ML19343B805
| ML19343B805 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 12/24/1980 |
| From: | Mills L TENNESSEE VALLEY AUTHORITY |
| To: | Schwencer A Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 8012300566 | |
| Download: ML19343B805 (13) | |
Text
_ _ _.
TENNESSEE VALLEY AUTHORITY e
C H A T T A NOOG A, T E N N ! '# S E E 37401 400 Chestnut Street Tower II Decer.ber 24, 1980 Director of Nuolear Reactor Regulation Attention:
Mr. A. Schwencer, Chief Licensing Branch No. 2 Division of Licensing U.S. Nuclear Regulatory Comission Washington, DC 20555
Dear Mr. Schwancer:
In the Matter of
)
Docket No. 50-327 Tennessee Valley Authority
)
During a meeting with your staff en the hydrogen issue on December 18, 1980, TVA comitted to provide information on equipment survivability for Sequoyah Nuclear Plant. Enclosed is the information you requested.
Very truly yours, TEl'NESSEE VALLEY AUTHORITY
[ '[ )l'}.b 3
L. H. Hills,'Hanager Huolear Regulation and Safety Sworn t;o subscri ed before r.e thic b day of n )f A 1980
[
lQht
'bb
- b LLL<
liotary Public Hy Comission Expires G-Enclosure u
g t7
[..
C C '.
t 8012300
[,,,... u opportuna y empiover
i l
1 ENCLOSURE 4
CALCULATED HEAT SINK TEMPERATURE WITH BURN (DERIVATION OF CURVE D FROM FIGURE B.10F TVA QUARTERLY iiEPORT) 1 A concern with Class 9 events is the potential temperature effects on equipment inside containment. One way to address this concern is to analyze an event and obtain an atmosphe:ic temperature versus time j
profile and then impose this profile on a piece of equipment.
Determining a temperature profile with reasonable levels of conservatism for a design basis large LOCA is relatively easy at this j
time because computer codes have been developed over the years that model the important phenomena involved in the accident. This is not the case for events that result in core degradation. The computer codes available to evaluate Class 9 events are in developmental stager and important models are not yet available. As an example, the '; LASIX 4
code does not model heat sinks within the containment. Energy and mass are added to containment and CLASIX calculates atmospheric temperatures after a hydrogen burn (Figure B.1, Curve E).
If heat
{
sinks are included, energy that was used to obtain high air temperatures would be absorbed by the structures and equipment. This means the calculated air temperature must go down or energy is not conserved.
1 Since the amount of energy determines the temperatures in either the atmosphere or the structures and equipment, ar. energy balance approach was chosen to evaluate temperatures inside contaircent. The energy releases are then partitioned between the various heat absorbers in the containment-(i.e., the ice, heat sinks, and atmosphere).
1 To obtain curve D, a small break LOCA with no high pressure injection (S D) was modeled with the MARCH computer code to obtain the mass and 2
energy releases needed for the analysis (the MARCH output was also usea in the CLASIX run). The first 2500 seconds of the accident was analyzed with a conventional containment compgter code (LOTIC). This code showed an atmospheric temperatgre of 160 F.
It was assumed that all the heat sinks were also at 160 F et 2500 seconds. The analysis was terminated when 75 percent of the zirconium in the core had been oxidized (9250 sec).
The MARCH results showed, for the time period of interest (2500-9250 sec)g 142100 pounds of steam were releasM with a total energy of 1.56 x 10 Bgu and 1620 pounds of hydrogen were released with an energy of 3 14x10 Btu because of the high temperature of the gas as it leaves the core. It addition curve D was generated assuming all the hydrogen burned with a heyt rate of 61,000 Btu /lb resulting in a burn energy of 9.88 x 10 Btu. Summgngtheenergyreleasesprovidesthetotalenergy release of 2.58 x 10 Btug Based on results from CLISIX, it was determined that 1.48 x 10 Btu was removed from the steam condensed in the ice bed. The air return fan also circulates air through the ice condenser at a rate of 4.8 million cubic feet per hour. This gives a maximum flow rate of 480,000 pounds per hour based on the upper compartment air density and the air return fan flow rate.
It'was assumed
~
in the analysis that the air entering the ice condenser was on o
the average 200 F hotter than the air exiting to the upper 7
compartment. This air flow results in the removal of 4.5 x 10 Btu.-
Taking the total energy from MARCH and pubtracting off tha energy removed in the ice bed leaves 6.49 x 10 Btu which must be distributed between the atmosphere and the heat sinks.
It was decided to neglect the heat capacity of the atmosphere to maximize the temperature rise in the heat sinks. The heat sinks used are listed in Table 1 and represent the structural steel in the Sequoyah lower compartment and the exposed portions of the steel containment shell in the dead-ended compartments. The volumetric heat capacity of the steel was calculated and an energy balance was used to calculate the heat sink temperature (as energy was released it was added to the heat sinks which would exist if one had an infinite heat transfer coefficient).
The results of tgis analysis showed the heat sgnk temperature increased by 115 F.
This was added to the 160 g initial heat sink temperature giving a maximum temperature of 275 F.
There are several conservatisms and considerations in the analysis that warrant further discussion. The results provided above assumed all the hydrogen generated (1620 pounds) was burned. The CLASIX results for this event showed only 940 pounds of hydrogen burned.
This represents a 40 percent conservat{sm in the energy released becruse of burning (9.88 x 10 Btu to 5.7 x 10 Btu). Taking credit for only the steel heat sinks is also conservative, since concrete surfaces are neglected. The results pgesented are based on a steel heat sink surface area of 50,000 ft. The total surface area of the major steel and concrepe heat sinks in the lower and dead ended compartments is 140,000 ft Curve D is a conservative and technically correct assessment of the energy distribution at Sequoyah during an S D event.
2 l
i
. hM:.:J c M i z x 5 5 ?.:.+>u-h.& % v & ad M
- Mda KGsG'.2.-
- TABLE 1 MAJOR CHARACTERISTICS OF STRUCTURAL HEAT SINKS INSIDE SEQUOYAH NUCLEAR PLANT CONTAINMENT - DEAD-ENDED COMPARTMENT Heat Thickness Transfer and Areg)
Material Structure (Ft (As Noted)
Containment Shell 3,045 7.8 mils coating 0.78 in. carbon steel 4,305 7.8 mils coating 1.1 in, carbon steel 4,305 7.8 mils coating 1.25 in carbon steel 3,780 7.8 mils coating 1 37 in. carbon steel 4,305 7.8 mils coating 1.51 in. carbon steel Crane Wall 7,255 1.6 ft. concrete 3,801 6.3 mils coating 1.58 ft. concrete Containment Floor 4,809 6.3 mils coating 2.1 ft. concrete Interior Concrete 9,870 1.1 ft. concrete 3,948 6.3 mils coating 1.1 ft. concrete 5,376 1 58 ft. concrete
1 2
TABLE 1 1
MAJOR CHARACTERISTICS OF STRUCTURAL HEAT SINKS INSIDE SEQUOYAH NUCLEAR PLANT CONTAINMENT - LOWER COMPARTMENT 5
Heat Thickness Transfer and Areg)
Material (Ft (As Noted)
Structure Operating Deck 7,507 1.1 ft. concrete 2,971 1.6 mils coating i
1.1 ft.' concrete 2,131 1.6 ft. concrete 798 6.3 mils coating 2.1 ft. concrete 2,646 2.1 ft. concrete 210 6.3 mils coating 2.1 ft. concrete o
Crane Wall 14,752 1.6 ft. concrete 4
3,570 6.3 mils coating l
1.6 ft. concrete Containment Floor 567 1.6 ft. concrete i
i 7,612 6.3 mils coating 1.6 ft. concrete Interior Concrete 3,780 1.1 ft. concrete 567 1.1 ft. concrete 2,992 2.1 ft, concrete 2,384 0.26 in stainless steel t
2.1 ft. concrete 1
2,373-2.t. ft. concrete 1,480 6.3 mils coating 2.1 ft. concrete 4
1 3
i f
i 5
, ~ -,, _, - -, +, ~,,
., ~... - -,, -, -,
.,-,,-.,....-,---,-.-,.-e.#r.,-,,
2,~~..
.~,e,m
,-4,
- ~ ~ _.-w.,.
TABLE 1 Heat Thickness Transfer and Areg)
Material (Ft (As noted)
Structure Miscellaneous Steel 12,915 7.8 mils coating 5.3 in. carbon steel 7,560 7.8 mils coating 0 78 in carbon steel 5,E50 7.8 mils coating 1.1 in. carbon steel 2,625 7.8 mils coating 1.45 in. carbon steel 1,575 7.8 mils coating 1 7 in, carbon steel
~
to nb m
s~
Ed h
~
e t
8_
8 E
/"
C y- -g -- - :. __
A 0
1000 2000~
3000 4000 TIME (SEC0!TDS)
A - Calculated Atmospheric Temperature Without Burn
- LOTIC S D case 2
- assumed as initial te=perature for heat sink calcuations (160 F)
B - Measured Atmospheric Temperature With Burn
- TMI - 2 (2l0* F)
C - Experimental Heat Sink Temperature With Burn
- Fenwal Phase 2, Part2, Test 2 (multiple burn)
- maximum igniter box internal te=perature (238 F)
D - Calcuated Heat Sink Te:perature With Burn
- TVA analysis
- maximum heat sirA temperature (275 F)
E - Calculated Atmospheric Temperature With Burn
- CLASIX S D case 2
- no structural heat sinks FIGURE B.1
~
EVALUATION OF EQUIPMENT SURVIVABILITY IN FENWAL TESTS AND AT SEQUOYAH NUCLEAR PLANT Lists of both the components that were tested at Fenwal for equipment survivability and the equipment in the test vessel required to support the combustion tests were presented in Tables 6 and 7, of Appendix B to the TVA Quarterly Progress Report on Hydrogen Combustion and Control. These lists included a description of the visible effects of their exposure to the hydrogen burns. Those components that showed any significant effects are reproduced here in Table 1 along with comments on the severity or implications of those effects. In addition, a discussion is presented of the applicability of these test results to similar key equipment currently installed in the Sequoyah containment. The list of key equip =ent at Sequoyah was taken from Tables 8 and 9 of Appendix B to the TVA Quarterly Report.
I This list of key equipment is also reproduced here as Table 2 along with a discussion of location and protection. As noted, with only three exceptions, the key components required to achieve and maintain cold shutdown following a degraded core, small-break LOCA event are protected from the environmental effects associated with such an event. The three exceptions are:
item 4, hot leg RTD cable; item 11, cold leg RTD cable; and item 10, core exit thermocouple cable. The j
hot leg RTD cable is currently encased in conduit inside the containment to within a few inches of the well in the hot leg piping.
The cold leg RTD cable is currently encased in conduit to within 20 feet of the cold leg piping. The core exit thermocouple cable is exposed "for several feet above the reactor vessel in order to allow removal of the vessel head. The insulation material for each of these cables is of superior quality construction designed for high temperature applications as detailed in Table 2.
TVA will, in its hydrogen research program, continue to evaluate these short lengths of cable for both short term and long term transient heatup due to hydrogen burning. The cabling will be encased in metal conduit or equivalent protection will be provided if shown to be necessary.
1
~
4 f
-+v+
r-e * -F e
"r**-**
- +4t*=-*
8-f"** * -
~TPe
e'M"-*?-~***----""--='7t-www
- t-wT'e'M"-
"*-r-'-y
>+-Mt-+re-ee---~~-rtv-e-F---t-*'7
'=
Y a
7d'*-"*i
'--T'W'
i^f
(
l Table 1 i
No. of Test Pelation to Sequoyah l,
M.pment Exposures Errect of Tests Comments Key Equipment (Table _2) l' 1;
Black plastic coated 1
Two scorch spots (2" by 1/2")
Surface errects only Used inside containment.
i All cables are wrapped or i
routed in conduit inside containment TVA igniter assembly 30 Assembly still functions well.
Scorching on transformer and Used inside containment Trensrormer coating scorched.
cover gasket indicates that as component or interin Transformer wires scorched.
a hydrogen burn occurred mitigation systen. How-Wrap on transrormer windings inside the igniter box. Box ever, igniter boxes scorched.
was intentionally not sealed inside containment are Glow plus connector scorched.
to see if this would occur.
complete sealed at 311 Transformer laminations corroded. Closer examination revealed openinCs and penetrations Cover garket scorched and that scorching was superri-with high temperature RTV
- hardened, cial and not functionally insulation.
[g Assembly exterior liChtly degrading.
corroded.
1
(
Duke igniter assembly 6
Cover seal burned, but no Igniter did not experience other obvious degradation.
any interral burning.
Wood block (4"x4"x5-1/2") 20 Thin brownin5 over much or Wood is not used inside wood surface.
containment.
f Thermocouple lead 30 Terlon insulation burned err Melting tempegature or Most thermocouples con-wires (first set) most or wires.
terlon is 620 F.
The tained inside containment I
diameter or the thermocouple have multiple sheathes wire and terlon sheath were and are enclosed in 0.032 and 0.058" respect-conduit (see item 10 in ively. Since the lead wires table 2 for exception).
were directly exposed to j-hydrogen burn atmosphere, the insulation could oe
- a expected to rail after
'f multiple burns.
[
Fan notor (1st)(1/150 20 Light oxidation over surface; Silver solder has a eclting There are no exposed j
ha shaded pole motor) soldered connections failed temperature in the rance or solder connections
)
on last test.
900 to 1400 F.
These inside containment
)
solder connections were
)
exposed to atmospheric 4
temperatures for some tests (i.e.,12". H ) as 2
high as 2000 F (calculated).
Fsn motor (3rd)(1/150 1
Failed after high temperature hp shaded pole motor) transient born test; soldered
(
connections detached.
f f
i 1
e
. _ _.. ~_
i TABLE 2 EVALUATION OF SURVIVABILITY OF SEQUOYAH KEY EQUIPMENT Key Equipment Location and Protection l
1.
Steam Generator, Located outside the crane wall.
Pressurizer, and Sump Transmitter cases are 1/4 inch thick steel j
Level Transmitters ~
cr cast iron. All cabling from the transmitters is run in conduit. The cases and conduit are sealed.
i 2.
Air Return Fan Motors Totally enclosed massive motor (1300 pounds). No exposed solder connection.
All control and power cables to the motor are enclosed in conduit and sealed.
3 Hydrogen Analyzers All components located in the annulus.
Components are not exposed to a burn.
4.
Hot Leg RTD's All' cables are enclosed in conduit except at the RTD well in the hot leg. Cable construction same as item 11.
j 5.
Gasket and Seals for The seals are not exposed directly to the l
Flanges,. Electrical atmosphere. The boxes or penetration I
Boxes, Air Locks, and assemblies will protect the gaskets from i
the Equipment Hatch thermal radiation from hydrogen burns.
i 6.
F cogen Igniters The igniter ' boxes are steel and have been sealed (igniter assembly used at Fenwal was not sealed). Cables to the box are i
enclosed in conduit and sealed.
7.
Electrical Penetrations Consists of a metal canister welded to the containment. Header plates are welded over the ends of the canisters.
i i
8.
Containment Isolation-The ccntainment isolation valves will be Valves Including in the required position prior to any Hydrogen Sample Valves hydrogen burn. All air supplies will be FCV-43-201, 202, 207, isolated and all relays and controls are and 208 outside containment with only power feeds to the valves (i.e., the valves cannot I
change position). The power feeds are routed in conduit.
9 Wrapped Cable All cables at the electrical penetrations are wrapped with 1/16-inch thick lead.
These runs are 'short (approximately 12 i
inches) and are all located at the containment wall outside the crane wall.
4 nr
~,&,+..,ew-e--e, p.
y,,n,
,,,em,,,,wp,-we,.v,,,m-,
w,en.,
,:e.,n,-.,,,,,_4-n.,,,mm...,
-,.g.
,,,,.,n.,w
,w,,w gy m--,,
-nw.w,,...,.,-,,,,,
lb. Exposed Cable - Core Runs of cable several feet long not in Exit Thermocouples conduit in order to allow removal of the reactor vessel head. Cable construction:
two wires 0.032 inch in diameter coated with 0.0005 inch polymide tested at 650 F.
for 1000 hours0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br />'. Each wire is wrapped with 0.06 inch silicon-impregnated fiberglass braid. The two-wire assembly is wrapped with a 0.02-inch' copper braid and a final coating of the silicon-fiberglass braid.
l The fibgeglass braid has been qualified for 900 F.
- 11. Exposed Cable Cold Leg -Twenty feet of exposed cable.
RTD's Construction of cable: 4 conductors, 22 GA, nickel-plated, silicon rubber-insulated, nickel clad copper shield, l
silicon-impregnated fiberglass wrap, stainless steel armor braid. Diameter of l.
outer sheath is 3/8 inch.
l-
- 12. Junction Boxes GA fourteen gauge steel. No solder connections are used'in any boxes.
e 4
I f
i
,----,r.,-m.
..m
,,.. -. - - ~
........_--.,-,-,,-.,-.,-_.,~~_.,._,..-_.,..-,,.___-,.-,.--,,--c.
SHORT-TERM EFFECTS OF A HYDROGEN BURN During a hydrogen burn a short duration energy release occurs. This energy release affects equipment in two ways. First, there is radiative heat transfer from the flame front as it approaches and recedes from an object. Secondly, there is conduction and radiation as the flame passes around an object.
A transmitter was chosen as a representative sample to evaluate the short term heatup effects of a hydrogen burn at Sequoyah. The transmitter was picked because of the relatively high thermal conductivity of the steel housing. The HEATING 5 computer code was used to model a housing seven inches in diameter, six inches long, and 1/4-inch thick. A temperature profile representing the flame front approaching, passing around, and receding from the transmitter was imposed as the -forcing function on the transmitter. The model was set up so that the temperature in the ~ transmitter wall was calculated at 0.05_ inch intervals to obtain a complete' temperature distribution.
The results of the analysis showed the maximum internal steel g
temperature is 265 F.
The - analysis also showed the maximum outside temperature of the housing is 407 F.
7 i
TEMPERATURE PROFILE USED TO EVALUATE I
I'.
SHORT TEMPERATURE EFFECTS
', l ON EQUIPMENT
)!
5000 0
4400 F
4000 i
if g
m E
s o
k m
3000 i
i F-4 cc i
m a.
E m
F 2000 i
4 1000 i.
~
l 300 F
0 i '
)
I I,,
t I
i l
e O
1 2
3 4
5.;
6 1
l TIME (SEC)
- i FIGURE 1 tt
.