IR 05000282/1990004

Insp Repts 50-282/90-04 & 50-306/90-04 on 900228-0409.No Violations Noted.Major Areas Inspected:Plant Operational Safety,Maint,Surveillance,Radiological Protection & Industrial Safety

ML20042G191
Person / Time
Site:	Prairie Island
Issue date:	04/20/1990
From:	Burgess B NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20042G187	List: IR 05000282/1990004 ML20042G186 ML20042G190
References
50-282-90-04, 50-282-90-4, 50-306-90-04, 50-306-90-4, NUDOCS 9005110203
Download: ML20042G191 (17)

Text

ggypge ---m-

. +

m

-

m

'

,

,

,

y pg/

,7

_

,

>

l + y, :A, :

1. .

y;,r

x

<

-

'

%

>

..,.

>

,

Qjjat/fl.

• Q A O

'

,.

Di

'

'

'I Ct

U.S. NVCLEAR-REGULATORY COMMISSION'

'-

-

4F. Q'

i

~

'

>

..~

r < -

-REGION III

-

g

,

[

b

' Reports No; 50-282/90004(DRP)'; 50-306/90004(DRP)

-[

'

r

'

p

, Docket Nos..50-282; 50-3061 License Nos. DPR-42;- DPR-60 d

..

..

.;

'

.

..

q

Licensee:. Northern States Power Company

o

.

414 Nicollet Mall.

'

!

L*

[

Minneapolis', MN-55401

{m

Facility.Name

. Prairie Island. Nuclear Generating Plant'

.

.

t j

L'<

LInspection Ati LPrairie Island. Site, Red Wing,'MN

'

x

'Insp'ection Conducted': -February 28 through April 9,:1990

,

,

.

,

' Inspectors:

P. Li'Hartmann

-

.;

'

o T. J. 0'Connor l

'

-v w

"i

' Mdd/f7M i

"I i Approved By:

B. L. : Bu'rgess, Ch f Reactor Projects lSection-2A Date

,

n

Inspection Summary.

L

.

'

. Areas Inspected: : Routine = unannounced inspection by resident inspectors of-

'

[

< Inspection-onFebruary'28'throughApri'.9,1990(Reports'-No.-50-282/90004(DRP){

-

50-306/90004(DRP)).

,

.

,

'

~

'

plan.t operational safety, maintenance, surveillance, radiological protection -

I

.

and. industrial safety.

l

?

.Results:. Unit l:has~ operated continuously-ate 100% during this.-inspection i

period-except feedwater perturbation testing on March 20, 1990, and-load

'following on March 31,.'1990.

Unit I has reached 45' days of.continu'ous j

,

. operation (at the end of;the report period.

,

Unit 2's operati_ng performance was marred.by reactor trips;on March 8,

March l9, and March 16,-1990.

These reactor trips havelbeen; attributed to

.

Lequipment1. failure and personnel error and are explained.in ' detail Lin

.

. paragraph. 4? of thisi. report.

Unit-2 was taken off line on April 6,1990 'to i

$,

trepair: leaking tubes: in the 21A Feedwater Heater and' clean condenser tubes.

}

Unit l2 was returned to power operation late on April 8',

1990.

,

,

f

,,

..] :

,-

..;;

s

..,

'i <

9005110203 900420

.

'

I PDR ADOCK 05000282 V7

.o.

PDC

.

.

-

'

$;p a

.;[ '

j

'

s

>

_.

W W 9 MM+P W_&f6M@@bW% ' sL C.

.

,

'" ' ".

'

'

' '

! W %.

'=

,

,.

,. *

m

-

t

<-

'

w m

%.c i - s e

m' ;

5.W.

g_).% > m.

4 4.4 Y-i s i

\\

,,

V:

"

L

.

Nf. f $l &! i'A, p.ol?.. f Y ! h'

PyMm.),,u qld a,

,

!

'

--

~.'

s

-

'

'

'

o t-ms in "

• 1 + i 6,h a$

,

+

y ua

~_a w*"Tvw s ~w.m

-

- _,

r

-

'

+-

,

fh W *,.DM%tfd*6 !E

'

'

,

<

'

'

<<

.

,

,g i&N

tf* Y ;- '

l

.

'i

.

'

"

>

hh@b'lCL $30filheQf areasMnspectedi: onefviolationiof NRC Lrequir

!6

.

1. p)yftsf 0uring lthe,lJan'uary/Februaryy refueling outage,llopeningsiin'the_3KC*l! Buil

'

4,'

p Auxt:liaryl J

m

.

RA 7 1 required by: Technical 1 Specification 3,6;E.21 : The root!cause wasipersonnell.

-

H%mW;a;; o

--e

.

,

Terrob'

'

W s :nk y ydidM._ '-

i

>

<

p;w f.p',_; y a m.

--

.A Q pai, _ ts su y

t';,.

-.;

w ~ g;;-

&y lY*

,

v

' & t l., ', a'

u : ;; W r i

p.m >. sn.

r

~

>

hp+ r; 's-V !

(

1. . h; 'm

.Nl

^..

v-

,

f -

,

r..

T $ = ::

d_=

,_

.z._

1 -

,--t,'

.,

4Ag

'

'

-

,

un

>

,

a b' D. l ;

f

'. ' '

'

,.

r-wa -

07- -

,

,

hs.

.t

. y; :

s

?

g -

(

,

i o

z..

• <

b.

f,.)

[

.

,

w-

_..)

,

Y

-

e,

'- ?

E n

.r.

, ~.

t

'?

y

';' { ;

. S:

'.

.

'.

{'

k F

t

.

v c

.m,

,

g 3_,>.

,

(,'f t.

i

.

gg

'

,

d-

.,, ' ' -

e

'_

g

.

,

s (l.

'

_'d

'

.

.

.

,

b

>

c-

).

f:hQ

}.

• _.s f,'

x.'J

i

.-

k t

5 :l g

f

.

Qhl;Q

)

7.

'7'

<

r

\\

L

,

g YN(. -

).$b

,

'h.',. t'..'.I

<

I

.cg'.

'

('.'.O

- t. /

'

. ->

g-1G6

[d '.b '

._

.-

,

,

h > r...,. lN y y y; -

,

,,

a

'.;..,6 l-dn.

E.. J.

q, ig :

,

-

+ -: -

,

,

d, '-

I

.Qr t'.,

^i

..

t'DD e

-

Aih,

,

i

.

. _-

2.

I

.1 wo

>

m -

,

,.

-

.kl

-

--

,

yr.

..i*. :. r :

-

-

.'#

,.

x k

r

, #,-j N-* ',, t

' h5 y a

'

g

.

gk;- l(*,^' {h E

'. ' t i

vl:: _ ' ~_, ;& s

%

_

i

'"

m

--

,

E,, f

,

.

.,

,

.j.

DETAILS 1.

_ Persons Contacted E. Watzl, Plant Manager

1. D. Mendele, General Superintendent, Engineering and Radiation Protection

1. M. Sellman, ' General Superintendent, Operations l

G. Lenertz, General Superintendent, Maintenance A. Smith, General Superintendent, Planning and Services R. Lindsey, Assistant to the Plant Manager i

1. D. Schuelke, Superintendent, Radiation Protection G. Miller, Superintendent, Operations Engineering

1. K. Beadell, Superintendent, Technical Engineering S. Schaefer, Superintendent, Technical Engineering s

M. Klee, Superintendent, Quality Engineering

[

R. Conklin, Supervisor, Security and Services j

G. Eckholt, Nuclear Support Services i

1. J. Leveille, Nuclear Support Services t

A. Hunstad, Staff Engineer

1. P. Wildenborg, Health Physicist t

Denotes'those present at the exit interview of April 12, 1990.

2.

Licensee Action on Previous Inspection Findings (92701)

{

l a.

(Closed) Unresolved Item 282/90002-02(DRP): Auxiliary Building l

Special Ventilation Zone Integrity (ABSVZ)

.l Inspection Report Nos. 282/90002 and 306/90002 identified an j

inspector concern over an opening created by removal of a blank i

flange, and subsequent-installation of the eddy current cable

!

connection flange. This modified flange is placed on the containment. vessel pressurization line to allow passage of eddy current cabling from.the auxiliary building to the containment.

l

,.

Review of the documentation supplied by the licensee confirms that j

while the eddy current cable connection flange was secured in place,.

,

containment integrity was not affected and that an opening'in the j

ABSVZ did not exist. The installation and removal of the eddy

. current cable connection flange creates an opening in the ABSVZ. The-

opening created in the ABSVZ during the installation or removal of i

the eddy current cable connection flange was not under administrative

-

control.

'

Technical Specification (TS) (Containment Systems) 3.6.E.2 states that openings in the ABSVZ are permitted provided they are under

direct administrative control and can be reduced to less than ten square feet within six minutes following an accident. Operations Manual Procedure D54, Control of Openings in the ABSVZ Boundary, I

y

.

.

.'..

.,.

,

,

~

Revision 4, implements TS 3.6.E.2.

Step 4.1 states that'a log (PINGP 751) shall be kept in the Control Room specifying the size

.

and. location of all openings in the ABSVZ. The log shall-include the time.and date that openings are made and when they are closed.

Contrary to the requirements of TS 3.6.E.2 and Operations Manual Procedure D54, between the dates of-January 17, 1990 and February 15, 1990, the installation and removal of the eddy current cable connection flange created an opening in the ABSVZ which was not administrative 1y controlled in that the opening was not logged to indicate the size, location, time, and date that the opening was

,

made or closed.

This is identified as Violation 282/90004-01(DRP),

and administrative 1y closes Unresolved Item 282/90002-02(DRP).

It should be noted that Violation 50-306/89017-01(DRP) was issued on June 20, 1989, which also documented the licensee's failure to administrative 1y control openings in the ABSVZ boundary during outage related activities.

Although both violations may be attributed to a lack of sensitivity to openings in the ABSVZ by operations personnel, an additional cause for this violation may be attributed to the work package which directed the installation of the eddy current cable connection flange.

3.

Operational Safety Verification (71707, 93702, 82301)

a.

Routine Inspection The inspector observed control room operations, reviewed applicable logs, conducted discussions with control room operators and observed shift turnovers. The inspector verified operability of selected'

emergency systems, reviewed equipment control records, and verified the proper return to service of affected components, conducted tours of the auxiliary building, turbine building and external areas of the plant to observe plant equipment conditions, including potential

~

-fire hazards, and to verify that maintenance work requests had been initiated for the equipment in need of maintenance, b.

Lockdown Search of New Administration Building On April 4, 1990, the inspectors monitored the activities of the security. guards as they conducted the lockdown. search of the new administration building. This search was conducted prior to its introduction into the protected area and looked for prohibited materials such as firearms, explosives and drugs. The security guards were assisted by local law enforcement officials and their dogs, one trained for explosives, and the other trained for drugs.

No prohibited materials were discovered, c.

Emergency Plan Drill On March 28, 1990, the resident inspector observed an emergency plan drill. The drill scenario and plant response were observed from

1

&

,

-

,

,

,

>

.

..

.

.

.

.

~

simulator's control room located in the training center. The

/.

inspector was satisfied with all activities conducted during the course of the drill scenario.

The inspector also attended the critique of the licensed operators conducted by control room-

.

-controllers.

The inspector was satisfied with the objectivity of the critique and the feedback received by the General Superintendent'

of Operations and the Superintendent of Radiation Protection.

>

4.

Review of Reactor Trips (93702)

During the inspection period three reactor trips occurred on Unit 2.-

The cause/ sequence of events were not immediately related to the December 21 and 26 trips on Unit 2.

Each trip is discussed in detail below.

h a.

March 8, 1990 Turbine Generator Lockout Trip On March 8, 1990, while at 100% power, the reactor tripped. The cause was a _ false main generator lockout which causes an immediate reactor trip.

All systems functioned as designed (except as noted below) for a secondary side initiated reactor trip, where a 30 second time delay for opening of generator output breaker is not available.

This 30 second time delay is designed to supplement / remove the decay heat following a reactor trip.

,

With the additional heat load to dissipate, several feedwater heaters and moisture separator reheater (MSR) reliefs lif ted, releasing steam outside the turbine building until reseating occurred. The relief valve on the suction of the 21 Feedwater Pump lifted and did not reseat, requiring system isolation and subsequent repair. Water hammer occurred in the 14 inch 24A Feedwater Heater to 21 Heater Drain Tank line during the cooldown and depressurization of the

.

.

secondary side.

Back leakage through a check valve was the suspected cause of the water hammer.

Check valves in-the piping leading to the 24 and 25 Feedwater Heaters and to the Heater Drain Tank were examined and tested.for operability. All check valves operated satisfactorily.

A gasket leak on the heater drain tank pump discharge valve 212HD1-1 occurred and was fermented..The 1icensee inspected the secondary systems affected and fot.nd no major damage.

~

'

The cause of the turbine trip at 100% power was a generator lockout signal generated at the generator bus duct cooling control panel.

,

The lockout signal was generated when an operations. instructor depressed the panel test push button.

Depressing the test push

button energizes the K3 relay (the attached schematic diagram).

,

,

When the K3 relay is energized, a "b" (normally closed) contact (K3-23) in the " generator lockout power supply" should open, thereby preventing the generator lockout relay (23X-2) from being energized.

Additionally, during actuation of the test circuitry an "a" (normally

'

open) contact on K3 closes which provides power to the K2-1 and K2-2 relays. When relays K2-1 and K2-2 are energized, "a" contacts K2-1 (23) and K2-2 (23) close, and would energize relays 23X-2 and 23X-1,

,

W.

.

l1.

e

=

.

.

'

'

.

.

..

hg

'

'

g.

J if contact K3-23 were closed.

Energization of relay 23X-2 would

>

cause a generator lockout (setpoint 85 degrees C sensed in the bus i

duct).

Er.ergiz& tion of relay 23X-1 would cause the autostart of a second bus duct cooling fan to occur (setpoint 60 degrees C sensed

.

L in the bus duct).

When the test button was depressed all functions occurred as designed

!

ll except the K2-23 "b" contact remained closed.

Once K2-2 was energized

'

'L (by K3),.the K2-2 (23) contact closed and energized the 23X-2 relay,

causing the generator lockout.

'

'

t The root cause of this equipment failure appears to be age. The if equipment was installed during initial plant construction.

In response, the licensee has electrically disarmed the test push

!

.

button on Unit 2.

The Unit I test button is SECURE-CARD tagged l,

again.tt usage. The test push button performs no valuable function

,

other than illumination of test lights which indicate the K2-1 and K2-2 relays are energized.

The setpoints for the 23X-1 (second) bus

duct cooler fan autostart ar.d the 23X-2 (generator lockout) which

are based on bus duct temperature, receive calibration during every

.

,

refueling outage.

'

The plant started up at 1027 hours0.0119 days <br />0.285 hours <br />0.0017 weeks <br />3.907735e-4 months <br /> on March 8, 1990, following an Operations Committee review which the backup senior resident inspector attended. The reactor was critical at 2208 hours0.0256 days <br />0.613 hours <br />0.00365 weeks <br />8.40144e-4 months <br />.

Heatup

and power escalation ensued.

The inspectors will followup licensee review and corrective actions by the Licensee Event Report (LER)

306/90001-LL.

b.

March 9. 1990 Reactor Protection Logic NBFD Relay Failure and Subsequent 7 0ss of the "B" Channel of the Reactor Protection i

System Reactor Trip The main generator turbine was test tripped at 0128 hours0.00148 days <br />0.0356 hours <br />2.116402e-4 weeks <br />4.8704e-5 months <br /> on March 9. 1990.

Following this test, the reactor was at about 6%

reactor power when preparations were being made to place the turbine online for full power operation. When the turbine'was placed on line a reactor trip occurred.

All systems functioned as designed; however, some anomalies occurred which are discussed further below. The cause~cf the trip was loss of DC power to'a portion of the "B" channel of the reactor

-

protection system (RPS) by a-fuse (F3) opening.

This loss of power removed power to e oortion of the "B" reactor prote tion system, and

>

resulted in the "B" reactor trip breaker opening. Tne "A" reactor trip breaker tripped shortly thereafter on a high r te flux in the power range nuclear instrumentation which sensed D rods being inserted.

The F3 fuse opening in the "B" train of.RPS was caused by the failure of the 2SV2XB NBFD relay, which failed as a short circuit.

.

L.u.Aa

p g

.

.

.

'

' '

,

.

.

..

This failure resulted in a high current drawn through the relay and m

opening the F3 fuse.

Loss of F0 de-energized the 12 B channel.

reactor trip relays, and associated permissive circuits and other activation functions. With the de-energization of the reactor trip relays, the "B" channel of RPS logic caused the undervoltage (uv)

i coil to de-energize and the shunt coil to energize for the "B"

reactor trip breaker.

Either the uv or shunt coil function will L

~ cause the breaker to open.

Three "first out" annunciators illuminated solid on the first out control _ room annunciator panel.

This panel is designed to indicate g

any activated reactor trip function by a solid annunciator window illumination, and the "first out" trip signal indicated by a

.

flashing annunciator window illumination.

The flashing "first out"

!

trip annunciator aids the operator in determining the cause of the

!:

reactor trip, which is later verified with the sequence of event E.

computer printout.

On the March 9 trip, three first out annunciators illuminated in a flashing-mode, indicating they were e

the "first out."

These alarms were:

1 loop low flow or RCP (Reactor Coolant Pump)

breaker open reactor trip; 2 loop low flow or RCP breaker open reactor trip; and RCP buses undervoltage..Following the trip, operators immediately verified power was in fact available to the RCPs and RCP operation was not impaired or suspended.. The reason these three annunciators indicated flashing was due solely-to the loss of DC power from fuse F3 opening. Three other annunciators on this panel indicated solid. These indicators were the:

high flux rate reactor trip; the turbine trip / reactor trip; and the source range high flux rate trip.

The source range detectors were energized when fuse F3 opened. This occurrence is not desirable due to the high flux present the short time prior to rod insertion. The design of the RPS system is conservative-in that the loss of both source range detector relays (SRB 1XB, SRB 2XB) caused the source range detector to energize when fuse FS opened. With this source range in operation at relatively high power, the source range logic tripped the "A" RPS train logic.

This reactor trip signal, however, did not cause the "A" Rx trip breaker to open since this trip was blocked.

The design auto-unblock does not occur until 10 EE-10 amperes on both intermediate range nuclear instrunients. The source ranger detectors operated with normal indication.

The turbine trip / reactor trip resulted from tne initiation of a partly operational logic train. When the "B" Rx trip breaker opened, the P4 permissive was met to pass a turbine trip signal, which occurred as designed.

Permissive P-9 will cause a reactor trip signal to be processed if reactor power is greater than 10%. During this event reactor power was less than 10%.

P-9, however, was i

$

I

'!

_

. - -

. _ _ _

_

_

3.

.

"

• *

,

,

.

.

q

satisfied when fuse F-3 opened causing both P-9 relays to de-energize (p9-1, PI-2) on the B train of RPS. With P-9 satisfied, a reactor

,

L trip s-igrial was processed.

F The high negative flux rate reactor trip was a valid signal sensed

-

L by the A train of RPS. The rods fell into the core when the "B" Rx

-

L trip breaker opened. The large negative change in flux caused this i

h trip to occur in the A channel of RPS and opened the "A" Rx trip

,

breaker.

>

?

Additionally, the turbine driven auxiliary feedwater pump

!

!-

autostarted.

This was due to the 2TD-AFD relay being de-energized

.!

by fuse F3 opening. The 22 TDAFW pump typically autostarts on a

reactor trip due to the low SG 1evels present because of the

.

" shrink" phenomenon. The operators stopped 22 AFW pump at 0236 by j

'

.

placing the control switch in manual, effectively removing the constant inutostart signal present.

i p'

,

Replacement of the failed 2SV2XB NBFD relay and fuse F3 were completed on March 10, 1990 at 1405 hours0.0163 days <br />0.39 hours <br />0.00232 weeks <br />5.346025e-4 months <br />.

Reactor prottetion logic

,

-

test was then commenced.

During tnis testing, visual inspection of

.

the protection relays identified two relays (P7-1XA, P7-2XB) which

.I wtre de-energized when they were required to be energized. These j

relays were P-7 relays, one in-each train of RPS.

r The failure of thne two relays had no immediate effect on the RPS

'

logic. This is due to the parallel design of the P-7 permissive

,

relays.

For P-7 to actuate, both relays in a train must y

de-energize.

The failure of one relay per train would not cause i

y nor prevent the intended function. With a relay failed in each b

train, the P-7 permissive was relying on one relay to bypass the

^

reactor. trip functions associated with P-7.

The P-7 permissive blocks the following reactor trips with less than 10% power range indication:

low pressurizer pressure, high pressurizer -level. -loss

-

of 1 RCP (low flow or breaker open) and;undervoltage on RCP bus.

. These relays were replaced and RPS logic was retested satisfactorily.

The operations committee reviewed the cause of the reactor trip, planned repair work and retesting of the RPS system. An additional DC meeting was conducted telephonically to review restart after the replacement and testing of the failed P-7 relays.

The plant was restarted at 0133 hours0.00154 days <br />0.0369 hours <br />2.199074e-4 weeks <br />5.06065e-5 months <br /> on March 10, 1990 with criticality achieved at 0215 hours0.00249 days <br />0.0597 hours <br />3.554894e-4 weeks <br />8.18075e-5 months <br />.

Plant heatup and power escalation occurred without incident. The inspectors will follow licensee review and corrective actions documenttd in LER 306/90002-LL.

c.

Unit 2 Rx trip _due to loss of Rod Control Reference Voltage

,

On March 16, 1990, the Unit 2. reactor tripped from 100*4 power.

The cause of the tri'p was loss of Rod Control Reference Voltage, which ultimatoly allowed two shutdown b-k rods (E-03 and I-11) to fall

,

.

x u

as..

...

-

-

-

--

...........

.

.

.

.

,

,

.

into the core. The insertion of the two shutdown bank rods caused a high flux rate (negative) reactor trip. All systems functioned as designed. The senior resident inspector was in the control room at

,

the time of the trip and observed the response to the event.

In *

pense to rod control electronic function problems which result ed in previous reactor trips (December 21, 1989, Deceuer 27, 1989, see report 282/89032; 306/89032) electronic moni.oring on the E-03 and I 11 rods was initiated and maintained.

Due o electronic noise on one channel of a multichannel recorder, an I&' technician with approval of the shift supervisor initiated a troub'eshooting activity to identify or eliminate the noise.

The 3&C technician discussed the possiteility that an urgent failure alarm may occur. This warning was based on the changeout of instrument leads to the multichannel recorder, one point being

-

Reference Voltage (V ref).

V ref is compared to voltage sensed at sampling resistors for each rod. When V ref went to zero, the demand for stationary coil voltage is zero. With the demand to the moveable coils also being zero (no demand for rod notion), a power cabinet urgent failure alarm occurs.

To measure V ref with the multichannel recorder, the unit is connected electrically as parallel circuit.

Removing the multichonnel recorder could cause a temporary porturbation in V ref and could cause an urtjent f ailure.larm. The 1&C technician discussed thi-s,70ssibility with the t.ead Reactor Operator (RO) prior to changing out the instrument leads.

The I&C technician procteded to change test leads and in an effort to gain information regarding the noise, connected an oscilloscope to the affected test channel.

When the oscilloscope was connected, the otrilloscope drew current away from the V ref circuit. The effect was to cause V ref to drop to near zero volts, causing voltage to stationary rod coils to drop to zero, Bec,ouse the stationary coil voltage and moveable coil voltage was zero, an urgent f ailure occurred '

for power cabinet 1AC. When the urgent failure occurred a hold current is applied to the moveable and stationary coils to prevent a rod drop from this logic failure.

In this event, rods E-03 and 1-11 began movement with the loss of V ref. The hold current supplied by the urgent failure stopped rod movement and prevented the two rods from falling into the core.

Total *od motion for the affected rods was about ten steps. The hold current is maintained until the urgent failure is reset.

The RO saw the urgent failure, and in accordance with system training, attempted to reset the urgent failure since.he understood the cause of the alarm to be temporary and the alarm condition had cleared. When the reset button was depressed, the hold current was removed which allowed rods E-03 and 1-11 to fall into the core.

Reset removes voltage to moveable coils only. During the event however, the normal voltage that would remain applied to the stationary coils was not available due to the oscilloscope connection.

The insertion of two full length rods caused a high flux rate

' - * - ' - - -

--

,:_____________n__._--_.-

- -. - - - - - - - -

u- - - - - -.-

p

a'

p

,

l

-

,

,

<,

k

(negative) reactor trip.

The root cause has several contributing

.

f factors.

The most substantial being use of an oscilloscope which

caused V ref to drop to near zero.

,

4:m t

b The I&C technician had decided to use a Hewlett Packard (HP) Model

-

(54502A) oscilloscope which was leased by NSP for use and evaluation.

!

Calibration documentation was supplied by the leasor to NSP. When

!

the I&C technician brought the HP oscilloscope into the rod control

!

room, he placed the inst.'ument adjacent to the Tektronix oscillescope

<

'

f.

Model (564B) which had been left in the rod control room for

"'

monitoring activities.

.In order to utilize the HP oscilloscope, the I&C technician needed

,

!;

'to attach an instrument probe to an input, jack on the front panel of

%:

the oscilloscope. The I&C technician removed the probe from the (

Tektronix oscilloscope and connected it to the HP oscilloscope, and

,

"

went on tc connect the oscilloscope to the multichannel recorder, s

By using the Tektronix probe, a low input impedance circuit (the HP L

oscilloscope) was connected in parallel to rod control. The result was the HP oscilloscope drawing most of the current away from rod control and lowering V ref to near zero.

The licensee verified by

'

test this action as the cause of the event. The Tektronix

-

+

oscilloscope had internal circuitry which provided 1 meg ohm (1,000,000 ohms) resistance which would have prevented the

'

oscilloscope from loading the rod control system. The HP oscillo w ope has a 50 ohm input impedance. The HP in a pouch attached to the oscilloscope) provided (probe (which was externally) an

. impedance of 1 meg ohm.

The labeling next to the HP probe input jack stated the input impedance as "50 or'I meg ohm."

The technician was not aware that the HP oscilloscope would have only a 50 ohm input impedance with a zero resistance probe.

The second contributing root cause was resetting the urgent failure

. alarms when rod control system conditions did not warrant reset.

This was from multiple causes, inadequate communication ar.J past licensed. operator training.

The inadequate communication occurred when the I&C technician did not accurately ammunicate to the RO the extent of his activities with rod control. The RO believed that the s

',

activity was extremely short in duration, i.e. only the time to b

change out the test leads on the multichart recorder. Although the I&C technician was only accessing test points in rod control with the oscilloscot.e, his activity lasted for a longer duration than what was communicated to the RO. - This information was the basis of the R0 decision to reset the urgent failure.

The R0 action was also based on operator training to reset the urgent failure when the cause was known and had cleared. When the urgent

'

failure annunciator alarmed, the RO believed the cause to be temporary, based on the information supplied to him by the I&C technician.

Following a short waiting period, he attempted to reset I

i ai

o

-

r

,

.

..

the urgent failure. -When the reset button was depressed, the hold voltage was removed from the moveable coils and the two rods fell

into tho.; ore.

'

The licensee has initiated multiple corrective actions. A temporary

!

change (TM-90-36) has immediately issued to clarify operator actions in regard to resetting urgent failure alarms.

Now operations must notify the I&C department prior to resetting the urgent failure alarm (except during rod control exercise tests). A cFange to the section work instruction (SWI) test instrument calibration control is planned to further control usage of test equipment.

In the past, this requirement was informal.

The licensee also intends to clarify

'

work control requirements of instrument monitoring.

This is to

'

ensure that troubleshocting of. test equipment connected to test points is adequately controlled.

The licensee also initiated a procedure change to address an

instrument suitability review specifically for adequate input impedance.

This action was initiated from the March 16, 1990, reactor trip and a separate instrument loading event which occurred on March 19, 1990.

The March 19th event occurred during Unit I troubleshooting of rod speed / response time to a reactor coolant

' temperature error (T error). When a newly acquired strip chart recorder was connected to an isolated channel of turbine impulse steam pressure, the rods began to automatically step into the core.

,

The I&C technician promptly removed the recorder fren the circuit and rod motion stopped.

The recorder has two alternative input

,

impedances 1 or 5 meg ohms.

The 1 meg ohm input impedance was first i

used.

Followf rg further review af d analyses the recorder was

'

connected with 5 meg ohm input resistance.

There was no effect on

.

T error.

The review disclosed the particular circuit was sensitive to instrument loading. The licensee believes the instrument review for input impedance applicability will address this event and the potential for other similar occurrences.

The licensee's efforts in this-area are important since the technology of I&C mecsuring and i

test equipment continues to improve, and as original equipment

-

ages, more measuring and test equipment will be purchased and. utilized by the licensee. The inspectors will follow the licensee's review and corrective actions (LER 306/90003-LL).

5.

MaintenanceObservation(71707,37700,62703)

<

Routine, preventive, and corrective mair.tenance activities were observed to ascertain that they were conducted in accordance with approved precedures, regulatory guides, industry coces or standards, and in conformance with Technical Specifications.

The following items were

'

considered during this review:

adherence to limiting conditions for operation while components or systems were removed fMn service, approvals were obtained prior to initiating the work, activities were accomplished using approved procedures and were inspected as applicable,

,

t

"

,

,

Y e

..

.

'

.

...

,

.

.-

functional testing and/or calibrations were performed prior to returning components or systems to service, quality control records were K

maintained, activities vere accomplished by qualified personnel, L'

radiological controls were implemented, and fire prevention controls were E

implemented.

Portions of the following maintenance activities were observed during the inspection period:

Repair to Traveling Water Screens Replacement of Reactor Protection Logic Relay 2SV2XB Replacement of Reactor. Protection Logic Relays P7-2XA and P7-2XB Replacement of Reactor Protection Logic Relay IRT2-XB L

Replacement of the 22 Cooling Water Pump seals, bearings and the pump alignment Replacement of the il Charging Pump Desurger Trouble shooting of the 11 Fan Coil Unit Breaker Rebuild of a safety relief valve for the Chemical and Volume Control System Replacement of Gaskets on Valves Associated with the 122 Spent Fuci Pool Heat Exchanger On March 20 and 21,-the licensee removed Emergency Diesel Generators (EDG) 1 and 2, respectively,.from service to perform a check of the upper crankshaft thrust bearing. Additional work on EDG 2 included the replacement of two fuel injectors,.the repacking of the air start valves, I

and the conversion of a low temperature lube oil heater shut off switch to a high temperature shut off switch. The inspector monitored various aspects of the maintenance activities including the restoration of EDG l

output breakers and. the realignment of various valves.

The fuel injectors were replaced to correct i condition which allowed' fuel to bypass the injector due to oad seals.

During engine testing it was

'

determined that additional injectors need to be replaced.

The inspector witnessed the performance of SP 1093.4, D1 Diesel Generator Manual and 4 KV Voltage Rejection-Restoration Scheme Test, BRKR 15-2 Rack-In Verification and Bus 26 Load Test, Revision 3, and SP 2093.4, D2 Diesel Generator Manual and 4 KV Voltage Rejection-Restoration Scheme Test, BRKR 16-7 Rack-In Verification and Bus 25 Load Test, Revision 3.

These procedures verified the operability of various relays in the plant load rejection and restoration scheme, verified the EDGs ability to carry approximately 2500KW and also satisfied independent verification that both output breakers associated with each EDG were properly racked in.

.No violations or deviations were identified.

'6.

Surveillance (61726. 71707)

The inspector witnessed portions of surveillance testing of safety-related systems and components.

The inspection included verifying that the tests were scheduled and performed within Technical Specification requirements,

u

-

m

-

-

'

.

.-,.

.

.

.

,

by observing that procedures were being(followed by qualified operators, that Lih:iting Conditions for Operation LCOs) were not violated, that

[.

System and equipment restoration was completed, and that test results were

acceptable to test and Technical Specification requirements.

SP 1116 Monthly Power Distribution Map SP 1295 Emergency Diesel Generator D2 Manual Start and Bus 16 Load Rejection and Restoration L_

SP 1015'

4KV Voltage & Frequency Test, Revision 12 i-SP 1089 Residual Heat Removal Pumps and Suction Valves From L

the Refueling Water Storage Tank, Revision 26.

SP 1102 11 Turbine-Driven Auxiliary Feedwater Pump Test.

Revision 39 SP 1158 Charging Pump Desurger Test, Revision 12 The inspector witnessed the performance of surveillance procedure SP 1588, Charging Pump Desurger_ Test, Revision 12 on April 3, 1990.

This procedure describes the steps necessary to check the desurger precharge and actions to take if the actual precharge pressure is less than required.. The desurgers are utilized in the discharge line from the positive displacement charging pumps. Desurgerb effectively eliminate water hammer common to positive displacement charging pumps. The charging pumps are considered operable with a failed desurger.

SP 1588 was performed on' charging pumps No. 11 and 12, both of which were determined to have desurgers with failed bladders.

Difficulties were encountered during_the conduct of this surveillance due to extra valves on the tubing used to direct liquid to floor. drains, lack of procedural-direction to address high pressure indications on the desurger, and the confusion with the opening and closing of a valve in one step. These difficulties were conveyed to the system engineer with the surveillance results.

SP 1588 is currently being-revised w'th the assistance of the operations staff. This revision should greatly assist in the performance of SP 1588. Additionally, the system engineer is pursuing greater reliability of the desurger materials hith the vendor.

It has been noted by the inspector that those surveillance, maintenance

- and. operations procedures revisions written by a system engineer with significant input by those responsible for their' usage are greatly improved over previous revisions, Although this process is more time-consuming, the quality of the end product appears to justity the effort.

No violations or deviations were identified.

7.

ESF System Walkdown and System Focus (71710, 61626)

The inspector performed a walkdown of the Unit 1 Auxiliary Feedwater System and observations included confirmation of selected portions of the licensee's procedures, checklists, verification of correct valve and power supply breaker positions to insure that plant equipment and instr. mentation are properly aligr.cd, and local system indication to

,

I

b

-_ -_

- _ - _ _

.

. -

I

'

'

'

.

,

insure proper operation within prescribed limits. The inspector utilized

.

C28-2, System Prestart Checklist, Auxiliary Feedwater System Unit 1, i

Revision 22 to conduct this walkdown.

Several minor discrepancies were noted between the procedure and equipment tag nomenclature. The licensee has initiated action to correct these identified discrepancies.

The inspector witnessed the performance of Surveillance Procedure SP I

'1102, 11 Turbine-Driven Auxiliary Feedwater Pump Test, Revision 39, on March 28, 1990.

SP 1102 fulfills monthly testing requirements of Technical Specification 4.8.A.

The surveillance was successfully

'

completed. The Plant Equipment Operator noted minor speed oscillations when the turbine was initially started with a duration of approximately 1 minute.

These observ4tions were documented on the cover sheet of SP 1102, Also recorded on the cover sheet was a notation reiterating pump

,.

seal leakat as having been previously d m mented on an outstanding work request.

A review of these comments by the system engineer was conducted and the determination made that operability was not affected. -The next performance of SP 1102 will be conducted in the presence of the maintenance staff and the system engineer for further evaluation. The inspector will continue to monitor surveillances and corrective actions associated with the auxiliary feedwater pumps.

Activities associated t

with the auxiliary feedwater pumps have been previously documented in para 0raph 4.d of Inspection Report Nos. 282/89031and306/89031(DRP).

.

8.

Licensee Event Report Followup (92700)

a.-

(0 pen) Licensee Event Report (282/90003-LL):

,

LER 90003 was issued to document the auto-start of the No. 12 Diesel Cooling Water Pump (CWP) on low het. r pressure during a

,

surveillance at approximately 0350 hours0.00405 days <br />0.0972 hours <br />5.787037e-4 weeks <br />1.33175e-4 months <br /> on March 23, 1990. The event was communicated to the NRC operation duty officer as

,'

required.

The autostart has been attributed to a loss of prime on the No. 11 CWP and subsequent ~ pressure degradation on the cooling water header.

The loss of prime on No.11 CWP was attributed to excessive air in leakage at the pump seals.

Investigations'into the seal water supply and air eductor system identified no problems.

Improvements in the eductor system have been made with the.

replacement of all piping with stainless steel and the addition of a back-up rotary screw vacuum' pump.

The auto-start of the No. 12 diesel cooling water pump occurred during the performance of SP 1106b, 22 Diesel Cooling Water Pump

Test, Rev. 28. After completing the required run time and in preparation for stopping the 22 diesel cooling water pump, the

,

,

procedure requires the pump which was stopped earlier to be restarted.

The procedure provides a note to the operators which states, in-part, to be sure that the pump that was just started (21 CWP) has taken its share of the load by observing a header pressure increase

<

,,

w

>

y,,

.

i

• ,7

-

..

a-

,

and a temporary speed increase on the diesel.

The procedure further states that a failure to have the header pressure increase or diesel

~

speed increase could mean that the pump is not primed and directs

,,i the operators to C35, cooling water system, for re-priming the pump.

^

The operators determined that adequate indication of proper 21 CWP operation was observed and commenced the shutdown of the 22 diesel

!

,

CWP. While reducing speed on the 22 diesel CWP, the cooling water header pressure decreased to 75 PSIG at which point the 12 diesel CWP autostarted.

The operators then discovered that the 21 CWP had

'

!.

lost its prime.

SP 1106b does not direct the operators to verify

!

that the pump is primed as indicated by the local sight glass as is

-

specified by C35, Cooling Water System.

'

Corrective actions associated with this event include examination of

-

'L the 21 CWP air eductor, calibration of the pressure' switch which

!

starts the 121 CWP_and increasing this setpoint to 80 PSIG cooling water header pressure.

Previously the.autostart setpoint was 75

PSIG for the 121 CWP and 75 PSIG with a 15 second timer delay for n

the autostart of the diesel cwp's.

The increase to 80 PSIG for the 121 CWP autostart will help miniinize autostarts of the diesel CWP

when decreases in cooling water header pressure are experienced.

The inspectors will followup licensee review and corrective actions

by the Licensee Event Report (LER) 282/90003-LL.

It should be noted that on April 21, 1988, both the No. 12 and

1

22 Diesel Cooling Water Pump autostarted due to low cooling water header pressure.

Low pressure occurred because the 11 CWP had

become airbound and stopped pumping.

Part of the corrective action i

-

for this event included the issuance of a PI Operations Note which briefly explained what had happened and provided special

.!

instructions for operating the eductor system and. for checking to i

see that 11 and 21 were operating properly. While the impectors do E

.not consider the event of March 23, 1990 a repeat of the

April 21, 1988 event, the inspectors feel that the corrective action taken should have addressed CWP operation during the performance-surveillance procedures.

9.

Regional Meeting

,

On March 23, 1990, at 10:00 a.m. CST the licensee met with Region III i

management in the Regional Office in Glen E vn. IL.

The purpose of the=

!

'

meeting was to discuss lessons learned from the recent cerctor trips.

Mr. E.G. Greenman, Director, Division of Reactor Projects, led the

'

discussion with the licensee. The licensee discussed:

the root causes of the recent reactor trips, continued use of Westinghouse (BFD and NBFD)

DC relays, vulnerability of the plant to secondary (steam side)

survaillances, and actions to address electrical and instrument aging

issues.

In particular, the licensee discussed the plans to replace the

,

NBFD relays which are installed in the Unit 2 RPS system (Unit I has more reliable BFD relays) during the upcoming (Fall 1990) refueling outage.

'

.

.

,

,_

__

_

-

_ _ _ _ _ - _ _ _ - - _ _ _ - _.- -.

y

.

if - ',

-.

>

,

,

,

,

l

• -

an These plans are potentially impacted by prouact availability,

!

modification engineering requirements and tny change in the outage date.

,i The personnel in attendauce were as follows:

!

-

I

"

.

NSP Title

t

.C.

Larson Vice President Nuclear Generation E.-Watz1 Plant Manager - Prairie Island (PI)

'

F D. Mendele General Superintendent Engineering and Radiation

,

U Protection - Pl.

';

L F. Bendell-

. Superintendent Technical Engineering - PI L

R. Lindsey Assistant to the Plant Manager - PI

'

E NRC

!

o r

E. Greenman Director, Division of Reactor Projects,

!

Region III-R.

Cooper Chief, Engineering Branch, Division of Reactor

,

.

Safety, Region III L

T. Burdick Chief, Operator Licensing Section 2

',

D. Dilanni

' License Project Manager, NRR P Hartmann Senior. Resident Inspector - Prairie Island

E. Schweibinz-Project Engineer, Region III

'7 l-D. Butler Reactor Inspector, Region III

'

"

10.

Exit (30703)-

<

'

The inspectors met with the licensee representatives denoted =in

"

t paragraph 1 at the conclusion of the report period on April 12, 1989.

The-inspectors discussed the purpose and scope of the. inspection and the

.

findings. The inspectors also discussed the=likely information content

.

,

of the inspection report with-regard to documents or processes reviewed

- by-the inspector during the inspection;. The licensee did not identify

any documents or processes as proprietary.

l

_

!

i

!

Ie v.

b i

i

N T

4 i +

F

.....

'

m.4 da m.:...a: mumm mov t.n unewue m gg myg,

< = _ -

-

BLoeER 50s.6 SN. 6 IRELAV586.

jagt g ETER I:108W skuessegadG SN.45 Agg. M AaB8.qsa.22

- *

,

T'

slI6750dj 0508 @7503 06ca, o

88E-40444 5t04 I

edE-404M SN.84 25P6 M)

' 217-t Lt (LI)

g

,

".

ZS g

Fj

,

oc3

%

A Z

6c (CAS)

gy E

G l

l'

RTD-4 KE-t

"m i

WC I

Ilc, !

(cA4)

W TED I

I 61za 5*

I l

PHASE

,

7c (cat)

TE4559a TEMR

!

n l

I RTM h*

gd

• 'stt-s

-'scz-2 e-Z

17 80*C.

jj

,

ll

3o (CAR)

er *

l pc5 Ite'

i

,

"A g

o bh h~

Sc

!

%

l

)

8o, l)

.

i

~

ca-x ca-xs

..

>

>-

i 8E" l

..

_ SPARE.._.

U l

.

o g

l

. _ _.

. DETECTOR..

9c

-

ClRCOiTE ~._i--

D l

l econ

A IW

-

og

l WM#

g 32 c s

_. _ _ _ _ _ _ _' _.

,hzaX-Z gn 2311

,

.

j

,

Il.

MG-C SG

>

' '

(' ")

t17 - 2 L2-

-

(LI)

' '

~

23N6 c

A-touwpost tocATEDW I

GER BUS DUCT 0000ER

.t10WER RB. PANEL

2 GENERATOR BUS DUCT COOLER BLOWER GONTROL

'

.

(X-HIAW 12-5)

"

.

g

8

'A-g q

@

v

4, F'

_

~

t i

• e

>

a O

'...

.

.__

__

II