles is good. Visual access from the computer console is good except for the in-core monitoring panel. This system, however, has its own audible alarm to attract operator attention.
g.
Overall, the control room lighting design is adequate. The combination of open grid dort below luminaries and t'a ? ft. X 4 f t. ceiling panels provide i
a co,afort4ble level of illumination, but does red i:: ;nn31 illumination ::n l
on the upper panels. However, this lighting arrangement does not ; rad /c3 excessive glare.
h.
No emergency equipment in the control room. No assessment or caa<;l es ton can be made as to the adequacy, accessibility, communication problems, etc.,
af prat-e.tive equipment at this time.
l
- i. The acoustic background (noisa) level is a comfortable 55 to 37 dbA. Alarm levels were about 15 dbA above this level. A potential problem exists vi th plant page system as a source for background noise with a peak of about 76 dbA. Announcemots over this sytems tend to reveberate throub,.it the control room.
2.0 THE MOST SIGNIFICAttf iLNAN FACTORS OEFICIENCIES OBSERVE 0 OW LW NE CONTROL W1 GESIGN REVIEW / AUDIT.
/. 1 CO4TROLS Specific Controls (CMC switches) were not easy to locate and identify (Categorj _
a.
b.
Clusters of controls had no demarcatii, color coding or other recognition aids. (C t tegory 2 ).
c.
Legend covers on CMC sw tches can oe inadvertantly interchanged. (Category 1. ).
i d.
Cannot distinguish momentary contact witches from continuous contact.
a (Category 1 ).
=-
t
^
4 g.
Difficult to identify a specific control in a large group of controls (Category 1 ).
a-h.
Cannot readily distinguish between controls for valves, pumps, fans, etc.
2 (Category 3 ).
i i
1.
Legend labels on the controls are inconsistent (Automatic sometimas means
]
closed) (Category 1 ).
- j. Component number (or other infonnation) that is engraved on the red and green lenses are difficult to read, particularly when the green lenses are not illuminated (Category 2).
I 4
2.2 "J"HANDLESWITCHES(SBM) a.
There are two reactor trip switches on the bench panel. One turns right (cw) to trip the reac' or and the other turns left to trip the reactor t
(Category 2 ).
l b.
It is difficult to read switch position labels while gripping the control (Category 1).
c.
The color flags are difficult to distinguish (Category 2 ).
9 y
-pw--
m,,r w
v ywr--,1r--w-1--
v'-
u-w--'r
--T-"-
y~
g
--ye=-My-
~-
- w.-tw
- Fyve-W
--c-t'p-t-4 9'*
2.3 HAGEN VALVE CONTROLLERS a.
Controllers which violate position convention are confusing with respect to open/close positions (Category 1).
b.
The setpoint scale is difficult to read on the automatic Iontrollers (Category 3 ).
i c.
Some 3 way valves are labeled open and closed when in fact they are always open in the same position (e.g., VCT Level Controller) (Category 1 ).
d.
There is no feedback from actual vs demand signal. This can cause transients when switching to the automatic mode (Category 2 ).
e.
Vernier scales tre difficult to read on manual controllers (Category 3 ).
3.0 DISPLAYS Some displays (vertical meter scales) are difficult to read, interpret a.
values, and require mental conversion (Category 2 ).
1 b.
Dual scales on one instrument are confusing (Category 2 )
c.
Many meters fail at mid scale (Category 1 ).
d.
ESF Status Monitor Panel indicators with low and bright intensity lights are 1
not consistent in meaning or intensity, and are difficult to use for pattern recognition (Category 1).
There are numerous hand made scales (Category 1 ).
e.
1 f.
There are no normal operating ranges or setpoints indicated on meters (Category 1 ).
1
.. 3.1 MODUFLASH DISPLAY SYSTEM a.
The percent scale is difficult to interpret (category 2 ).
I b.
The system cannot be easily related to specific annunicators on the Main Centrol Board.
(e.g., no indication on annunicator tiles as to whether that parameter is monitored by the Moduflash system).
(Categcry 2) c.
Flow readings must he converted fran a precent scale (Category 1).
d.
Distinction between setpoint and flow / temp selection is poor (Category 2).
e.
The crientation of the status indicators on "A" train and "B" train is not consistent (e.g., CC pump SUCT-loop and DISCH loop) (Category 2 ).
4.0 CONTROL / DISPLAY RELATIONSHIP Layout of controls and associated displays have no consistent pattern a.
(e.g., display directly above control or to the right or left)
Category 2 ).
b.
Display is not always within convenient viewing distance when using a related control (e.g., pressurizer pressure indication is several feet from controller) (Category 2 ).
Strings of controls and displays are not consistently oriented (i.e.,
c.
vertical, horizontal, matrix, etc.) (Category 2 ).
w y
e y
g,_,
y-
-r--
w
- + -
w
. 4.1 SERVICEWATERPANEL( 604&605)
SW Pumps A, B, C have a Eonfusing instruction plate on the panel a.
(e.g., the bottom line says:
" Pull to lock in slow speed pull position." Misinterpretation could cause operator to Tock the pump in slow speed position when it should be locked out of senice)
(Category 1 ).
.b.
SW Pump "C" Discharge status light is essentially a caution or warning, yet it is a white legend and is not located adjacent to pump switch (Category 2 ).
c.
SW Pump Alignment indicators do not illuminate either red or green (some had both illuminated). These indicators should be set to illuminate either all red or all green for each train (Categon 1 ).
4.2 H.V.A.C. PANEL a.
Mimic is color coded but the coding is not carried out over the controls and the panels (Category 2 ).
b.
Damper Controls and Mimic for LOOP A, LOOP B, TRAIN A, TPAIN B, do not appear to correspond. Cannot tell on the mimic what is a loop or what is a train. The control labeling is very confusing and not apparent on the mimic (Category 2 ).
c.
There are large numbers of fan controls grouped together and it is not -
easy to find a particular fan by number. Also, the arrangement of those controls does not correspond to their relative position on the 1
mimic (Category 2 ).
j
i
- d.
Considerable glare on mimic,~particularly at the top of HVAC panel (Category 2 ).
Mimic board is too high for a 5th percentile person to rend (Category 3 ).
e.
f.
CMC component identification numbers are hard to read when legend is not illuminated (Category 3 ).
4.3 REACTIVITY CONTROL SYSTEM PANEL (620 & 621) a.
BANK SELECTOR control position indicator is difficult to detennine j
because of poor labeling (Category 1 ).
b.
NIS RECORDER PEN SELECTORS control position is difficult and awkward 1
to determine because alpha numerics are sideways.
Also, selector control must be turned though SOURCE range to go froa. INTERMEDIATE TO 1
POWER (Category 2 ).
c.
Part length Rod control is not used. (Category 2 )
d.
Boric Acid Pump controls are 6-8 feet away from BORATE control yet they may have to be used if emergency borate flow is inadequate. (Category 1 ). -
There are 8 vertical meters for POWER, INTER & SOURCE range reading e.
% RANGE and % FLUX. These generally must be compared for differences and are not well designed scales for this purpose (Category 2 ).
.r
. 4.4 EME?.GENCY FEE 0 WATER SYSTEM PANEL ~T622 & 623)
~
a.
Arrangement of control strings is not always optimum EF Isolation Frcm FDP to SG =
A B
C EF Isolation From TDP to SG =
B A
C
.(Category 1 )
b.
Minimum specified turbine flow of 100 GPM is at very bottom of meter scale (Category 2 ).
EF SG Pressure and SG Pressure (Narrow Range) must be compared but c.
they are two different types of scales, even though they cover the same range of values (Category 1 ).
d.
Identification of' A, B, C, steam generator :on.ponents is difficult (Category 2 ).
1 4.5 FEEDWATER SYSTEM PANEL (625)
The switch handles for the bypass controls are the same as the control switch. Search time might lead to confusion (Category 2).
e
++
w
~
w
- + -
1 4.5 FIRST OUT PANEL (626) 4-There is no identification of the first out alarm panel on the control board (Category 1).
4.7
' CONDENSATE PUMP PANEL (627)
The control board layout could lead to confusion because the master condensate controller is below pump C (Category 2).
4.8 i AIN STEAM PANEL (629 & 630)
The main steam isolation valves are identified A, B, C horizontally while the status light indicators show channels I. II, III.
(Category 1).
4.9 RA3IATION MONITORING SYSTEM Detectors are not associated with their related annunicators, such as, detectors RMA-11 is located in Section No. 4 of the pane.1 while the corresponding annunicator is located in Section No. 5 RMA-6 is located in Section No. 5 of the panel while the corresponding annunicator is located in Section No. 6.
(Category 2).
4.10 STEAM GENERATOR PAMEL (624) i a.
Reset for Feedwater Pump A', B. I n C is difficult to identify.
It is
~
within a cluster of several CtC switchs.
(Category 1).
l
11 -
- b.. By-Pass Flow Meters are difficult to identify and read when operating flow controllers (Category 2 ).
There are 3 sets of meters which are used for comparison thecking c.
and which have similar but different scales.
These are-Main Steam HDR Press (0-1300 PSIG); FW Pump DISCH HEAD (0-1500 PSIG); FW Pumps A, B, C (0-1800 PSIG) (Category 1 ).
d.
The master controls for a series of valves is located on the left hand side for (Feedwat' r Pumps,) and on the right for others (e.g., Condensate Master e
Pump).
Also, " Master" displays are located inconsistently (e.g., FW Boost Pump DISCH Press is in the middle) (Category 2 ).
4.11 POWER PANEL (636 thru 638)
On the Diesel Generator A controls, the " Exciter Reset Shutdown" label a.
should be labeled " Engine Shutdowt. :xset" to be consistent with diesel generator B controls (Category 2 ).
b.
Most wattmeters and voltmeter indicators on the panel are miniature and wide range and do not pennit accurate readings which are sometimes required (Category 3 ).
5.0 LABELING AND CODING There is no hierarchial structure of labeling (e.g., from System, Subsystems.
a.
to the ca.'conent label) (Category 1 ).
,p 7
1 e
b.
Little or no use of demarcation lines (Category 3 ).
c.
Use of mimic diagrams on the control panel is minimal (Category 3 ).
d.
Labels are not consistently positioned above, below, tb the right or anywhere (Cate' gory 1 ).
Labeling techniques are not consistent'(e.g... black on white and white on e.
black labels, etc.) (Category 1 ).
f.
Labels are not permanently attached (Category 1 ).
g.
Labels tend to be wordy, inconsistent in the use of abbreviations, and have some misspelled words.
(Category 1) 4 h.
Some labels are missing (Category 1 ).
- f 1.
Some labels are obscured by controls (e.g., Barre.1 Switches) (Category 1 ).
- j. Essentially no color coding for location and identification of systems or functions is used (Category 3 ).
k.
Labels under low mounted strip chart recorders cannot be seen (Category 1 ).
1.
Qual function vertical scales are not clearly labeled to identify the function of each scale (Category 1 ).
6.0 WORKSPACE AND ENVIRONMENT a.
Visual access from the Shift Foreman's desk to main control board is restricte by a support column (Category 3).
. b.
Storage of essential material (emergency operating procedures) is undefined (Category. ).
c.
There is inadequate knee space a't computer console (Category 3 ).
d.
There is no emergency equipment in the contrcl room (Category 1 until reviewed) 6.1 HAZARDS a.
Mats arcured the bench console and long telephone cords present a potential tripping hazard.
(Category 1 )
b.
"J" Handle controls (e.g., diesel generator panel) are mounted close to the edge of the bench boards. They present a hazard to personnel as well as inadvertent control actuation (Category 1 ).
6.2 CCMMit41 CATIONS a.
Comunications to the Emergency Operations Center and Technical Support Center are via the page phone system (Category 2 ).
b.
There is no storage or accountability for sound powered phones in the control room (Category 1 ).
c.
Incoming telephone calls to the plant are transferred to the control room at night (Category 1 ).
d.
No procedures for communication control during an emergency (Category 1 ).
e.
There is no direct comrunication from the main control room area to the HVAC panel area (Category 1 ).
e 4 7.0 ANNUNCIATORS a.
Annunciators lack prioritization (Category 1 ).
~
b.
Audible alarms are not directipnal except for Radiation. Monitoring System and the HVAC (Cagegory 2).
c.
Several annunciator windows are not located on the same panel with their associated controls (Category 3 ).
~
d.
Up to 3 different styles of print and character size are used on annunciator windows which varies the readability envelope (Category 2 ).
e.
Fire and Security alarms systems could not be reviewed. This system was not turned over to operations at the time of our audit.
(Category 1 ).
f.
Numerous annunciator windows are hanging open and others were missing tiles and bulbs (Category 1 ).
8.0 PROCESS COMPUTER IBM Selectric Typewriter (printing speed) is the limiting component in the process a.
computer system for displaying or printing numerous multiple process alarms (Category 2).
Analog trending capability is limited to 4 parameters on 2 process computer b.
controlled - 2 pen strip chart recorders (Category 2).
There is no CRT graphic display capability (Category 2).
c.
d.
No color coding used on CR.T. displays (Category 2 ).
Not all overhead alarm parameters are monitored by the process computer e.
+
(Category 2 ).
f.
Operators have received only a 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> training course on.the use of the Summer process computer at the system level.
(Category 1) g.
Data point addresses are not cross indexed. Present isdex requires excessive operator search time to locate individual addresses (Category 1 ).
h.
No labels or other means on panel to identify what four para. meters are being monitored by the two 2 channel computer controlled strip chart recorders (Category 1 ).
i The incore thermocouples on the process computer are limited to 700 F (Category 1 ).
9.0 OTHER OBSERVATIONS There are 3 channels of vertical indicators used to display RCS flow.
a.
The arrangement of ?.hese meters and the flow indication is not logical and is confusing to coerators reading the display (Categott 2 ).
b.
There is operator confusion concerning the manual initiation of Tafety Injection (SI) (Category 1 ).
Reactor Building (RB) temperature display on vertical panel is wide range.
c.
To detect incremental change in RB Temperature when RB fans are energized requires an operator to go to the HVAC panel where a narrow range tempera-l ture indication is available (Category 2 ).
o 4 -
d.
RB cocling units are cooled by industrial coalers.
Controls for cooler pumps are on CP 606 (sloping panel) while the industrial cooler teaperature alarm windows are'on CP 610 (7-8 feet away) (Category 2 ).
Emergency procedures require operators to rely on pattern recogaltion to e.
identify Train A and Train B initiation on 6 ESF status monitor light panels.
Patterns are confusing and difficult to recognize on display panels.
Patterns are shown and described in Attachment 1 to E0P-1, and these patterns were even difficult to interpret in the E0? (Category 1 ).
f.
There is construction in the area of the Refote Shutdown Panel, therefore, ve did not complete our review of the panels, lighting, and communication syste-(Category 1 until reviewed).
g.
There is a temporary ventilation fan mounted in the back of the main control board.
(Category 1).
h.
The plant page system in the control room is too loud, approximately 76 dbA, also the CR has reverberation (Category 2).
i.
F.mergency DC lighting - Illumination levels were between 2 & 11 footcandles on control and display surfaces (Category 2).
j.
Emergency AC lighting levels were not measured or evaluated. Actuation of this system was not accomolished. The pplicant should provide us a copy of its survey results for our review and evaluation prior to January 15, 1981.
(Category 1)
o
- 17 :
k.
No system is provided to test all the status an'd positilon indication lamps (Catagary 1).
'~
1.
The pressure power operated relief valve controls and associated isolation valve controls (445A, B, C) are inconsistently labeled.
(Category 1) m.
The strip chart recorder for rod position and insertion is scaled fron 0 to 1007,, which equals 240 steps. This is not consistent with the position indication on the main control board for full out pcsition (100".) which is equal to 228 steps (Category 1).
n.
There is no CRT trending capability or graphic display capability (Category 3).
10.0 SUSC00 LING MONITORS A Westinghouse systes is expected onsite by 10/30/80, installed by 12/1/Ls and operational by 1/1/81.
11.0 REACTOR VESSEL LEVEL INDICATION A Westinghouse system is expected on site by 11/30/80, installed and operational by 1/1/81.
T mL w
2
18 -
i 12.0 INCOPE THERMOCOUPLE SYSTEtt DISPLAYS 4
A.
Limited to 700*F.
B.
All 51 themocouples are monitore,d by process computer (PC).
Thermocouple readings can be displayed on either or botN' CRTS for a C.
maximum of 30 at a time.
D.
PC can provide long and short term themocouple maps and data on the trend typewriter in the CR and on line printers.
4 f
a n
g o
6 4
s,--
p
-- -+ -,,-
e-
,mm--
p
,