ML20217C216
| ML20217C216 | |
| Person / Time | |
|---|---|
| Site: | Fort Calhoun  |
| Issue date: | 07/09/1991 |
| From: | Gates W OMAHA PUBLIC POWER DISTRICT |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| LIC-91-185R, NUDOCS 9107150005 | |
| Download: ML20217C216 (35) | |
Text
.
~.. _
.. ~..
~. _..., _ _ _ _ _ _ _ _....
Omaha Public Power District 444 South 16th Street Mall July 9,1991 Omaha. Nebmska 68102 2247 LIC-91-185R 402/636-2000 U. S. Nuclear Regulatory Commission Attn: Document Control Desk Mail Station PI-137 Washington, DC 20555
References:
1.
Docket No. 50-285 2.
Letter from NRC (A. Bournia) to OPPD (W. G. Gates) dated November 7, 1990.
Gentlemen:
SUBJECT:
Additional Information on the Revised Fuel Handling Accident for Fort Calhoun Station In Reference 2, Omaha Public Power District (0 PPD) was requested to provide additional information on two items concerning the Fort Calhoun Station Fuel Handling Accident analysis. The first was to submit a revised evaluation of the fuel assembly radiological consequences with all fuel pins failed in the assembly.
Secondly, additional information on the ability to contain and control a potential release was requested.
The radiological consequences were recalculated assuming failure of 176 fuel pins (complete assembly) rather than 88 fuel pins previously analyzed.
The maximum resultant dose was 5.94 REM thyroid dose at the Exclusion Area y/g values based on NRC approved methodology.
oundary.
lhe dose consequences were lowered through the use of revised Also, experimental results in iodine decontamination of water pools were usea in the input assumptions.
The experimental work and the decontamination factors were previously reviewed I
and approvea by the NRC for the BG&E Calvert Cliffs Plants.
The resultant dose consequences of the revised evaluation meet the "well within 10CFR Part 100 limits" criteria of the Standard Review Plan.
In the calculation, summarized in Attachment 1, it is recognized that the VA-66 filter unit would be used during refueling and irradiated fuel movement operations for mitigating the consequence of an accident.
The analysis conservatively takes no credit for the capability of VA-66 in reducing iodine activity.
This assumption is considered conservative since tracer gas testing (summarized in att. 2) demonstrated as much as 40% of releases from the fuel pool would be directed through VA-66 filters.
910715o005 910709 5
p 45-5124 DNWreng tgopportung U'
U. S. Nuclear Regulatory Commission LIC-91-185R, Page 2 OPPD respectfully requests NRC review of this information by August 8, 1991.
If you have any additional questions, please do not hesitate to contact me or members of my-staff.
Sincerely,
- b. $. /W W. G. Gates Divisian Manager Nuclear Operations WGG/sel c:
LeBoeuf, Lamb, Leiby & MacRae R. D. Martin, NRC Regional Administrator, Region IV W. C. Walker, NRC Project Manager l
R. P. Mullikin, NRC Senior Resident Inspector l
~ _-
-ATTACEMENT1 14.18 FUEL HANDLING ACCIDENT (IN SPENT FUEL P0OL AND CONTAINMENT) 14.18.1 General
-An analysis was performed to determine the consequences of a fuel handling accident (FHA) in either the spent fuel pool area or the containment. A fuel handling accident is defined as dropping of a spent fuel assembly onto the floor resulting in the rupture of the fuel cladding of a fuel assembly in spite of many physical limitations L
imposed upon refueling cperation. Administrative restrictions on retueling procedures provide additional margin.
Before refueling operations start, the reactor refueling cavity is filled with approximately 215,00 gallons of-borated water. The boron concentration of the refueling water is increased to at least the minimum refueling boron concentration which is sufficient to maintain the reactor subcritical by more than Sh including an allowance for uncertainties, with all CEA's withdrawn (Ref. 1).
Periodic checks of refueling water boron concentration in the spent fuel pool area-further ensure that the fuel will always be in a subcritical geometrical array with k,, < 0.95, even though zero ppm boron concentration in the spent fuel pool: water is assumed in the spent fuel design basis.
The maximum elevation to which the fuel assemblies can be raised is limited by the design of the fuel handling equipment to assure that the minimum depth of water above the top of a fuel assembly required for shielding is always present (Ref.1).
This constraint applies in fuel handling areas inside containment and in the spent fuel 2001 area.
In addition, direct radiation monitors located at the fuel landling areas provide both audible and visual warning of high radiation levels in the event of a low water level in the refueling cavity and fuel pool.
The possibility of denage to a fuel assembly as a consequence of mishandling is minimized by thorough training, detailed procedures and the design of_ fuel handling equipment incorporating built-in interlocks and safety features.
Should a fuel: assembly be dropped or otherwise damaged during handling, a radioactive release could occur in either the containment or the auxiliary building.
The ventilation exhaust air from both of these areas is monitored before release to the atmosphere The containment release path to the environment is ducted (Ref. 2).through the Auxiliary Building to the discharge stack for the HVAC system.
The release path from the spent fuel-pool (SFP) to the environment is through other ductwork in the Auxiliary Building to the same discharge stack (release point) (Ref.
9).
In the event that the gaseous effluent monitors indicate activity levels in excess of the limits in the Technical Specifications, the containment ventilation flow paths will be closed automatically and the auxiliary building flow paths will be closed manually.
In addition, the exhaust ventilation ductwork from the spent fuel storage area is equipped with iodine absorbers (VA-66) which will be manually activated from the control-room whenever irradiated fuel is being handled.
14.18-1
The likelihood of dropping a spent fuel cask into the spent fuel pool is extremely low.
The spent fuel storage area crane is conservatively rated for the weight of the cask and is equipped with overload alarms and safety devices.
The safety features incorporated into the design of the main hoisting system of the crane preclude a cask drop accident by preventing a load drop in the event of a single failure in the hoisting or braking systems.
Interlocks prevent the crane hook from traversing any part of the spent fuel pool, thus precluding the possibility that any load can be dropped on the spent fuel.
The interlock consists of a key operated override switch which will permit bypass under administrative control for operation of the crane over the limited area of the pool used for spent fuel handling, cask loading and unloading. Operation over the entire pool area is also permitted in conjunction with a key operated interlock override switch and strict s
administrative control. A crane supervisor must be present when this switch is overridden.
14.18.2 Methods of Analysis Calhoun, plant-specific gas gap activity could be calculated for Fort Before a i
a bounding estimate of expe_ ed fuel temperatures had to be derived. This was achieved by compiling bounding radial fuel temperature distributions for Fort Calhoun fuel rods from available distrib(Ref. 5 and 6) computer code outputs.
The temperature FATES utions are bounding in the sense that they provide conservatively high temperatures which, subsequently, yield conservatively high gas gap activity levels. They also bound both the C ' fuel design and the Exxon-fuel design.
Exxon fuel temperatures used in the analysis are based on Exxon fuel characteristics obtained from OPPD for the Cycle 11 safety analysis and are presumed to be bounding for all Exxon fuel used at Fort Calhoun.
It should be noted that the temperature distributions bound fuel in fuel rods restricted to an internal pressure of no psia, in Fort Calhoun Station) greater than nominal RCS pressure (2100 I
Fuel temperature distributions were taken from the previous FATES analyses identified in Table 14.18-1.
TABLE 14.18-1 FATES ANALYSES UTILIZED AS E E fAM
._...DD1Griotion A
Exxon Fuel, Hot rod in hot bundle from Cycle j
11 safety analysis.
l B
C-E Batch M Fuel, Hot rod in hot bundle from Cycle 11 safety analysis.
C C-E Batch N and P Fuel, Design Analyses.
l I
l 14.18-2 l
The available data was manipulated to give conservatively hot fuel temperatures for a combined radial fall-off curve representing C-E Batches M, N, P and the Exxon fuel. This method clearly bounds assembly average condition and will bound minor variations in the fuel characteristics of later fuel batches.
The fuel temperature data for C-E Bctch M and the Exxon fuel (in the reactor during Cycle 11) were derived from a direct comparison of batch M and Exxon fuel temperature calculated in the Cycle 11 safety analysis with the power history (radial peaking factor fall-off curve) associated with the safety analysis.
fuel temperatures were calculated in the Batch The Batch N and P(N/P)The design an& lysis power history maintains rod N/P design analyses.
power (radial peaking factor) as high as possible without rod internal pressures exceeding the reactor coolant system pressure (d to calcu) late 2100 psia.
In the design analysis the fuel rod dimensions were biase a maximum pressure inside the fuel rod. This approach tends to make the fuel temperatures 100-200 F higher when the fuel to clad gap is open compared to the results from a licensing analysis. This temperature difference decreases as the fuel to clad gap decreases.
The Batch N/P rod power (radial peaking factor fall-off curve) is higher than that of the Batch M and Exxon fuel for a rod average burnup range of 25 to 45 GWD/HTU.
In order to achieve sufficiently high Exxen fuel temperatures, the fuel temperatures were ratioed up to the higher power levels.
The highest fuel temperatures were taken from the FATES cases of Table 14.18-1 based on selected rod average burnup categories as defined below:
0-25 GWD/MTU:
Exxon and CE fuel temperature data was taken from the Cycle 11 Exxon fuel and Batch H safety analysis (Cases A and B res)ectively). The power levels are the same as Batch N/P (Case C). Tie higher Batch N/P temperatures however, are due to the biased rod dimension (yielding conservatively high internal rod pressures).
26-39 GWD/MTV:
Temperature data for the CE fuel was taken from the Batch N/P Design run. The power levels and fuel temperature, in general, are higher.
The Exxon fuel temperatures are also higher at these burnups (Exxon fuel temperatures from case A were ratioed up).
45-55 GWD/MTV:
C-E and Exxon fuel temperatures were taken from cases A and B.
Fuel temperatures were hotter over the major portion of the rod for these cases.
14.18-3
N'
- The rods were modeled in '20 equal length axial nodes. Nodes 1--and 20 correspond to the ends-of the rod while node 4 through-17 make up the central-part of-the rod,--Fuel temperatures were taken from nodes 17, 18,- 19 and 20 since they generally tend to represent the highest-
. temperature at given' local power levels. These temperatures are-conservatively high for the purpose of calculating the fission gas gap activity because-they are based on the hot rod in the hot bundle analyses.
The fuel temperatures are consistent with the radial fall-off curve present in Figure 14.18-1 and Table 14.18-2.
TABLE 14.18 RADIAL PEAK _lblG 1 ACTORS
.VERSUS BJRbVP R00 AVERAGE RADIAL BURNUP PEAKING' FACTOR GWD/MTU 8.14 1.000 14.71 1.000 19.00 1.000 25.00 1.000 26.0t 0.986 28.00 0.973 30.00 0.954 32.00
'0.934 33.00 0.931 34.00 0.923 35.00 0.916 36.00 0.904 37.00
.0.892 38.00 0.878 39.00' O 866 40.00 0.797
-42.50-0.797 45.00 0.794-47.50 0.743 50.00 0.694 52.50 0.681 55.00
-0.665 14.18-4
Local nodal burnup versus rod average burnup is given in lable 14.18-3.
TABLE 14,18-3 LOCAL NODAL BURNUP VERSUS R0D AVERAGE BURNUP ROD AVERAGE LOCAL BURNUP, GWD/MTV BURNUP GWD/MTV N0 DES 1.20 2.19 3.18 4-17 0
8.14 4.07 6.51 8.14 8.96 14.71 7.35 11.76 14.71 16.18 19.00 9.50 15.20 19.00 20.90 25.00 12.50 20.00 25.00 27.50 26.00 13.00 20.80 26.00 28.60 28.00 14.00 22.40 28.00 30.80 30.00 15.00 24.00 30.00 33.00 32.00 16.00 25.60 32.00 35.20 33.00 16.50 26.40 33.00 36.30 34.00 17.00 27.20 34.00 37.40 35.00 17.50 28.00 35.00 38.50 36.00 18.00 28.80 36.00 39.60 37.00 18.50 29.60 37.00 40.70 38.00 19.00 30.40 38.00 41.80 39.00 19.50 31.20 39.00 42.90 40.00 20.00 32.00 40.00 44.00 42,50 21.25 34.00 42,50 46.75 45.00 22.50 36.00 45.00 49.50 47.50 23.75 38.00 47,50 52.25 50.00 25.00 40.00 50.00 55.00 52.50 26,25 42.00 52.50 57.75 55.00 27.50 44.00 55.00 60.50 i
14.18-5
Local nodal power versus rod average burnup is given in Table 14.18-4.
TABLE 14.18-4 LOCAL N0DAL LINEAR HEAT RATE VERSUS R0D AVERAGE BURNVP R0D AVERAGE LOCAL NODAL LINEAR HEAT RATE, KW/FT-BURNVP GWD/MTV N0 DES 1.20 2.19 3.18 4-f7-8.14 5.62 8.98 11.23 12.35 14.71-5.62 8.98 11.23 12.35 19.00 5.62 8.98 11.23 12.35 25.00 5.62 8.98 11.23 12.35 26.00 5.54 8.86 11.07 12.18 28.00 5.47 8.74 10.93 12.02 30.00 5.36 8.57 10.71 11.78 32.00 5.25 8.39 10.49 11.54 33.00 5.23 8.36 10.45 11.50 34.00 5.18 8.29 10.36 11.40 35.00 5.15 8.23 10.29 11.32 36.00 5.08 8.12 10.15 11.17 37.00 5.01 8.02 10.02 11.02 38.00 4.93 7.89 9.86 10.85 39.00 4.87 7.78 9.73 10.70 40.00 4.48 7.16 8.95 9.85 42,50 4.48 7.16 8.92 9.85 45.00 4.46 7.14 8.34 9.81 47.50 4.17 6.67 7.79 8.17 50.00 3.90 6.23 7.65 8.57 52.50 3.83 6.12 7.65 8.42 55.00 3.74 5.98 7.47 8.22 The Exxon fuel is always at a higher temperature than the CE fuel at a given burnup and power.
The temperatures derived through the analysis are considered conservative because of the following:
(1) They were taken from the hot rod in the hot bundle analyses.
(2)
Batch N/P temperatures were based on a maximum pressure design analysis which has resulted in higher fuel temperatures.
(3) The fuel temperatures were taken at the top of the rod where temperatures are the highest.
The maximum fission product activity in the gas gap of fuel assembly was calculated for use in calculating the dose rate at the Fort Calhoun site boundary following a fuel handling accident.
14.18-6
ANSI /ANS 5.4 (Ref. 7) provides an analytical method for calculating the release of volatile fission products from oxide fuel pellets during normal reactor operation.
When used with nuclide yields, this method will give the so-cal ed "ga) activity," which is the inventory of volatile fission products t1at could be availab'e for release from the fuel rod if the cladding was breached.
TheStandardconsidershigh-temperatures (uptothemeltingpoint?and low-temperature (where temperature-inde endent processes dominate releases and distinguishes between shor half-life (i.e., half-li e less than one year) and long half-life (i.e., half-life greater than one year nuclides).
ANSI /ANS-5.4 is intended to give a best-estimate prediction of the release fraction under steady-state conditions and is not applied if abrupt temperature changes are involved.
In certain calculations, best estimates are not practical and approximations are made that lead to conservative over-predictions.
The fission product release fractions were calculated for several burnups up to 55 GWD/MTV to determine when the maximum gas gap activity can be expected. The fuel rod temperatures used in the release calculations were obtained from the data presented above.
As shown in Table 14.18-6, the available data revealed that the maximum fission product release fraction can be expected at a burnup of 36 GWD/MTV.
TABLE 14.18-6
SUMMARY
OF IS0 TOPIC RELEASE FRACTIONS Burnuo Eaual Volume Radial Temoerature Distribution
- Isotope **
MWD /MTU T1
_l2__
T3 T4
__T1 T6
,.I-131 1E-131M 19-L31 30000 2380 2045 1745 1455 1209 9/3 0.1200 0.0626 0.0291 36000 2272 1955 1673 1401 1173 973 0.1224 0.0624 0.0299 40000 2014 1745 1518 1291 1091 918 0.0561 0.0270 0.0122 50000 1831 1618 1418 1236 1055 909 0.04C8 0.0223 0.0100 55000 1772 1573 1382 1201 1055 891 0.0488 0.0233 0.0104 Temperature in degrees Fahrenheit The long lived isotope Kr-85 is less temperature sensitive and more burnup dependent.
Its release fractions was calculated separately per the method of ANSI /ANS-5.4 at a burnup of 40,000 MWD /MTV and found to be 0.075.
1 14.18-7
~.
+
The corresponding fission product inventories were calculated for the peak linear heat rate for a fuel rod at 36 GWD/HTU (9.84 kilowatts per foot It was assumed that the entire fuel rod length as well as every other). rodinthefuelassembly,wasoperatingatthislinearheatrate to conservatively overestimate the fission product release.
The fission product ir.ventory was subsequently multiplied by the release fraction to obtain the gas gap activities.
The calculated gas gap activity for the hottest fuel assembly (i.e.,
greatest activity) after a three day decay is presented in Table 14.18 7.
TABLE 14.18-7 FORT CALHOUN FUEL ASSEMBLY GAS GAP ACTIVITY lsotope Eg] ease Fraction Calculated Activity (Cil I-131 0.122 4.53 E 04 Xe-131m 0.064 9.41 E 02 Xe-133 0.029 2.00 E 04 Kr-85 0.075 1.58 E 03 Fort Calhoun plant saecific fuel rod temperatures as a function of rod burnup were extracted from FATES computer code runs.
From this data, plant-specific fuel rod gas gap ar.tivities were calculated for I-131, Xe-131m, Xe-133, and Kr-85 as discussed above.
The method of calculating the radiological consequences of the fuel handling accidents is identical to that used previously for Fort Calhoun. The only changes incorporated since the previous dose calculations have been the use of plant-specific gas gap activities and updated atmospheric dispersion factors (Ref. 8).
To determine the extent of damage possible to a fuel assembly during fuel handling, it is assumed that the fuel assembly is drop)ed during handling.
The worst fuel handling incident that could occur is tie dropping of a fuel assembly onto the ficor. After striking the pool floor vertically, the assembly would rotate into a horizontal position; during this rotation, it is assumed that the assembly strikes a protruding structure.
For this analysis, a line load is assumed.
To obthin an estimate of the number of fuel rods which might fail in the event a fuel assembly were dropped, the energy required to crush a fuel rod and bend the entire assembly has been determined. The point of impact was assumed at the most effective location for fuel rod damage, i.e., the center of percussion.
Resistance to crushing offered by the fuel pellet is considered in the analysis.
Failure of the fuel tube by crushing absorbs the least energy;l rod f ailures.hence, the model results in a conservative upper limit for the number of fue This failure mode is applicable to the outer row of fuel rods only.
Since it is not possible to apply a line load beyond the outer row of fuel rods, the failure mode of rods in other than the outer rows will be bending rather than by crushing.
l 14.18-8
The energy absorption cap (Ref. 13)d b ability of the cladding was determined for various times in the fuel cycle' Conservatively assuming-that the kinetic energy at impact is fully absorbe y the portion of the fuel assembly between the-spacer grids :a strain energy vs -deflection distance relationship is1 developed.
Comparison of the relationship to the cladding energy absorption capability yields a deflection distance of the rod at which-the ultimate strain is produced in the cladding which causes clad failure.
These values are then compared to a geometric model.- of the fuel assembly in-order.to determine fuel rod' deflection and-the number of rows that would fail while absorbing the impact energy applied. This model-predicts failure of 1.6 rows. :Thus, no more than 28 fuel rods, i.e. two outer rows of rods, would be expected to fail.
Consistent with this discussion, the dose calculation presented here assumed-all 176 rods in a single fuel assembikl assembly) is a NRC review requiremen would fail and contribute to the gas release.
The failure of 176 rods that is consistent for all power p'(fuants (Ref. 12).
-14.18.3 Results
-The radiological consequences for a fuel handling accident in the spent fuel pool area and in containment are presented in Table 14.18-8.
TABLE 14.18-8 4
RADIOLOGICAL CONSEOUENCES OF A POSTULATED FUEL hah 0 LING ACCIDENT Dose Tvoe At the_EM;t Dose (Rem) 10 CFR100 Limit Thyroid 5.94 300 Skin 0.191 Whole Body -
0.072 25 At the LPZ:
Thyroid 0.106 300
-Skin 0.0034 Whole Body 0.0013 25-As the values in Table 14.18-8 show-the doses calculated for the spent fuel pool' area are well within the limits of 10 CFR100 (Ref 4) using conservative assumptions.
For the fuel handling accident in containment, the very conservative assumptions of all rods failin! ion yields' doses that also remain well within with no credit taken for containment isolation or atmos)here filtra the limits of 10 C R100,-
14.18-9
--e
14.18.4 Radioloaical Consecuences The radiological consequences of a FHA in the spent fuel pool area and containment are identical if no credit is taken for the charcoal filter (VA-66 in the spent fuel pool area. While VA-66 provides some filtration of the iod)ines released, a conservative assumption of no credit for iodine filtration is utilized.
The assumptions and input values used in the calculation of the radiological consequences are listed in Tabla 14.18-9.
The activities tabulated are those released to the water in the fuel pool not those released to the building atmos)here; the iodine activities released to the atmosphere would be lower than t1ose released to the water in the fuel
- pool, Since the fuel handling process is under water, essentially all of the iodine released from the damaged rods would be retained in the spent fuel pool water because of the preferential distribution of iodine in water over air.
Experiments in which air-steam mixtures containing iodine were bubbled through a shallow depth of water indicated that, for pools very dilute in iodine (such as spent fuel pools), only about lx10* of the iodine reached the water surface in the bubbles Additional evidence fn om two deliberate fuel melting excursion (Ref.10). experiments in the CABRI pool reactor indicated that only 5x10* of the iodine released from the melted feal elements escaped to the air above the core (Ref.11). To account for this preferential retention of iodine by the pool water, a decontamination factor of 1x10 is assumed, which corresponds to the lower value suggested in 5
Reference 10. No additional cred3 for iodine plate out on the Auxiliary Building surfaces was taken.
14.18-10
TABLE 14.18-9 PARAMEl(RS USTD IN EVALUATING THE RADIOLOGICM CONS LOVENC iS OF A FUEL HANDLING ACCIDENT PARAMETER INPUT ASSUMPTION BEFERENCE Source Data:
Power Level, (Mwt) 1500 USAR Radial Peaking Factor
> 1.65 Reg. Guide 1.25 Burnup (MWD /MTU) 16000 EA-FC-90-094 Decay Time (hr) 72 Tech. Spec.
Failed Rods 176 NUREG 0800, 15.7.4 Gas Gap Activity:
Isotope I-131 4.53 x 10' EA-FC-90-094 Xe-131 9.41 x 10' EA-FC-90-094 Xe-131 2.00 x 10' EA-FC-90-094 Kr-85 1,58 x 10' EA-FC-90-094 lodine Gab Activity (%):
Inorganic l-131 99.75 Reg. Guide 1.25 Orgaitic l-131 0.25 Reg. Guide 1.25 Activity Release Data:
Gas Activity Released 100 Reg. Guide 1.25 to pool (%)
Minimum Water Depth 23 Reg. Guide 1.25 Above Damaged Rod (ft)
Pool Decontamination Factors:
Noble Gases 1
Reg. Guide 1.25 lodine 1000 Reference 10, 11 Spent fuel Pool Area Filter Efficiencies (%):
Inorganic lodine O
Conservative Assumption Organic lodine O
Conservative Assumption Noble Gases 0
Conservative Assumption Atmospheric Dispersion Factors:
4 EAB (sec/m')
2.55 x 10 Calc. FC-05526 LPZ (sec/m )
4.53 x 10~'
Calc. FC-05526 2
Release location / Timing Ground Level Reg. Guide 1.25 Within 2 hr.
14.18-11
References:
1.-.." Refueling System", Fort Calhoun Station, Unit 1 Updated Safety Analysis Report, Section 9.5, Revision 7/89, 2.
" Radiation Monitoring", Fort Calhoun Station, Unit 1. USAR Section 4
11.2.3,. Revision 7/89.
3.
- Assumptions Use for Evaluating the Potential Radiological Consequences i
of a fuel Handling Accident in the fuel Handling and Storage Facility for Boiler-and Pressurized Water Reactors",-Regulatory Guide 1.25 (Safety Guide-25),'U.S. Nuclear Regulatcry Commission, March 23, 1972.
4.. Title 10. Energy, Code of Federal Regulations Part 100, Reactor Site -
Criteria, January 1979.
5.
" Fuel Evaluation Modal:
CE Fuel Evaluation Model Topical Report", '
CENPD-139-P-A, CE Proprietary Report, July,-1974 6.-
" Improvements-to Fuel Evaluation Model", CEN-161(8)-P, CE Proprietary
~ Report, July 1981.
7.
"American National Standard Method for Calculation of the Fraction l
- Release of Volatile Fission Products from 0xide fuel", ANSI /ANS-5.4-1982.
.8.
" Atmospheric Dispersion USAR Calculations", OPPD Calculation FC-05526,
- 11/2/90.
9.
- P&lD Drawing, GHD&R 11405-M-1
- 10.. " Iodine Cleanup in a Steam Suppression System", Diffey,_ H.R. et. al.,
- International Symposium on Fission Products and Transport Under Accident Conditions, Oak Ridge, Tennessee, CONF-650407,<Vol. 2, Pages 776-804 (1965).
11.
" Efficient des Pieges A Charbon Impregne vis-a-vis' des Produits de Fission Deg SM-110/10, ges lors D'u Accident de Pile _ Piscine", Dadillon, J. Paper Proceedinos Treatment. of-Airborne Radioactive Wastes Symposium, New York,~1968, IAEA.
(
12.
" Standard Review Plan for the Review of Safety Analysis Reports for L
Nuclear Power Plants LWR Edition",- NUREG 0800, June 1987.
L 13.
" Fuel Handling Accident and 8ounding Source Term", Engineering Analysis (EA) FC-90-94, Revision 0,1.0/3/90, h
i 14.18-12 L
--^---w r-7m er,-ee-
+r.w s
c.
+-
r.
,m--
, w
i 1.2 1.1 -
i4 o
u--o un i.l ll.13 s
t O.9 (J
g rJu g
r a
h 0h o
N 3
"x
%'N
- 0. 7 --
O.6 -
0.5 -
O 20 40 60 ROO AVLHAGC QUHtAJP, (AyQ/g{U l
Radial Peaking Factor Omaha Public Power District Figure vs. Bumup Fort Calhoun Station-Unit No.1 14.18-1
ATTACHMENT 2
\\
SUMMARY
OF THE i
TRACER GAS CHARACTERIZATION OF THE SPENT FUEL POOL VENTILATION SYSTEM at FORT CALHOUN STATION March 25 to March 29,1991
Introduction and Summary In order to assess the quantity of gas released over the Spent Fuel Pool which passes through the Fuel Handling Area Charcoal Filter (VA-66), and ensure that a fuel handling related radioactive release could be at least partially contained / controlled in the Auxiliary Building, a series of tracer gas tests were undertaken. A tracer gas, sulfur hexafluoride (SF ) was released over the Soent fuel Pool at a known mass flowrate. The resulting tracer concentrations passing through VA-66 were measured along with duct flowrates while measurements were taken to assess the ability of the HVAC system to control and contain the release.
Since it is easily detectable in minute quantities by means of electron capture gas chromatography, SF, is an ideal tracer gas for ventilation system performance investigations.
Analytical sensitivity to this gas ranges from 10 parts per million to approximately 50 parts per trillion. Thus, for reasonable injection concentrations, dilutions on the order of 10,000 can be easily measured.
All tracer gas measurements were performed by means of 3roprietary chromatographic instrumentation manufactured for field use by
.agus Applied Technology, Inc. (LAT).
From this data, the ratio of the mass of tracer gas passing through the VA-66 Filter to that released above the Spent "uel Poo! to that was found to be 41% 14%.
Additional measurement disclosed that the flow through VA-66 with two su) ply fans (VA-35A, B) and three exhaust fans (VA-40A, B, C) running was varia)le and resulted in fluctuations of approximately 25% in the flow through VA-66, it was found by comparing hot wire anemometer traverse data to flowrate data obtained by a mass balance tracer flow measurement, that a satisfactory estimate of airflow though VA-66 could be obtained by differentiating the flow measured downstream of VA-66 and the flow measured in the exhaust riser from the 989' level.
In the course of the measurements it was demonstrated that no re-entrainment of ventilation exhaust gases from the plant stack enter the auxiliary building through the fresh air intake VA-38.
VENTILATION TESTING TECHN10VES There are three principal techniques for quar +.fying the air leakage / ventilation rates within a structure:
1: the tracer oilution method, 2: the steady-state concentration method, anu 3: the constant concentration method. The tracer dilution method is a direct way of measuring the air leakage / ventilation which exists within a building under ambient flow conditions.
The steady-state concentration method is an indirect method; i.e., it measures the equilibrium tracer concentration within a ventilated area. This concentration can be related to the air leakage / ventilation rate if the tracer release rate is known. The constant concentration method is also an indirect method.
It measures the amount of tracer as a function of time required to maintain a constant concentration within a ventilated zone 1
i or zones.
The quantity of tracer injected can be related to the ventilation rate.
This third technique, the constant concentration technique, requires very sophisticated computer controlled release and analysis equipment.
At this time it is primarily a research tool and will not be discussed further.
TRACER DilVT10N METHOD The tracer dilution method has been used for a number of years to measure air leakage rates.
It is particularly useful in a single zone or at most in a small number of closely spaced zones. This technique entails introducing a small amount of. tracer gas into a structure and measuring the rate of change air cha)nges per hour, abbreviated ACH) can be determined from the lo(generally (decay in tracer concentration.
The air leakage / ventilation rate garithmic decay rate of tracer concentration with respect to time.
The principle of the tracer dilution method for measuring air leakage / ventilation rates may be developed briefly by considering the average rate L at which air leaks into a test volume.
This must equal the average rate at which air leaks out unless there is a steady increase or decrease in pressure within the volume.
The rate of change in the total amount of tracer in the volume is, dQ/dt (C, - C.)
L (1) where Q is the total amount of tracer, and C, and C, are the concentrations of tracer outside and inside.
If V is the total test volume, equation (1) may be expressed as, 1/V dQ/dt - dC,/dt (C, -C)
L/V (2) where L/V is the air leakage / ventilation rate in air changes per unit time and is often designated as I.
If the outside concentration of tracer is small enough to be neglected, equation (2) reduces to, dC/dt - -C L/V (3) where we have suppressed the subscripts, and tactily assumed that C always refers to the inside concentration.
Integrating equation (3) leads to,
C C,exp (4) where C, is the concentration at time t - 0.
Equation (4) may be rewritten to give I
L/V 1/t log, (C/C)
(5)
This equation is the theoretical basis for tracer dilution studies of air exchange.
Figure 1 provides an illustration of the use of the tracer dilution method to infer the air leakage rate.
2
CONSTANT FLOW METHOD A second method used to infer ventilation rate entails the use of constant flow source of tracer gas and measurement of tracer gas concentration build-up. This experimental technique is often called the steady-state concentration method.
If a constant flow of tracer S is injected into a volume V, beginning at some time t - 0, then the time rate of change in concentration within the volume i: given as,
-C L/V + S/V (6) dC/dt It is again assumed that the tracer concentration in the supply air entering the volume is negligible.
Solving this equation yields, exp
(-
L/V t)]
(7)
S/L [1 C
0.
For large values of time if it is assumed that C
. at t corresponding to steady-state conditions, the ventilation rate at any point is simply, S/C (8)
L The basic idea behind the test is shown in Figure 2.
In Figure 3 we provide a plot of equation (7) using values typical of those which might occur within a plant environment.
For convenience in interpretation, portions of the curve are denoted as Transient Region, and Steady State Region. The Steady State Region is of most interest, in this region the equality expressed in equation (8? is satisfied. Thus, interpretation of tracer concentration data un terms of leak rates can only be properly undertaken for data which lie in Steady State Region, i.e., af ter tracer initiation or ventilation change transients have damped out.
For regions in which L/V is a large number (i.e., high ventilation rate),
this technique is easier to use than the tracer dilution method because tracer concentration decay occurs rapidly in regions of high air change rate.
Often this rapid rate of decay precludes the measurement of but a few concentration decay data point in a meaningful time span, while a single measurement after attainment of steady-state in the constant flow source technique yields an air leakage rate.
EFFECTIVE LEAKAGE RATE It is possible to characterize leakage flowrates from a controlled area to adjacent areas by injecting tracer gas into the controlled are airflow.
Subsequent measurement of tracer gas within non-controlled areas provides evidence of leakage between controlled and non-controlled areas.
In addition, since the ventilation rate within an area is presumed known, measurement of a steady-state tracer gas concentration would allow calculation of an effective leakage rate L using equation (8).
3 I.
_____.____m_..-____%_
The effective leakage rate can be defined as the fraction of the total leakage from the controlled area into the non-controlled area. This rate can be inferred by using the steady state concentration technique. Knowledge of this rate allows assessment of potential-health hazards.
The concentra[ ion of any substance leaking into a non-controlled area from a controlled area can be found if tracer concentration within both the controlled and non-controlled areas are known.
The effective leak rate is then given by, L.,,
C _,,,,,u/ C.,,,a x L (9) where C o....,,,,
- Concentration of Tracer Gas in Non Controlled Area (Measured)
C,,,,,a
- Concentration of Tracer Gas in Controlled Area L.,,
- Effective Leak Rate into Non Controlled Area L
- Non-Controlled Area Ventilation Rate This concept of effective leakage rate can be expanded to encompass a duct inleakage measurement in which a negative pressure return duct passes through a potentially contaminated area.
In this case, tracer gas is seeded into the contaminated area and the resulting tracer concentrations in both the contaminated area and the duct are measured.
Equation (9) can be used to calculate inleakage by recasing it to the following form:
L a,,,
(Ca,m/C,,,)
x Qa,
(10) where L a,,,
Duct inleakage rate
=
C Tracer concentration in duct
%a C,
-= Tracer concentration in containment area 4
Duct flowrate Q%,,
Thus, even relatively small amounts of duct inleakage can be measured by establishing an appropriate value of containment area concentration.
DILUTION RATIO Often it is desired to know the probable concentration of given airborne containment at a number of locations which are separated from a given source location. An injection of tracer at a source concentration C,,,, resulting in a measured tracer concentration C
,,w at a location of interest yields a dilution ratio, D, given by, C.J C,,,,,
(l1)
D Most of the measured tracer data in this report are presented as dilution ratios.
4
i i
The significance of providing data in the form of-dilution ratios is that for a given location, if one knows the source concentration, the actual concentration at the sampling location can be easily calculated by multiplying the source concentration by the dilution ratio.
TRACER GAS MONITOR Testing of air samples for the presence of tracer gases in this study was performed by means of a prop ~rietary field-usable four channel electron-capture gas chromatograph manufactured by LAT.
A schematic diagram of an electron capture chromatograph is provided in Figure 4.
Operating characteristics of each channel of the LAT monitor are provided in Table 1.
All output from each chromatographic channel is displayed on a strip chart recorder, where relevant peaks are measured and recorded in a data 109 In general, the electron-capture gas chromatograph utilizes the high electron affinity of gases with halogen group elements to provide a measurable signal.
In the unit utilized in the study, all samples are injected by means of either glass or polypropylene syringes.
Injection is through a rubber septum located on an external sample fitting.
This septum prevents spurious contaminants from diffusing into the chromatograph and producing anomalous signals.
The gas chromatographic column, in simplest terms, operates to separate the various gaseous components of sample by selectively slowing down some gases relative to others.
The column can be thought of as a device to output the distinct componer.ts of a gas sample in a definite order.
The detector portion of the chromatograph consists of a tritiated t2 6anium foil encased within an electrically-conductive housing.
Specific pulse generator circuitry energizes the detector, initiating a flow of electrons. A collector wand within the detector receives the electrons and establishes a current flow which is amplified through an electometer circuit.
Should an electronegative gas flow through this stream of electrons, the number of electrons being collected, and hence the current, is decreased in proportion to the concentration of the gas resulting in a measurable signal on the strip chart output.
5
TABLE 1 EL S
R PH Variable
.Spniflqation Detector Type 300 mci TiH' constant-frequency Concentric Detector Carrier Gas Type Oxygen-free Nitrogen Carrier Gas flowrate 100 ml/ minute Column Type 1 meter Solid Sorbent Type Power Requirements 110-125V AC, 50-60 Hz. 3 Amps Output Ranged Electrometer Output to Strip Chart 6
i 1
{
l l
i DT[81PENTAL RLSMLIS In order to assess the of vicinityoftheSpentFhuantitkwht!otentialtocontaingasreleasedintheh passes through th al Poo Charcoal filter,ht to minimize disturbances to the ventilation system due to a series of tracer gas test were undertaken.
Testing was performed at nig maintenance and high personnel traffic during normal dayshift activities.
On each of three successive nights a different test was performed within the Auxiliary Building on the fuel handling ventilation system to provide information required to allow calculation of the ratio between the gas released over the Spent fuel Pool and the amount of this gas passing through VA 66.
Tracer gas injection was accomplished by means of a Matheson Model 8270 Mass flow Controller designed to inject and control the mass flowrate of gas over a range of 0-5 Standard Liters per Minute (SLPM) (0 to 0.0353 Standard Cubic feet per Minute (SCFM)).
For all of the testing performed at fort Calhoun Nuclear Station, the tracer injection source consisted of an aluminum cylinder containing a mixture of 18.47% sulfur hexafluoride (SF ) in nitrogen at an initial pressure of approximately 1000 psi. All of the tracer SCFM). gas injections were performed using an injection rate of 1.01 SLPH (0.00653 This tracer concentration and injection rate provided the optimum balance between the injection flowrate and the resulting concentration through the Auxiliary Building on one hand and the total consuuption of tracer gas during the proposed testing on the other.
A simplified schematic drawing of the tracer injection manifold is provided in figure 5.
Tracer samples were taken from the VA-66 ducting by means of a recirculating diaphragm aump, he discharge side was connected to the ductwork downstream of VA-housing, w111e t were withdrawn on the discharge side /4" diameter y means of disposable 66.
Connections were made using a 1 olypropylene tubing.
Samples of the pump polyproplene syringes in conjunction with a sampl fitting.
A schematic illustrating the sampic manifold is provided in Figure 6.
INITIAL TRACER RELEASE An initial tracer release test was performed on the evening of March 27, 1991.
The tracer release point was at the level of the Spent fuel Pool on the west side, midway between the north and south edges.
For this test, samples were drawn at many locations in the Auxiliary Building.
These samples were taken at timed intervals by means of disposable polypropylene syringes.
The testing performed on March 27 was designed to elicit information on how the ventilation system and the building would respond to a tracer test. As such, an incomplete set of data were obtained.
It is clear from these data that the test was not run sufficiently lon estimate of the equilibrium value of tracer concentration to be made. g to allow an The equilibrium value is required since this value when combined with the volumetric flowrate through the ductwork, allows the mass flowrate to be calculated.
This mass flowrate can then be directly compared to the mass flowrate injected above the Spent Fuel Pool.
7
A second observation regarding this test is that even if sufficient data had been obtained to estimate an equilibrium concentration value, it would still not be possible to reliably calculate a mass ratio, since durin the performance of the test, the air flow through VA-66 fluctuated ap)roximatel 25%.
During this test it was not possible to maintain a constant flow tirough VA 6 with three exhaust fans (VA-40 A, B, C) running.
MSji RATIO DETERHitB110N On the evening of March 28, 1991, a second tracer injection test was performed with sampling at the locations indicated in Tables 2 and 3.
Sampling for tracer during this test was performed over a one hour period. A different interpolations of tie same data. gure 7. plot of the tra history upstream of VA 66 is arovided in fi The two curves shown represent An estimate of the error of this measurement can be made by looking at the RMS errors associated with the individual components making up the ratio.
Since the ration can be written as Mass Ratio (Cm Om)/(Cwc, Qwcr)
(12) the RMS error can then be expressed as i(fm' 4 E o,'
Ewe,'
+ Eon, ) "'
(13)
Em>
+
1 where Tracer Injection Concentration at Spent fuel Pool C;
Q; Tracer Injection flowrate Tracer Concentration Upstream of VA 66
- Cwe, Flow in Duct Upstream of VA 66
- Cwe, Enu.
Root Mean Square Error Estimate E;
Error in Injection Concentration Co, Error in injection-Flowrate Exc3 Error in Duct Concentration Eoo Error in Duct flowrate Estimates of the various errors are provided in Table 5 and lead to a value of Enus i0.095 or 19.5%
(14) 8
.~
TABLE 2 SAMPLING LOCATIONS 81mg location VA-38 Downstream of VA 38 liVAC Supply (fan)
VA-40 Downstream of VA-40A, VA 40B, VA-40C Junction (Exhau.t fans)
VA-66 Immediately Upstream of VA-66 Round Riser Exliaust Riser Adjacent to North Side of Cask Decontamination Room Square Riser Exhaust Riser Adjacent to Round Riser from 989' Level Station A NW of fuel Pool on 1025' Level Station B SW of Fuel Pool on 1025' Level Corridor 26 Adjacer.t to Open Hatch at Corridor 26 Room Al 100 Center of Room Al-100 (Auxiliary Building Operator Area 9
TABLE 3 Concentration at location Time VA 30 VA-40 YA-66 Round Scuare Station Station Corridor Room im).n)
Riser Riser A
B 26 Al-100 0(BG) 2. 2
O.59 2.1 1.3 0.14 0.49 5
0.99 1.5 0.88 0.41 10 10.5 17.8 4.8 16 15 18.1 47 17.8 23 20 27.5 83.5 20 205 25 34.2 240 28 30 39.0 280 250 345
>200 0.27 45 0'"
51 380 30 440 440 60 0'S 84 470 500 500 500 All Concentrations in Parts Per Billion (ppb)
Taktn from Duct inside Auxiliary Building Taken of Roof NOTE:
Background (BG) sample taken BEFORE onset of tracer injection.
10
TABLE 4 JNlEAKAQE VPSTREAM Of VA 66 f.lLTER TRAIN Tracer Tracer VA 66 Duct VA 66 Duct Injection injection Flowrate Concentration Hass j
Rate (SCIE 1 Concentration (%)
(SCfM)
(pob)
Ratio (7d 0.0353 18.47 5366 470 39 0.0353 18.47 5366 520 43 Sample Calculation Tracer Injection Rate 0.0353 x 0.1847 0.00652 5366 x 4.7 x 10' Tracer flowrate lhrough VA 66 0.00252
=
Ratio 0.00252/0.00652 0.387 39%
TABLE 5 ESilMATES FOR RMS ERROR CALCULATIDS Eva 10.01 a
Em 10.01 Enva i0.05 Eon fo,og 11 I
-~,--
~.
What can be seen by comparing the error estimate with the two values of the mass ratio provided in Table 4 is that the RMS ERROR is approximately the difference between the two estimates. Therefore the best estimate that can be made is to average these two ratio and apply the RMS ERROR calculated above to yield a value for the mass ratio of 41% 14%.
Return air samples were taken downstream of VA-38 inside the Auxiliary Building and showed evidence of tracer concentrations.
In order to investigate whether this tracer was entering the ducting within the Auxiliary Building or whether these non-zero values were evidence of re-entrainment of tracer expelled through VA-40 via the plant stack, and re entering through the fresh air inlet, VA-3B, two of the measurements (at 45 and 60 minutes) for VA-38 obtained during the test were taken at the air supply outside on the roof of the Auxiliary Building.
Both of these samples showed no evidence of tracer gas thereby eliminating tiie possibility that the concentrations measured within the Auxiliary building were at least partially 4
the result of reentrainment of Auxiliary Building Exhaust.
CONCLUSIONS Two major conclusions can be drawn from this limited study on the Auxiliary Building ventilation system.
The first and most important is that the VA 66 filter will process approximately 41% by mass of any gaseous release from the Spent fuel Pool.
The second significant conclusion is that the flow through VA 66 was not particularly stable with two supply and three exhaust fans in operation and that in fact, the flow through VA-66 for at least part of the test differed very little from that obtained with only two supply and two exhaust fans o)erating.
Current procedural direction is given to operate with two supply and tiree exhaust fans during fuel handling o)erations.
This is to obtain maximum flow through VA 66.
These tests indicate tTat the optimum mode of operation is with less fans operating.
The tests have indicated that the radiological releases from a fuel handling accident will be filtered, to a limited degree by VA 66 and that the release can be contained / controlled in the Auxiliary Building without adverse effects on the health and safety of the public or operational personnel in the Auxiliary Building or the Control Room.
12
~. - - _. -
FIGURE 1 TRACER CONCENTATION DECAY _TE_ST.
AIR L.EAKAGE BY CONCENTRATION DECAY (ASTM E-741-83)
, :::t. : ::.:.d:.. i (1!.:m:.:.::::tt: :T. :.y;I:*tilll. a:::
Fl/O:$ii!!li@....'W...*:::*:.i !!.!;:.:.. e '.:gi i.!
.wi
- ~
.. ::Q".:..::a.
......,.;.;.,.',$i.W.Y.if.7 q:!i!!!!!!.dI.
j "'i.i.
!!)A!!!!.
'D7.!!!}' i
)d(*i volu=e
- Y
!!7.fh.I.iir,*)pi::< !!I i N.Nf[N.'Ie'l$.i:'YIt.:h5 hj::
Leak Rate
- l
.ts.. :*;;:....!!.. :".: !"h.::.,.!!,:,:!;'
.if:$ns!!l,i!ii!"..li!!:i j
i:.
f.9:::j.:
- 1)::....
3.,.
- i::
- e....:n: ::! ; ;..
- i...:;. ~. W:.......<:*.:n.".::'...:e:g......
itete;.:
. =.......i::' *:t. ;.n..:.n.:. *,
. ". :*...t:: :
- q.
>.'.:::: : :.:. ;l::.,,
' ::.*:.:::ll..;
.:: 'A..?.:.' ;,.a.1".;' :,;.:.
- .~t
- .:v...
! $,i$,
'l i f'.'ilh!M..b,! :ii !
e
\\
I.:.
'r'f' :.
.:i);ii:'
I*
I"I'CS
- E"
$.l;{,';t,..
/...II'!.(!. *'!!T..:.!j:'!.!.!,'$'hi.
h*i
- !*I
,'I[,.
W*; */
2.
HomogeniIe
'.:.I..i.i::$
I...
I:!,i.ih "'
' :.#$fI.!ill:. E ini l:
)*
Measure Deesy Homogeneous Concentration 4
Best for Relatively low C
C, exp (- h t) t.eak Rates Concentration Time D
0 5
Slope
- Air Change Rate C
3 (Infiltration)
C 10 3
j 5
C.
20 W
3 C) 30 m
O Time (Minutes)
FIGURE 2 LONSTANT INJECTION FLOW TEST CONSTANT FI,0W TEST 1
/
l Volume Y
I
- eak Rata = L L
..) !
n:::
Y
.. Y' M;. :: :.
..0 P:-
k:;,.)ppn&Q
'.l.,i,j I
1.
Inject Tracer at Constant neurate (P) l 2.
Howevenire 5F now 3.
Measure Concentration in Aoss 5
source 4.
Best for Relatively High Leak Aates w
J P
- Coevillbrium "
C
- - - - ~ ~ ~ -
millMus 5
.?
5 W8 i
i e
i i
e Time
_ ~ -. - _ _ = -. = _ -...
FICURE 3
.i PLOT OF EQUATION (7)
USING TYPICAL INPUT val,UES 10.0 i.
~
Transient Steady State Region Region t
t W
kW Dl g.
8 Steady Stato e
Valuees2.5 x 10~T M
.E 1.0 e
3
,e co C
1 I
4 0.1 0
0.5 1.0 1.5 2.0 2.5 Time (hound l
l i
l t
, _ _. ~... _ - - -,.., _,. _ _ _ _
-..,_,.,.m.-,
. -.. - - -.. - ~...,. -. -.... -. -.
. -.. - ~. ~. - -..... - - - -. - -..
i FIGURE 4 l
SCHEMATIC DRAWING
-0F AN ELECTRON-CAPTURE GAS CilROMAT0 GRAPH Vent Sample in A/
l Carder Carrier 9
Flowmster Shut-Off i
j Valve i _
V i
3 G.C. Column Sample r-----"
E.C.
Valve Sample.
+
Inject Valve TiOMi Detwor i
Signal l-----
(May Be Heated) a 3
Metenng Valve
' f Sample To Pump ->
Pump Signal Electmnics
/Y Carner input
Release l
Point i
f Flow Control Element l
l.
l l-
- 1XTO
?
00 00 i
l 1
Flow Control
- Panel and Readout i
Source of SF in N g
2 1
l 1
l l
L l
~_-
...,._. _.-,... - -... - _ -~.. _ _ _... _.. _ _ _,., - _ - _. - ~... _ _ - - _ _ _ _ _. _ - -.
FICURE 6 SCllEMATIC DRAWING OF TRACER SAMI'!,ING MANIFOLD VA66 l
Sampling Syringe l'
G A
Septum j
Fitting Pump
{
l FIGURE 7
.. o.%.
s.
CONCENTRATION DATA AT VA-66 FOR TEST ON MARCH 28, 1991 1000 e
i i
i e
i i
i a
i
---~~~-_
i
/
p/
Cf 100 C
\\
o
~
i x
l w
8 ua 2:
ceo c
8 d
~
te m
i 10 l
l' l
1 I
1 1
3 f
I l
l j
f 0
10 20 30 40 50 60 70 80 90 100 Time (min) l NOTE: The two curves shown represent different interpolations l
of the same data.
l i
- - - _ _ - - - - _ _ _ _ _ _ - - -,, - _ - - - - - - - - - - - - - - - - _ - - - - - - - - - - - - - - - - - - - - - - - - -