ML20101F181
| ML20101F181 | |
| Person / Time | |
|---|---|
| Site: | McGuire  |
| Issue date: | 11/18/1991 |
| From: | Benson L, Srinivasan J WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | |
| Shared Package | |
| ML20101F179 | List: |
| References | |
| NSD-TB-91-09, NSD-TB-91-09-R00, NSD-TB-91-9, NSD-TB-91-9-R, NUDOCS 9206240322 | |
| Download: ML20101F181 (11) | |
Text
____ _ _ _ _ _ -.
OVER TE'.1PERATURE DELTA-T (OTDT) SCALING TB. 91-09-P0 WESTINGHOUSE NSSS PROCESS CONTROL SYSTEM U#*
1 m e/91 ALL PL ANTS WITH W OTDT REACTOR TR!P FUNCTION 492!320 References PLANTTECH SPECS PL&S. SCALINGA"ects Safety No] l Sneet Yes X MANUALS. NRC INFO NOTICE 31-52 3 c, g Re'ated Ecuroment INTRODUCTIQN NRC information Notice 91-52 discusses recent events where improper scaling of the Over Temperature Delta-T (OTDT) protection channel allowed the average temperature (T Avg) lead / lag compensation module to saturate before the T-Avg input reached the upper limit of its range. Saturation of this module prevented further reduction in the OTDT setpoint as T. Avg continued to increase. This channel was, therefore, ineffective in performing its intended safety function.
Although all reported incidents involved 7300 series process sa ipment, the potential for a similar situation rdso exists for 7100 and Foxboro equipment. For 7300 equipment, the T-Avg saturation condition was eliminated by red;stributing the gains on the OTDT setpoint summing amplifier and the s
lead / lag compensation mo<Nie. The input resistor of the OTDT summator was changed from SCk ch. s to 24.9k ohms and the gain of T-Avg lead / lag module was reduced by a factor of 50k/24.9k - 2.008.
These changes ensured that the OTDT setpoint would reach the trip setpoint before the T-Avg module output saturated.
This Technical Bulletin addresses these modifications and identifies a potential transient concern, solution and recommendations.
BACKGROUNQ The OTDT Trip is dt. signed to provide primary protection against departure from nucleate boiling (DNE) dunng postulated condition ll events in Westinghouse reactors. The trip function sperates by comparing the temor, ature difference (DT) between the hot leg and the cold leg of each loop to a calculated f.one. Informaton, if Aequired, may to Obtained from t*e ongnator Teiechone 412-374-5602 or (win) 284-$602 ongnator A pproval
. S Snnivasan. Control Systems Analysis L R. Benson Domestic Customer Pre;ects N P. Muener, Mgr, Control Systems Design Analysis s.mer weerroue E een Cecmanon ect :ts eracecos -m aay w aca~y or recrosertawn *+ <secoct ee accuan :ccc.evess :r.sevess at me normason :ceta.rw e es recon or assu rae any resocesosty er aaocy cr camage amen *av wt trom me usa of wm c'a axn NSO 2067 0 01M 9206240322 920615 PDR ADOCK 05000369
NSD-TB-91-09-RO Page 2 of 8 setpoint (OTDT). A reacter trip is initiated wnen two or mere locp DTs exceed their setec4nt.
Several terms such as, loop average temperature (T-Avg), pressurizer pressure, and neutron flux d stnbution in tne ccre (F DELTA i), are factored into the OTDT tnp serpoint ca'cu!ation (Equation-1),
The setpoint is tyoically expressed by the followirg equation:
{K - K (1 + Tau 3s)/(1 + Tau 2s)(TAvg*IPe +
33 -
3 2
K (P - Pp,) - F(Delta 1)j'[ DELTA-T)
(Ecuation-1) 3 where:
K,K,K3 are gains, 3
2 Tauis, Tau 2s are the leadMag time constants on T-Avg, Tp.,
is the reference T-Avg, typically nominal full power T-Avg P p.,
is the reference pressurizer pressure, typicaily nominal pressunzer pressure F(Delta l) is the Delta-1 penalty 1
DELTA-T is tha full power DT T
is the m6asured Average temperature Avg P
is the measured Pressurizer pressure.
When considering implementation of the OTDT setpoint in the protection system, Equation-1 can be reduced and wntten in the voltage form as follows:
VoToT - G. [B - G, V + G V - V (Deta-0]
(Equation-2) 3 s
i p
p F
where:
l VoToT OTDT setpoint in volts l
G gain on the OTDT summer 3
i t
NSD-TB 91-09-RO Page 3 of 8 8
bias on the OTDT summer
=
3 G,
gain on the T-Avg module
=
GV t
t voltage output of the T-Avg lead / lag module G
gain on the pressurizer module
=
p V3 volta 7 squivalent of the pressurizer pressure
=
V (Detaa:
voitage equivalent of the F(Delts-l) penalty F
=
The setpoint calculation in its simple form is shown in Figure-1. As can be seen from the above equations and from Figure-1, as the T-Avg increases, the OTDTsp decreases; as the pressurizer pressure increases the OTDT increases and the Delta-l penalty abays decreases the setpoint.
3p DKQUSSION lf the gains and biases in Equation-2 are not distributed properly some of the terms in Equation-2 could reach their maximum (saturate), before the OTOT setpoint reaches its tripped condition. As an example, consider the typical case where the input range of T-Avg is 530 F to 630 F. This corresponds to O to 10 volts for 7300, or 1 to 5 volts for 7100 or Foxboro equipment. If the plant is operating at a reduced power level and is at the middle of its T Avg range (580 F) and the gain G on the T-Avg module is 1.0 (i.e., greater than one), then for the 7300 equipment the output of the T-Avg module in steady state will be 5 times 1.6 - 8.0 volts. If the power levelis then increased and T-Avg increases to 595 F (65% of its range) the T-Avg module output at this new steady state should be 6.5 times 1.6 - 10.4 volts. However, due to the hardware limitation the output will reach its maximum of 10 volts and its overall contribution to the setpoint will not change for any further temperature increases.~
This condition can be resolved by distributing the gains in the Equation-2,i.e.. decrease the C and 3
correspondingly increase the gain G The bias B, gain G, and gain on '/ (Dema i) also need to 3
3 F
be reduced accordingly.
Transient response situations must also be considered to assure proper operation of the hardware. Specifically, during a transient the amplification associated with the lead / lag compensation unit, (used to anticipate the temperature response of the Reactor Coolant System) could cause saturation preventing further OTDT setpoint decreases on additional temperature (T Avg) increases.
t m
NSD-TB 91-09-RO Page 4 of 8 To illustrate this saturation effect, consider a temperature transient superimpcsed on the in;tial steady state conditions (580 F) used in the previous example. Figure 2 gives the output of the T-Avg modu!e, with the leadag compensat:en, to a 2 F/sec temperature increase (pcstulatec rod withdrawal event) for a typical equipment seti ? (G; - 0.8, L/L = 28/4). Note tnat even tnougn the lead / lag module gain is less than 1, the T-Avg module saturates afir acout 19 seccnds (t - 5 sec). At this time the input T-Avg has only reached 608 F (580*F 14 sec ::mes 2"Fisec),
o which is only about 80% of its possible range.
Even though, the T-Avg module saturation during transients may be unavoidable, the ga:ns can always be redistnbuted such that the OTDT setpoint reaches a minimum to ensure a tno, i e.,
OTDT setpoint reaches the minimum of its range (0 volt, or 1 volt), befcre the T. Avg module saturates (refer to Figure 3). A technique for achieving this is outlined in the following section.
1Q18) TION The unconservative impact on the OTDT setpoint calculation causeu by steady state er transient saturation of the T-Avg lead / lag module can be avoidN as follows:
1.
Set the gain on the T-Avg lead / lag module to be less than unity. This will keep the T-Avg module from saturating over the entire input range of T Avg in steady state.
2.
Evaluate Equation-2 to determine the Bias (8 ) and OTOT summer gain (G ) such that 3
3 the summer output rear' v a minimum to ensure a trip condition (output equal to 0 v, or 1 v) before or as the og lead / lag mocule output reaches saturation (output scaal to 10 v, or 5 v). This is done with tne prsssure and Delta-l inputs ta the Summer at. ting, to the maximum extent, to keep the setpoint above the trip value.
The second step is illustrated in the following example using the 7300 equipment voltage ranges (0 to 10 v).
Initial conditions (referenced to Equation-2):
I a)
The output of the OTDT Summer is at the minimum, V ror s 0 o
b)
There is no Delta-l penalty, Vg(o,qa i) - 0 v c)
The T lead / lag module reaches saturation, G,
- V - 10 v Avg t
l l
d)
G V is evaluated at the maximum pressure (usually 2500 psig) p p
i l
i I
NSD-TB-91-09-RO Page 4 of 8 To illustrate this saturation effect, consider a temperature transient superimpcsed on the initial steady state conditions (580 F) used in the previous example. Figure 2 gives the cutput of the T-Avg moou!e, with the lead. lag compensation, to a 2 F/sec temperature increase (pestulated rod withdrawal event) for a typical equipment setup (G - 0.8, L/L - 28/4). Note that even though tne lead /iag modu!e gain is less than 1, the T-Avg module saturates after acout 19 seconos (t - 5 st~). At th's time the input T-Avg has only reached 608 F (580 F + 14 see tmes 2 Fisec),
o which is only about 80% of its possible range.
Even though, the T-Avg module saturation during transients may be unavoidable, the ga:ns can always be redistributed such that the OTDT setpoint reaches a minimum to ensure a trio, i e.,
OTDT setpoint reaches the minimum of its range (0 volt, or 1 voit), before the T. Avg module saturates (refer to Figure 3). A technique for achieving this is outlined in the following section.
SOLUTION The unconservative impact on the OTDT setpoint calculation caused by steady state or transient saturation of the T-Avg lead / lag module can be avoided as follows:
1.
Set the gain on the T-Avg lead / lag module to be less than unity. This will keep tne T Avg module from saturating over the entire input range of T-Avg in steady state.
2.
Evaluate Equation-2 to determine the Bias (8 ) and OTDT summer gajn (G ) such that i
3 3
the summsr output reaches a minimum to ensure a trip condition (output equal to 0 v, or 1 v) before or as the T Avg lead / lag module output reaches saturation (output eoual to 10 v, or 5 v). This is done with the pressure and Delta-1 inputs to the Summer acting, to the maximum extent, to keep the setpoint above the trip value.
The second step is illustratad in the following example using the 7300 equipment voltage ranges (0 to 10 v).
Initial conditions (referonced to Equation 2):
l a)
The output of the OTDT Summer is at the minimum, Voror s 0 b)
There is no Delta-l penalty, V (Delta-l)- 0 v F
l c)
The T lead / lag module reaches saturation, G V - 10 v Avg t
i d)
G V,is evaluated at the maximum pressure (usually 2500 psig) p L
t
NSD-TB-91 09-RO Page 5 of 8 Solving Equation-2 under these conditions will give a value of G.
3 G -[B 10 + G 10} s 0 3
3 p
(Equation-3) or, G a[G 8 + (G G;) 10)/10 (Equation-4) 3 3
3 3
The value of products G B, G G and G G can be determined by cornparing Equation 2 3
3 3
p 3
to Equation 1 reduced to voltage form, using plant specific parameter ranges and equipment type (7300 in tnis example). Once G is calculated, the bias B and the gains G: and G and Foenag) 3 3
o need to be calculated based on Equation-2 and the products G 8, G G and G G.
3 3
3 p
3 t
RECOMMENDATIONS Scaling of the OTDT channel shoulo be examined to confirm that saturation of the T-Avg lead / lag modulo will not occur in the steady state and that the channel gains are distributed such that, during transient conditions, saturation of this module would occur only after the channel has developed a trip setpoint.
Proper operation of the steady state is assured if the gain of the T-Avg leadAag module is less than unity.
Equation-2 can tm used to verify the proper functionality of the OTDT channel under transient conditions. VoTDT s calculated using the plant specific values for the terms, G, B, and G V
i 3 3 o
p (under maximum pressure condition) and assuming the maximum value at the output of the T Avg leaoAag module (10 v or 5 v) and the r '1imum value (0 v or 1 v) of V (Dema-0. If the value of F
"VoTDT* is equal to or less than the minimum of the equipment (0 v for -*300, or 1 v for 7100 or Foxboro) then the OTDT scaling in done properly. If the value of "Voror" does not meet this criteria, then the gain G and bias B must be adjusted as outlined in the solution section. Once 3
3 G is calculated, the gains G,, G, and Fo, tag) need to be calculated based on Equation-2.
3 l
l l
I l
)
. e 6
OTDT CIRCUIT BLOCK DIAGRAM DELTA-TEMPERATURE (DT)
T-AVO N
LEAD / LAG SNPUT T-AWG MODULE H
DELTA OIM ooTo pgyg FUNCTION H
OTDT SETPotNT SP REACTOe oToy FIDfLTA-8 GENfMATOR suuusNO AMP mp esSTAatE g
1
= /
StA8 x
.-M..
AM9 l
l l
l FIGURE 1 JStO91A
.a -
m J.
m
)
OTDT Temperature Lead / Lag Compensation Gt = 0.8, dT/dt = 2*F/sec, L/L = 28/4 d
12
--~'- ----
10 Output (Tavg Lead / Lag) 8 input (Tavg = 580 + 21)
E o
6 i
4 1
l 2
n O
O 5
10 15 20 25 30 35 40 45 50 s
Time (seconds) 1, I
l FIGURE 2: OTDT LEAD / LAG MODULE SATURATION i
i
)
i
,~,.,---.-----.--..----------w--
a...-~--
+ - + -
---~..--.----#-
Maximum Rar ge (10 V or 5 V)
Tavg Lead / Lag Output y
Module Saturates Tavg Ramp input Trip setpoint reached before module saturates OTDT Setpoint Output Trip Setpoint Reached A
Minimum Range (O V or 1 V) l l
l
- _1
_.. _.1
.1.
a Time FIGURE 3: ILLUSTRATION OF SOLUTION
.