ORNL-TM-2478.txt

 

 

 

 

ORNL-TM-2478

DESIGN, CONSTRUCTION, AND TESTING OF A TARGE MOLTEN SALT FILTER

R. B. Lindesuer and C. K. McGlothlan

 

| LEGAL NOTICE

| This report was prepared as an account of Government sponsored work, Neither the United
: States, nor the Commission, nor any person acting on behalf of the Commission: :
: "A. Makes any warranty or representation, expressed or fmplied, with respect to the accy~
racy, completeness, or usefulness of the information contained in this report, or that thé ase
“of smy information, spparatus, method, or process disclosed In this report may not infringe
- privately owned rights; or . i .
| B. Assumes any Labilities with respect to the use o, or for damages resulting from the
: use of any Information, apparatus, method, or process disclosed in this report.
. As used in the above, *‘person scting on behalf of the Commission’’ includes any em-
.| ployee or contractor of the Commiasion, or employee of such contractor, to the extent that
- | such employee or contractor of the Commission, or employee of such contractor prepares,
1 disseminates, or provides access to, any {nformation pursuant to his employment or contract
!frflfll the Commission, or his employment with such contractor. ’

" MARCH 1969

OAK RIDGE NATIONAL LABORATORY
Oak Ridge, Tennessee
- operated by
UNION CARBIDE CORPORATION
for the
U. S. ATOMIC ENERCY COMMISSION

DBIRIBUTION OF THIS DOCUMENT IS UNLT
 

 
 
 

 

"

.

- ABSTRACT

INTRODUCTION .
'DESIGN CRITERIA. . . . .
Physicel Ieyout of the System .
-Sequence of Operations. . . . . .

Queantities of Corrosion

* * * .

Pressure Drop . . .

Temperature . . . .

Pressure. .

- ® ® .

Maintenance . . . .
EXFERIMENTAL PROGRAM . .
Reduction of Corrosion Products
Selt Filtrafion Studies . . . .
FINAL DESIGN . . v v v v v e o o & &
Analysis of MSRE Fuel Cell Salt
Filter Elément Design . . .
Electric Heater Design . o .
Instrumentetion Design . . . .
FABRICATION
Procurement of Meteriels. . . .

Shop Febricetion. . .
Quelity Assurance .

* - & . L]

~ Schedule and Cost .
INSTALLATION . . . . . .

_ ACKNOWLEDGMENT . . . . .
REFERENCES . . . . .
 APFENDIX (Febrication Drawings). . .

CONTENTS

1ii

- *

. - ® - . .

. . * * . -

¢ 8 s = @

. .. s & s @
o o s e 8 @
. . e » .

. . * 2 & *

* 2 . @ - * . ®

 

Products to be Handled.

»

*

&

4

D
N

00 ONOWVI VI VT VTV NN N }n

n P NN
OOOCX)-QLI:;

N D N NV N
O 0O N & M

 
 

 
 

 

)

3}1.

DESIGN, CONSTRUCTION, AND TESTING OF A LARGE MOLTEN SALT FILTER

R. B. Lindauer and C. K. McGlothlen

ABSTRACT

The Molten Salt Reactor Experiment uses mixtures of fluoride salts
as fuel, Routine on-site processing of these molten salts results in
formation of corrosion'products. This report describes development
design, construction, installation, and testing of a large salt filter to
remove these corrosion products ‘The filter is designed to remove

approximately 15 kilograms of corrosion products from 9000 kilograms of

flush and fuel salt at a temperature of 1200°F.

 

~ INTRODUCTION

‘The fuel in the Molten=Salt Reactor Experiment (MSRE) is & molten
mixture of fluoride salts (LiF, BeFp, ZrF;, end UF,). The UF, required
for eriticality is less than one mole percent of the mixture. The MSRE,

. & forerunner of breeders operating in the thorium-233U cycle, started up

with ®35y, Sufficient 233U later became available and the experimental
program of the reactor was expanded to include operation with this fissile

material.!,® The changeover involved stripping the original UF, from the
other fluorides (carrier salt) by the fluoride volatility process, in an

on-site processing plant;3 then sdding Z33UF.-LiF as required.
~ The fluorinetion of the salt is accompanied by.formation of corrosion-

-_product fluorides which if left in the carrier salt would interfere with

the routine monitoring of corrosion during reactor operation. In principle,
the corrosion products could be removed simply by reducing them to in-
soluble metallic form, then filtering the molten salt. A problem was the
filter. Smell sintered-metal filters had been used extensively to filter
 

 

molten selts &t temperatures to 1200°F, but the design of a filter for
this high temperature and large enough to handle around 15 kg of corrosion
products in 9000 kg of selt was a different order of magnitude. This
report tells how such a filter was successfully developed and used, It
describes the concept, development tests, final design, construction,
installation, and operation.

DESIGN CRITERIA

Fhysical Layout of the System

Figure 1 shows the piping lsyout of the filter, storage and processing
tenks. The selt inlet line of the filter is sbout 6 £t higher in elevetion
than the meximum selt. level in the processing tank. From the bottom of
the filter the molten salt drains by gravity to the storage tanks, about
20 £t below in another cell,

Sequence of Operations

Two TO-ft> batches of salt were to be processed — the flush salt
and the fuel salt. Salt properties are given in Table 1. The operation
was to begin with transfer of an entire salt batch, by heiium_gas pressure
from one of the storage tanks, through the filter to the processing tank.
At this time the salt should contain essentially no solids and should
back-flow through the clean filter element with very little pressure drop.

The salt was to be sparged with gaseous fluorine to convert the UF,
to volatile UFg which leaves the salt to be collectéd on NaF. During this
fluorination corrosion of the Hastelloy-N(a) vessel would produce NiFgp,
FeFp, and CrFs, 811l of which are soluble in the.salt.' Since thesé‘

 

(a)N1,<Mo, Cr, ‘Fe (70 - 18 - 7 - 5%) specially developed alloy with
high temperature strength and corrosion resistant to molten salt,
Commercially available from Haynes Stellite as "Hastelloy-N“ and Interw
national Nickel Co. as INCO-806. _
"

L

n

OPERATING AREA

ORNL DWG 68-8661

""'/////////////

7777

 

 

 

 

 

 

 

 

 

 

 

 

 

FUEL PROCESSING CELL SALT STORAGE
| ~_RING JOINT CEL
’/‘ FLANGES 71
SALT /|
INLET.__ /
/|
)
N Y
i V]
I /
| | A
' REDUCTION | ~
w1 /\ Rou
l v - TANKS
af V
1
1
=

Fig'.rl. Ph}réibél ‘Layout of the System. =~

 
 

 

soluble fluorides would interfere with corrosion monitoring during re-
actor operation, they would havé to be removed from the salfi. The NiFo
would be reduced by hydrogen sparging to metallic nickel and the FeFy
and CrFz would be reduced with pressed zirconium metal shavings to Fe .
end Cr metal. The salt batch would then be filtered to remove these

precipitated metals before béing sent to the salt storage.tanks.l

Table 1
Properties of Fuel and Flush Salts

 

 

Fuel Salt™ Flush Salt™

Composition, mole %:

LiF 65 ‘ 66

BeFs 29.2 34

ZrF4 5 : ' . ‘ t

UF4 e 0.82 ~ 0.03 .
Average Physical Properties: @ 1200°F @ 1065°F *

Viscosity, Ib/ft-hr - 18 20 .

Density, 1b/ft3 147 126

Liquidus Temperature, °F 813 856

 

s

®after 15,000 hours of fuel salt circulation in the reactor
piping system and before fluorination of the salt.

Quantities of Corrosion Products to be Handled

Since salt of this composition had never been fluorihated_in plant-
scale egquipment the exact corrosion rate was not known, However, because
of the smaller surface]volume ratio it was expected that corrosion would
be less than the 0,5 mil/hr experienced in small scale work.

Assuming a corfosion rate of 0.1.mil/hr and & total fluorine sparge
time (flush + fuel salt) of 70 hours, it was estimated that 15 kg of O/
 

¥

#

[ 1]

corrosion prcducts as metal would be produced. One filter element should

have sufficient volume on the upstream side of the_filter media for this
welght of metal if the bulkrdensity is not lower then 50% of the solid
metal density. A iarge safety factor in the filter capacity was the

large amount of metels remeining as sediment in the feed tank during small

scale filter tests.

Pressure Drop

The filter element must withstand a pressure differential of 18 psig
tending to collapse the filter during filtration. This is 150% of the
minimum'pressure at the end of the filtration when the transfer gas
pressure releases through the loaded filter.

The filter element must also withstand & pressure dlfferential of
30 psig tending to burst the filter media during transfer to the processing
tank, This is 150% of the salt head from the bottom of the storage tenk

r‘t0~the‘filter element.

Temgerature
The filter element must have the required strength at 1200°F, which

is the maximum expected temperature during salt transfer. .
Pressure | |

The maximum,pressure on the filter housing will be 30 psig.
Maintenance

, The fllter element must be replaceable by remote maintenance in
case of plugging.
EXPERIMENTAL FPROGRAM

**The'selection'of'a-high;temperature'filter media for molten salt is
limited to materials that have high temperature strength corrosion re-
sistance in molten salt, and adequate filtration efficiency. The tempera-

) ture and corrosion requirements limit the material selection to high-nickel

'alloys such as Inconel and Haetellpy-N. Inconel was chosen because of

 
 

avelleblility. To select the type and size of filter medie-that would
" satisfactorily remove the corrosion products expected in fluorination of
the MSRE fuel eand flush salts required an experimentsl progrem.t The
experimentel progrem was composed of two parts. In the first part,
‘molten salt was prepared to simulete the conditions expected after fluori-
‘nation of the MSRE fuel salt. The second part consisted of salt-filtration
tests of two types of filter medie under conditions which epproximated
those anticipated in the reactor application. -

Reduction of Corrosion Products Fluorides |

The experimental program wes carried out with e 104k-kg bat;h of the
fluoride mixture, IiF-BeFz-ZrF, (65 - 30 - 5 mole %) which was prepared
from the component fluoride salts by routine production procedures. After
treatment to remove -oxygen and sulphur impurities, CrFp, FeFp, and NiFo
were added to the salt-mixture to simulate the conditions expeCted after
fluorination of the MSRE fuel salt. The salt mixture was hydrofluorihated
 to insure complete dissolution of the corrosion product fluorides.
Anaiysis of a filtered sample of the salt taken after dissolution.is
compared with expected concentrations in Table 2. "

The NiF2 in the selt mixture was reduced with a hydrOgen sparge and
the gas effluent periodically anelyzed for HF. Since equilibrium data
produced very low hydrogen utilizetion during reduction of FeF- and CrFo,
deta were obtained on reduction of these fluorides with pressed zirconium
metel turnings added to the salt fiixture. Analysis of filtered eamples
teken neer the conclusion of the zirconium addition period are elso shown
in Table 2.

Salt Filtration Studies

Filtration tests were conducted using 40-;; pore size sintered porous
nickel (Micro Metallics Corp.) and two grades (20 and L4l-, pore size) of
sintered fiber metal (Huyck Metals Co.). Each filter was fabricated as &

2-5/8—in. dismeter plate so that geometric surface areas of all filters
~would be identicel. :

The experimental assembly used for the filtration tests was essenti-
elly thet used for the routine production of fluoride mixtures. Tests were
 

 

 

 

 

(_i‘. 0 » H " "
Table 2
Corrosion Product Fluorides in Batch of Salt for Filter Media Test
\ | Estimated Results of Analysis of Filtered»Samples* (ppm)
: Quantity Concentration After Hydro- After Addition After Addition
Material Added (g¢) @ (ppm) fluorination of 253 g Zr of 327 g Zr
FeFz 50 286 620 60 Couy
NiF5 800 - 4680 3700

<30 - 36

 

Concentrations reported on.a metal basis,

 
 

 

made under static conditions by allowiné the melt to remain quiescent for
& minimum time of 4 hours prior to trensfer, and also made by rapldly
sparging the melt with helium just prior to salt trensfer et 650°C. The
pressure drop across the filter varied from 22,5 to 20.8 psi &s the level
of salt in the treatment vessel decreased A summery of the filtration
tests is shown in Table 3.

The fiber metal medie (designed FM-250 by the menufacturer) which head
e porosity of T8% end a stated removal efficiency of 98% for particles
larger than 10 microns in dieameter was recommended for the MSRE. Fil-
tration times for this material were &bout 1.19 and 1,36 heurs per cublc
foot of selt mixture (Runs T and 4). The occasional plugging of the
filter in Tests 4 and 5 suggests that the loading capacity of the 40-micron
filters may be &bout 50 to T5 grams of metal particles or about 9 to 14
grams per square inch of filter surface. |

Samples of the salt mixture were taken before the first filtration
experiment and downstream from the filter plate after Tests L4, 6, end T.
Apalysis of these samples are given in Table 4. Only 1.8 to 3.4% of
metals reduced from solution passed through the FV-250 fiber metal media.

It was concluded that the FM-250 fiber metal medie would filter as
well as the porous metal medis that had been used successfully to filter
small batches of the original salt loadings for the MSRE and was less
susceptible to plugging.

FINAL DESIGN

Analysis of MSRE Fuel Cell Salt Filter Pressure Vessel

Calculations were made to determine that stresses in the pressure
vessel for the filter would be within the allowable stresses for Class-C
vessels of Section III of the ASME Boiler and Pressure Vessel Code,®
Design deta are given in Table 5 and construction detalls are shown on
- Drawings E-NN-D-49036 and E-NN-D-L9037T.
 

h « <& -

Table 3
Summary of Filtration Tests

Salt Composition: LiF-BeFo-ZrFy (65-30-5 mols %)

Wt of Salt Mixtures | 10k4,1 kg
Volume of Salt at 650°C: 1.7 ££°

Indicated Pressure Differential: ll psig forepressure vs vacuum

1

 

 

| ) o Pore Transfer Weight
Test . Filter - Diameter = Salt Time Gain .
No. \ Material | Microns Conditions Hours Grams - . Remarks
1 Monel fiber \ 1 20:\‘ ~ ~Static - 22  Test terminated after 2 hrs.
- metal S o | Essentially no salt transfer.
2 Porous o Static 0.5 7 No visible material on filter or
Nickel evidence of failure.
3 Porous %o  Agitated 1.75 L
Nickel ' . _ :
L 347 88 fiber L1 ‘Static 2.0 7 Test stopped after 90 kg transfer.
metal - '
5 " Porous | Lo ‘Static 2,17 63 Filter plugged after L0 kg transfer,
Nickel S - : Material on filter predominantly :
L e - ' salt,
6 Porous ko ~ static . 1.84 19 Balance of salt transferred
- Nickel filter ruptured.,
T 347 SS fiber 41 Agitated = 2,0 67
- metal |
8  Porous - 40 - Static 1.75 22 80 kg back transfer of salt from re-
Nickel ‘ celver to treatment vessel. Filter

plugged.

 

 

 
 

 

10

Table L

‘Summggy of Analytical Results
During Filtration Tests

 

 

 

 

Sample Intervel Impurity Concentration (ppm)
Filtered Sample - Cr Ni Fe Total
Before Test 1 26 36 LL 106
After Test L 15 84 66 165
After Test 6 16 256 132 Lok

After Test F | 17 19 Lo 85

 

»
Teble 5
Design Data for Iarge Salt Filter

*¥ n

 

Construction Material

Pneumatic Pressure Test |
;Pressure Vessel without Filter Element, psig
- Filter Element Inner Core Cen, psig

Helium.vacuum Leak Rate to Inside
- Pressure Vessel (without filter element), cc/sec of helium
~Filter Element Inner Core Can, cc/sec of helium

Design Temperature
. Pressure Vessel
Filter Element

Design Pressure, psig

Operating Temperature,‘°F‘Msximum

vOperating Pressure, psig Maximum
FILTER PRESSURE VESSEL

Size

Length

Salt Inlet and Outlet Size
Access for Filter Element

Flange Seal

FILTER EBEMENT

Filter Media
Porosity, %
. Mean Pore Size (Microns)
Filteration Rating when Flltering Liquids

Inconel 600

625.

100

<l x 1078
< x 10-8

1200
1200

35
1200

35

6 inch, Sch. 40, pipe

T feet ~10 inches

1/2 inch, Sch. 40, pipe

6 inch, 3oo 1b, ring Joint
flange (special)

R-45, O-king, Copper, Leak~
detected (special)

M-250 Feltmetal
8L

29
10

 

‘Nominal (98%), microns

1T

 
 

Table 5

(continued)

 

FILTER ELEMENT (continued)

Tensile Strength @ 100°F, psi
@ 1200°F, psi

Modulus of Elasticity @ 1200°F, psi
Thickness, inch

Pressure Drop for Clean Water
@ 10 gpm/f+® in psi
@ 100 gpm/ft® in psi

Total Filtering Areas, f£tZ

Filter Media=-Perforated Metal Support
Thickness, inch
Open Area, %

Quantity of Salt to be Filter, kg

 
 

3

)

¥

at

N

13

Stresses in the vessel are due to the 35-psig internal pressure and

‘the axial temperature gradient along the vessel from the heated section

at 1200°F through & 25-in, insulate:d section and & T-1/2—in. bare metal
section at 200°F on the end, The axisl temperature gradient causes dis-
continuity stresses because of the resulting differential radial expansion
of the vessel. | |

The temperature gradient was determined through the use of the "Astra
Heating" computer program st approximately 1l+in., increments for the
32.5-in, on length outside the heated zone. The maximum temperature
differential of 56°F per linear inch is near the heated zone at the |
highest temperature and lowest elloweble stress, There is no tempera-
ture gradient across the wall of the pipe so there are no radial thermal
stresses, ' '

The sum of the stresses due to pressure and the temperature gradient

‘are less than three times the Code allowable stress for primary stresses,

so the design is therefore satisfactory.
Filter Element Design

| In small scalerbatchroperations, sintered porous nickel filters
have been used successfully as reported earlier. However, it was decided
to investigate the use of fiber metal filter media for possible improve-
ment in capacity. Although experimental tests did not show any signifi-
cant improvement in performance, it was decided to use the fiber metal
because of 1ts adaptebility end availability to fabrication requirements.

(A cylindrical porous filter would have to be fabricated by the mamu-

facturer.) Inconel was chosen as the material of construction not only

e,because of its strength at elevated temperature, but also because of its

:corrosion resistance to molten salt

 

* O S Ll :
Generalized Heat Conduction Code written for the IBM-6090 Computer

- by ASTRA, INC. under contract to the Atomic Power Development Associates
- of Detroit, Michigan, (Fermi Reactor); reported as The Heating Program,
- ‘Heat Engineering and Transfer in Nine Geometries, R. R. Liguori and

'J. W, Stephenson, ASTRA, INC., Raleigh, North Carolina, Jamary 1, 1961,
 

 

 

1k

Both the burst (from internel pressure) and the collapse strength
of the inner and outer filter elements were considered in the design of
the filter element. Design data are given in Table 5, The burst
strength is important only during back. flow when transferring selt to the
processing tank, The formula |

: 28t
| P - dila

 

was used in the calculation where s is the tensile-strength of the fiber
‘metal at 1200°F.

x 8 (fiber metal @ 100°) = %%nga-x 800 = 571 psi
2

s (Inconel @ 1200°)
s (Inconel @ 100°)

The outer filter element was calculated to have & burst strength of 25 psi
and the inner filter 34 psi with no safety factor. The worst conceivable
case would be 1f the filter element were almost completely restricted from
the flush salt filtration and the fuel salt wes then transferred to the
processing tank. - Since the burst strength of the outer element is less
than the 30 psig specified in the design criteris, therbuoyancy of the
filter in the salt on transfer to the processing tank is necessafy for
 the filter to meet the burst strength requirement. The filter element
has & weight of 57 1lbs and a horizontal area of 25 in.=%, Construction
detalls are shown on Drawing E-NN-D-49038 and Figures 2 and 3. A pressnre
of only 2.3 psi will therefore cause flotation of the element and by-
passing of salt through the seat. This pressure is further reduced by
the buoyancy of the element so there is no danger of ruptnre of the ele-
ment if the filter is above the melting point of the salt. |
| To pfevenf collepse of the filter element it was necessary to provide
a perforated back-up plate against the inner filter surface.  The maximum
external pressure is applied to the filter at the end of filtration when
the transfer gas pressure blows into the gas space above the filter ele-
ment. At this time there will also be the maximum restriction of the
filter from collected solids. In calcuiating the collapse strength, the

3
formula? ps = 2Bt was used where E is the modulus of elasticity,

(1-m®)D>
+t and D are the thickness and diameter of the element and m is Poisson's
 

O
o
&
0O
=
O
X
O

 

Filter Element, Top and Side View.

Fig. 2.
 

 

 

 

 

 

 

 

 

 

 

 

 

 

' - Fig. 3. Filter Element, Bottom View.

 

 
 

 

.-

ot

»

17

~ratio (0.3). E for the: fiber metal 1s 1.75%: ‘of s011d Inconel and is

3.5 x 105 psi @ 1200°F., Considering the fiber metal without the backup,
the outer element has a collapse strengthrof‘only 6.5 psi and the inner
element 16.3 psi without a safety factor. .Since the minimum final trans-.
fer pressure is 12 psi for the fuel salt (without any flow AP), it is
evident that & backup is required. An 0.078-in. thick Inconel perforated
sheet with 32% open ares was used as a baekup because of availability.

This plate alone has a collapse strength (for the larger element) of
131 psi. To provide & safety factor of U4 the operating procedire speci-

fies & maximum trafisfer plus'gas space pressure of 35 psi.
The filter element can move upward during salt back-flowing and
by-pass salt through the seat, but it has to reseal after flotation to

" minimize. leakage in the main direction of salt flow. Mating spherical

seating surfaces are provided between the filter element and pressure

vessel for ease in self-sealing.
Electrlc Heater Design

" Electric heaters are provided to permit preheating of the filter and

‘to maintain the temperature of the salt during a transfer. Maintenance

requirements are minimized by designing the heaters with excess ceapacity

- 80 they can be operated at reduced voltage to promote longer 1life. Also

duplicate spare heaters are installed and connected ready for use with
only minor, out-of-cell wiring changes. The tubular heaters are 0.315-in.
OD Inconel sheath with nichrome elements rated 500 watts per_foot of

‘heater. A layer of stainless Steel'shimistock is installed between the
heaters and the thermal insulation'to prevent the heaters'from being in
_ direct comtact with the insulation, |

" The controls that: are used limit the available installed capacity
to 3500 watts; 2koo watts in the lower control (FSF-1) and 1000 watts in

- the upper control (FSF 2) (See Drewing E-NN-D-49036.) Control of the
'electrical input to the heaters is from manuel powerstats. The voltage
 setting required was determined during startup testing, ‘and manual stops
- -were set to limit the controls to hold the temperature to 1300°F or below,

"During;startup testing, 1600 watts for the lower section and 500 watts for
 

18

the upper section were required to preheat the filter. Only the heaters
on the lower section are required for normal transfer of salt. If salt
freezes 1in the upper section of the filter because of unforeseen diffi—
culties, heaters on the upper control will be used to preheat this

section.
Tnstrumentation Deslgn

A helium supply is provided and instrumented to assure the presence
of & gas cushion in the filter at all times and to purge the connecting
1ines-with helium before filtration. Tempersture instruments are pro-
vided both to indicate that the equipment is above the salt melting point
and as &an indication of salt level, | ,

~Figure 4 shows the helium supply to the filter. The solenoid valve
is interlocked with the salt freeze valves to prevent the accidental
pressuring of molten salt from a salt storage tank to the reactor by
helium-through the filter., Accidental filling of the reactor is further
prevented by the pressure control valve limiting the pressure to 15 psig —
sufficient to £ill the reactor only 1/2 full — and the flow restrictor
fwhich would limit the salt transfer rate to 5 liters per minute — only
'30% of the normal reactor fill rate. The check valves prevent back-flow
of radioactive gases to the operating area., PI-B indicates the pressure
in the gas space above the filter element. This pressure is transmitted
to the operating area by a transmitter contained in & cubicle vented to
the processing cell. _

Temperatures are monitored by eleven thermocouples attached to the
outer wall of the filter housing. Nine of these are read out on &

O - 2000°F recorder. The other two (at the same elevation as Points T

and 9) are connected to temperature switches which actuate annunciators

Bl

afid provide interlock contacts in the pressurization and vent valve
control circuits. Since the upper heater will be normally off, tempere-
ture points 5, 6, 7, 8, and 9 will be heated only by conduction unless
the salt level rises higher thanrnormal. If this occurs there will be an

- alaym and the pressurization valve will close when point T reaches 1000°F.

' When point 9 reaches 350°F there will be an alrm, the pressurization valve Qsj,

%
 

 

19

ORNL DWG. .68-8643

  
     

50 PSIG HELIUM SUPPLY

[ S}

. PRESSURE CONTROL
VALVE

15 PSIG

| X—@ SOLENOID VALVE

—'|

 

 

FLOW
RESTRICTOR

<« CONTAINMENT
CUBICLE

CHECK VALVES

 

 

 

-l

PROCESSING CELL

 

 

 

 

 

 

»

 

y | - S Fig. 4, .H.e'.l;ium.'St_ipply"I'r.lstrumentation.. _
 

 

20

will close and the selt-tank vent valve will open. Rise in salt level
could be caused by & restricted filter element, lesking ring-joint flange

and/or lesking check valves,
FABRICATION

Procurement of Méférialslr

All selt-containing parts.are_fabnicated from Inconel-600 materiel.
Pipe, fitting, and plate were puiéhasea.from vendors &ccording to ap-
" pliceble Americen Society for Testing'and Materials Specifications8
with certified chemicel and physical properties. After receipt in ORNL,
the meteriel received addition liquid penetrant® end ultrasonic inspection,l®
This inspection is routinely given material used to fabricate MSRE equip-
ment that will contain high-temperature radiosctive gases and fluids.
This additional inspection is considered partial insurance against equip-
ment feilure and the need for expensive and time-consuming remote

4

maintenance, , |

The filter media (sintered Inconel fibers) wes pnrchasedfin x ' .
20-inch sheets from Huyck Metals Company. This special material was
selected and purchased after nmumerous consultations with the vendor about

its unique design properties.
Shop Fabrication

The filter housing and element were fabricated in Oak Ridge Nationel
Laboratory shops from Drawings E-NN-D-49036, E-NN-D-49037, and E-NN-D-49038.
High quality nuclear fabricetion specifications and techniques that met
~or exceeded the requirements for Class "C" vessels of Section III of the
ASME "Pressure Vessel Code" were used to fabricate the filter. Forming
snd welding of the fibrous Inconel filter-media metalll required the de-
velopment'df'special fabrication techniques. Special procedures were &lso
utilized to prevent the filter-media from becoming contaminated with *
foreign meterial during fabrication, |

Weldsl! of all Inconel pressure-containing parts were liquid-penetrent C;j“
inspected® and radiOgraphed.12 Other Inconel welds received liquid-penetrant -
 

 

"

21

inspection. Welds of the fibrous metal filter-media were visually in-
spected. All welding was performed by welders qualified in the weld pro-
cedures specified on the drawings.

A pneumatic test was performed on the pressure vessel, without the

filter element, as specified in ASME Code., The inner core can of the

filter element was pneumatically tested at 100 psig to insure against
failure at operating pressure and temperature.

After pneumatic tests, the pressure vessel and inner core can of the

filter element were separately lesk-tested by evacuating the inside and

flooding the outside with helium. Heliumqleakage to the inside was less
then 1 x 108 cc/sec of both subassemblies,

- A polydispersed aerosol, dioctylphthalate-(DOP) was forced through
the fibrous metal filter element at various stages of fabrication and
monitored to determine if any large cracks had developed during fabri-

cation.
Quality Assurance

The salt filter is designed to operate at 1200°F temperature and

35-psig pressure and contain highly radicactive fluids and gases. Since

remote maintenance of radicactive equipment is difficult and time-consuming,
every effort, within practical limits, was made to obtain the necessary
quality level to minimize maintenance on the filter during the life of
the experiment. |

Once the adequate quality level was established necessary material

~requirements,.fabrication,specifications, and test procedures were se-

lected end placed on fabrication drawings.-rThese drawings, special in-
structions, end inspected material were delivered to the shop for fabri-

. cation, Discussions between;sbop_management,!inspectors,tand MSRE

, _representatives nere arrangedtto diScuss'the quality of fabrication de-
.. sired, possible prdblem areas, and. completion schedule._ This was done to
."“reduce the possibility of misunderstanding and the possible reduction in

uality and/or additional cost,and additional time to rebuild the filter.
One of the most imp0rtant factors for good quality assurance is the

establishment of & strong, working relationship between project management,
 

 

22

shop management, and inspection personnel to make sure the requirements

‘on the drawings are followed and the results reported,
Schedule and Cost |

~ Conceptusl design and development tests began in lste November 1967..
They progressed concurrently until completion in late Jemery 1968, Pro-
curement began in mid-Jafiuary 1968 and meteriels were received and in-

_ spected.in time to begin fabrication in late February. Febrication of

'one_filter assembly and one spare filter element was completed and tested

by late March 1968. ' |
Fabricetion costs of one filter housing and two filter elements con-

sists of $4800 for meterial, $1300 for material inspection, and $5200

for shop labor. - |

INSTALTATION

The filter assembly was installed in the fuel-processing system
~after all other piping was completed (see Fig. 5). Since the fuel-
préceésing cell is comparatively smell and crowded, some difficulty wes
~encountered in locating thevasseMbly. One of the prime requirements in
locating the filter was to provide direct access from above so that filter
“elements could be replaced by remote maintenance techniques. Close co-
ordination between the installers and the remote maintenahce'group in
locating the filter assembly resulted in the filter assembly being in-
stalled in & location which 1s accessible for element replacement with
minimum 4ifficulty. ‘ _

Welding caused some problems since part of the piping system was
contaminated with small quantities of non-radiocactive salt from & previous
test run. This salt hed to be completely removed;from_the_immediate weld
area to prevent contemination of the connecting welds. Valuable assistance
~was given by the welding inspection group on the best methods to follow in
cleaning and welding in this contaminafedvpiping system. As & result
of close cooperetion, all welding passed.the necessary inspection and

test specifications.

)
 

 

[ %

 

 

i
~
-
~

 
 

 

2L

PIANT PERFORMANCE

On May 6, 1968 an'6pportufiity was provided for detefmining the
temperature distribution prior to radioactive operation. To compensate
for past and anticipated removels of salt from the reactor fuel system,
3-1/2 £t3 of clean carrier salt (67 mol % LiF, 33% BeF2) was passed
through the filter. Figure 6 shows the temperature distribution over the
upper half (gas space) of the filter. Also éhown are the temperatures
obtained later with the flush and fuel salt filtrations which correspond
closely; From these temperatures, the temperature switches at points T
and 9 were set to alarm and stop salt transfer at 1090°F-and 350°F
respectively. The temperature 2 inches below the ring joint (Pt. 9)
was only 2505F indicating that the uninsulated flange joint was probably
below 200°F. The remote leask detection on the flange indicated no leakage
during the heat-up, transfer, or subsequent cool-down. ) /

During August 1968, both flush and fuel salt batches were tfansferred
to the processing tank through the filter, fluorinated, the structural
metal fluorides reduced and the salt filtered'as it was returned to the
reactor drain tanks. In each of these operations there was no detectable
pressure drop across the filter and the filtration wes accomblished in
about two hours. The amount of metals removed from the salt is shown in
Table 6. The total of both batches is 10 kg. Since visual inspection of
the processing tank after_filtraticn was not possible it is not known
how much of the reduced metals remained in the tank and hofl'much is on
the filter element, |
 

 

.

g

TEMPERATURE - Of

1200

ORNL DWG, 68-8662

 

 

1000

 

900 —

 

 

¥ SALT TRANSFER
AUTOMATICALLY
STOPPED IF

 

TEMPERATURES
POINTS

 

700

600

500.

400 ‘

REACH THESE

 

 

 

 

 

300

100

- TOP OF

INSULATION —— o]

 

 

R-J FLANGE

 

 

 

 

 

 

 

 

 

 

 

10

18

s 20 25 0 35
DISTANCE ABOVE SALT INLET, INCHES

Fig. 6. Salt Filter Temperatures.
 

 

26

Table 6

Radioactive Salt Filtration

Chenges in Concentration

 

 

 

 

 

 

ppm

Ni Fe Cr Total

 Flush Salt -

Before Reduction o 516 - 210 133 859
After Reduction and Filtration 26 1k 76 243
- Removed by Filter 490 69 5T 616

Fuel Salt
Before Reduction 8o koo 420 1660
After Reduction and Filtration _60 110 3L 204
Removed by Filter 780 290 386 1456
Weights Removed, Grams
Ni Fe Cr Total
From Flush Salt 2110 297 245 2652
From Fuel Salt | 3900 1hk50 1930

7280

 

%

Y
 

27

ACKNOWLEDGMENT

We gratefully acknowledge the contributions-of_the following Reactor

Division personnel:

C. W. Collins for assistance in preparing the section on
Pressure Vessel Design.

P. N. Haubenreich for assistance in preparing the Intro-
duction and report content.

T. L. Hudson for assistance in preparing the section on
Electric Hbating Design

B. H. Webster for assistance in preparing the section on
Installation,

J. H. Shaffer and L. E, McNeese of Reactor Chemistry
Division for performing and reporting the developmental work.
Informaetion contained in their CF Report has been condensed and

" included in this report.
 

 

10.

11.

 

28

REFERENCES

J. R. Engel, MSRE Design and Operations Report, Part XI-A, Test
Program for =>>U Operation, USAEC Report ORNL-TM-230k, Osk Ridge
Nationel Laboratory, September 1968.

P. N. Heubenreich et al., MSRE Design and Operations Report Part V-A,
Safety Analysis of Operation with U, USAEC Report ORNL-TM-2111,
Oak Ridge National Ieboratory, February 1968.

R. B. Lindauer, MSRE Design and Opereations Report, Pert VII, Fuel
Handling and Processing Plant, USAEC Reports ORNL-TM-90T7 and 90T
Revised, Oak Ridge Nationel Ieboretory, May 1965 and December 1967.

J. H. Schaffer and L, E. McNeese, Removal of Ni, Fe, and Cr Fluorides
from Simulated MSRE Fuel Carrier Salt, ORNL-CF-68-1- 41, April 16
1968 (for internel distribution only). ,

- Nuclear Vessels, Section III, ASME Boller and Pressure Vessel Code,

American Society of Mechanicel Engineers, New York, 1963.

V. M. Faires, Design of Machine Elements, MaoMillan, Néw York,
1948, p. 91.

Ibid, p. 463,

Nonferrous Metel Specifications, Pert 2, ASTM Standards, American
Society for Testing Materisls, Philadelphia, Pa., 1961.

Tentative Methods for Liquid-Pentrant Inspection, MET-NDT-L4, Inspection
Specifications, Metals and Ceramics Division, Oak Ridge National Iabora-
tory, Oak Ridge, Tennessee, August 8, 1963 (for internsl distribution

only).

Tentative Methods for Ultrasonic Inspection of Metal Plate and Sheet,
MET-NDT-1; Metal Rod and Bar, MET-NDT-2; Metal Pipe and Tubing,
MET~NDT- 3, Inspection Specifications, Metals and Ceramics Division,
Osk Ridge National Iaboratory, Osk Ridge, Tennessee, 1962 and 1963
(for internal distribution only).

Procedure Specifications for Direct Current Tnert Arc Welding of Inconel.

Pipe and Plate, PS~1, Inconel Tubes, PS-2, Inconel to Stell, PS-k46,

- Ok Ridge National Laboratory, Oak Ridge, Tennessee (for internal

distribution only).

Tentative Method of Controlling Quality of Radiographite Testing,
MET-NDT-5, Inspection Specifications, Metals and Ceramics Division,
Oak Ridge National Laboratory, Osk Ridge, Tennessee, 1962 (for
internal distribution only).

rp
 

APPENDIX

 Fabrication Drawings ~— Salt Filter

E-NN-D-49036 — Assembly
E-NN-D-49037 — Details |
E-NN-D-49038 — Filter Element Subassembly — Details
 

 

 

  

 

 
 
 
 
 

31

 

e

o

 

 

 

 

 

 

  

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
    
    
  
  
    
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
  
  
 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

’ I DUFPORT _ RING. How. ¥ s xhr < et co PARTS LIST
. F Bk 4o Pet- 12k 0 s | PART | DWG NO. REQD DESCRIP] ’
. 't Ao Boo 8'd L. Bouhe BRoM WOM] & aTL NN | STOCK $17E | MATERIAL
¢ 2% 8 %% S e B | |09y HOLD - Bowh  ASSEMBLY =
v MUY & MeLT_ Ramuuc w18 1a — —_a £ ‘ £-800"_ 8.4, .BLNG FlLaNOE:
ir 1aixl <. sia ” € Wi 2 ioimson &ie. i LGSTR E e
. t i 2 € Bk " = STRuCT, WL - L WAL BUARD ; Ol THE |1
N 8480 Lr=g D LNIY_CiNCI4 ANCWOR . K% AT - r ST e Sinl e
! i THERMOGOURL . mv_leC. [ -4 i BNE ST . Avihs 313
W by || ) LaTen as 4A4 100G T of-ossond weve 7 | % (2800 B.J. WM. FAN
! H 1 2 : | LRATLR oL 4ALZG ; | NoTE & ) 3 T G A 40 mime . <o} La.
1 | VAT Cexionma 3 | — 4 F LOVER CONTAINNENT Tl ARy —_—
- "y o TR ATION - CARIVTANF o' b S mral Tan__ NoTh 1o gt £ 2CH 4D MR L:1 1o
. | / < i F TNGULATION - CABIYTANP 100" k& TOTAL THE, NoTe 9 ‘: VOLD Wiha Shyet x sjir x B Tuk
1- | " — ] LENIRR CAR  Buofa: ARy . —
i . - ' &1 ©BCH 4D M. W. PR CAP
| 44— i *2 PATHE QLD Wy Wy u b, Tow )
|y | 6 T ADamTe R 1%, DA X 1y L )
_‘ | ; 7 i ADAPTE R Sown 1] w
. “a~] & P-RING  R-48 G- 3OO" "
- ik A [T
T nfi%“‘-‘ | _,.1‘.._‘ 9 Reomu-ie] -Ja BAMOTE MeLT 3-8 FART He. 4-40T76-% %fi@
' A 0 kione-w 19 Uhe PART Mo, B4OTIL - W o] 3
i n-4e0t% FATRR AWMLY T o —ve—
! : -
Ssgrion TAAT %) ‘ = ‘\ '
SBCTION A 0 I -
u A
!
i /
ek e .,
ih T -
5h Twd Devd h aci 40 fo' Rw. BlRow ok et
1 T Owolm me acu 4D PIPE . . ;¢
% I |y BLS Dk .0l W Tuhlr W
ot man| koo « 080 w TURA LG, ™1
Ineewnocoupye |- sacy do Fime &4 Lo < W,
W o < & 1 T | GAM AL Bw PEE cAW c '
‘—om T DOy l ! ATRAR Bh K LU TUR £nty, |
L Ho o\Fy Procibuliy | | shxexy The S wriJ
L _TWO_G RWRL Syl | —
TDATED Wit i [
P
|
1
{
i

 

 

 

 

-5
I

Qi)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[

 

rn

-

 

 

S

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
 

 

 

 

 

 

 
 

 

 

 

 

 

 

. Ve ~ ! .
- ,,_._....-m-.-..,_‘_»_i‘-._“ ]

et S B N N o Ty

 

 

 

   

 

 

 

   

e . raan s e

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. CEEEAL AN ROBIAL

So UREMMOIE LN CIIDE Y BOBIND (VENEOW PTMSW
AN}, LAAK 90 1B DaTOR GMALL A Lo T

ey W L'z 35 anfeee oF TRESL -

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6. W0 VENER. 39 SRSNNEN T - o
CTAS O VENELS oF SWTIG 1L OF TR Adsy “TREASE
VESENL g, = :

7. v ». SYIMALL BN 18,

B* « W00 ¥ - RIW ~ 4K hhaoe.

. . (B RNLL B * -
su---u-ul.l-.. . . u

S FREEAVIEN TR OF 5 2% OMLNE FIFDR PUNLATIN

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. REFERENCE DRAWNGS aeen
; 03¢ Bt Krrwmnt Losonsrony
“d+'l—4! - OPERETRD Y
Thelicats Crtex Creanee Covronation
) L

 
   

 

 

 

  

 

 

 

 

T

 

 

 
       

 

 

 

 

 

 

320 SN \ . .H.“ +
! Tt was o
BOMMRCHL PSR, < -:-m | S—E— MOWAS & -S| Emtit-e8 n.w yw
™ moos noY %0 08 VP OB ORieth min st |R MACHINED SURFACE sy - » 259" 0y | 1048 .
LA} WPON BB DF THE PORSNSbG © SMALL NOT EXCEED: - -
o (ASA S, L 1952) ok b ont xe srrwons !

 

 

 

 

 

 

 

 
 

 

 

 

. ) ] I ' PALTS LIST :

s . { PART [ owG No. [ REQD | DESCRIPTION ] $10CK SZE | MATERIAL
; frass —a - § LasTare  nouany .
. . WALLG AR BUGWN |l----—-—-—0L LD AR ——————————

ity M1 . ! .
Y/ |
£ 7o \ il
= R
|

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

. 0
) & b *
i : \‘ X — r \\
5 - TR q - & B jown, et
I 1 N T et e
v . +
/ 4 \\ : . /
’ ' N, /
\ . 14
"/’ ;
P, ?‘ .
I» .
A 1
y .
' ® Loves Couramusur
- BCALW v o
el NOTE: AFTAR WALDING § RAPORA HALLIAING  SPURBICAS
L -___/ RADWUE TTRAM RALIAVE AT 1L00'F FOR 1 RS N
- e o g st - -‘“_

ot s

MIGON ATHOWPHLRE , RUENACYK GCOOL.

BECTION 'C-C' (£45036)
PULL G B .

B (s-as0me) -

   

 

fon om

 

 

 

 

 

 

 

 

 

 

1
i+
{

3
|
A

 

i

ioa ok Tuey 1 uoLae
SO ABPART - %O ~»
%40 TAPPAR WHOLL

 

: BOTTON DR { TP oA e WO ————vey
oA LW el o o ¥” ‘{
T HOLAS \Bet ADARY . WM on RAD-

   
  

.*i-'%:/

 

e f imnm

rma
L .

 

 

 

 

i THE AVERAGE PiToW DA M INDICATRD
HAABURINE (%) DIA'S $0° APART., SWALL BR WVITWIN

 

 

 

 

 

 

 

e ol T.000] OF THR HNOMINAL DA, :
4 HOTE 1 Tue AVERAGE PITEH DiA. M INDICATRED . 2 NEFERENCE DANWINGS | NUMBER
- WY WLABIRING (3} DIA'S GO° APABT WHALL B @ . —) 1 1 £ax Riset Manienad Lasasasent
- S WAHIN T 0L 0f Tul WoM. Dik. § SHALL R . wCart | 1§t 60 e =7 [t : AT &
) CONCEMTRIC AROUT Tulh & WITHIN .GaB T.\.&. =t - - -
©® ' ' ' = v Yuien Cansiat Caavsnitina
*uLL. SCALN

 

 

 

 

 

 

 

.‘ i -y . .
- WTERNAL FADAICATION § EXTENNAL PABMICATION QA% 1008, VOUEIEY .
®©_%%W&m.fi__‘?fl.z_ T e [ — -
' m:_ 1 DCN BAlS T Ty [
VNLENS OTHERAME SPECINER * o . o
L SACAR ALL S EDGES | PRACTIONS & il
8 MACHINED SURF wan o Qb
N

= - i s o BRE 1D LER (T 68
g
NOT DCEXDX DL m~flqgn; o

i
El
T

I
i
!

innsaarh R
jr aw
e

 

it
i

 

 

Tug. Proctssing. Cli
s Line 110 _Sacy Tuges DS

3

 

'I!
Ff
|

 

 

 

 

 

 

 

 

 

 

 

 

 

0k ?,7' 3 " [pavsieimanr| T
-Wuu—unu o il gl . SN D-49037
- S MPO794 RNOS

 

 

¥

 

M

™= g nF
—c

o —

   

33

 

 

 

 

 

 

.

 

 

R

%’e
:
d
§ ;2
: o~ )
1 < it 3
@.- 2% N
I \
@_m \
¥ @1 N
T O
!
|
1l

i | '
s by N )

T T R
ey

R e

?

R ST Y Rt e ey

 

 

 

 

o e N

_7%®

(e

 

 

 

 

 

i

 

 

 

L EiLTeE R Assrms i
NOTR © WALD ‘B INDICATE BUGOHEDITRED WEDING SLQULNECK

 

 
      
      
 

i‘lfllfi. L .. ‘ .
AT TNy fi'fim
b A JUOM. Lo (PART WD} d-

 

) vom. 10, m——j

 

8 (PART 1N )——

 

/%//m-—'

1\

i
d

 

 

 

 

 

 

b s f (a7 A}
“

 

 

@ @ funee Gy, s @Q_HLELQI.L
WO SCALG ne BCaLe .
MOTE : ENDS OF PARTE I-¥,1kiD, Wil 7o BA SLS T WVITH MATING
PARTS 4% BLOWN ON ANWARLMPAY. '

‘——‘i uon-\o ugmfi I——-&fh How 1D SRk MNOTE ——n

 

 

@ L\E"!\NG ;l\h

 

 

20° (PARTY tg)—— ——e e

PESFORATED & o
- I
SEITMETAL

 

 

— PERFORATRD .
SREUTHET h\j

! __'/;'—//7
L=

Bl

1y

4

.z
o mom =T (PARE AP
&

 

 

 

 

 

 

 

 

. .__)/"

84 (Auer 1) o]

20" (PART IFd - ——]

 

 

 

 

Q)CDS\.&%L.E.LLE 2 (Z:)(EB O&TEE FiLTER
No .

NOTR: EMOD OF PANTE b, W, {126 , kT To BEL SUP BT WITH MATING
PARET AB BUOWN ON AsSeEmMbLY. v |

 

SEAM VRLDS OF SATMETAL ( PLEFCRATLE FLATE 7O BA LOCATLD iBa’ |
APARY, ROLL ALL RCOA% OF RALETHMETAL To & TUWCKNAYY oF 080 GOR V".J
|

]

T I

ULl SCALR

—tan0: 1940 SIA

 

  

Lo THER

PARTE LISY .
PART | DWG NG. | 50D DESCRIPTION T STOCK SIZTE | MATERAL
1 pusowe|j 2 | FuTeR  Assammy
1t || upTiNGg  maaL Bou x TLe AVTI )

[T} I | vo® crosues maTe G D x 1 ALTH & 168

[ 1 | curer " eviLinpre s, 0ia x 4l s?"r.‘.“i._!'

Tt t | SUTER RILTER, cownl 4810 X 20 SRk & o)

e | | outen  miLTRE. bcovel 4% LO. X 8 5e . NoTICRE & ST
e t [ inam  FILTER , INCOMEL 8300 X 20 et Pra e .o o

"7 i |[inwes  piitem,meonnl  8%o.x 72 TRV ' %

. T | wib mina 4% o x by et
i 1L v, mike thow x*h S e
1140 I §INNER CYL. 410 x48' é-?-m"bu:uhfi
[ 17 INNERE CvL, 2R Lo x 48 STH.
it I | cLosumE  PLATE t o x Yo S D Eh]
118 I | cLosure wmino TN T
"t 1 CLOBUR.E  RiNG 4 o x%a ASTH B ILe
[ irie ¢ | vas meo Bom x 1l INCONEL
(‘A e ié 1 Yoo X.006 wALL TR o L. l:l'::Azhfl.
e MOTER § PARTY  SLOWN AIIJ.FO- 1) AMBENELY . i
4
T uoles O
or sROW
: SUREACE A’

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
     

T f A TURL ST
WOLES ON 4991 TIA.

.
-
Lok iR

UOLES ON 360 DiA.

 

 

 

 

 

 

5.957- 5-927 DA £9. Boyyg ey
L9317 Oia. . .-_.__q'n._r_'.
— B.A%0 DA !
{ITREN . — n&j . f o RAD.

 

>
7
-

]

 

   

 

BIL - 2-801 Bk~

-

  

 

| @CLG&HB; .
FUL SCALE

r——-——t-fl" - 4. 82T O ——
e 4, 78 DUA — ———-l'

 

 

=

 

Lol —-|-—‘,vu GAP

 

 

 

Rz

——

 

w——— 4} o ————

my,g,hg__z.l.u_@_

 

e — 80 = .M DA,

&\.a?ugg Bing

5.0t

p— 1% - 5140 DR ——=

 

 

 

 

 

 

~8
dog
4
A—q— e L A T
Qk D ey

5. %06

e Do
-2 B“\:————-——-lj

e 4000 - 5990 DiA.

 

 

 

 

 

 

 

 

 

 

 

-
\ sl ¥ vl - r §- e
jee-215%-3 ey - - e e -
fra-i L ¥ ; - z ey
> P, ey PR WrSRATION,
m BRANGR.
STDWAL FAbcanion | EXTORMAL anicaTion § 1 o ao
& FPECTION "

 

. _L ; ~ ‘u———————; LYY M—-———-I !
T"‘L[ P

@W

VT - - —% -l_l

 

3! AFTER WELD

e

 

 

R\ P

N

 

 

; pe—dOGL = 458 Cib—-—
B e 4, 578 = 4,868 DIA. —~ -
| —tan- o i
i ket BT DA— e — *Ii
!
ke g L~ 802 DR e
@ Tbe > e
N o - = =

  
  

   

| -
21-788 - 2.-15»6 Duu—fiL
[T Wt = 2. B0 DHA— ;
§ —te1m-2.008 QIA.——J
t—L—_—J.m‘l = 5.417 DA
3.150 - 5.340 DIA.

[ --—4.000 « 3.906 DA ————]

 

 
  

 

 

 

 

 

| : Pumsonw

ARALFRRLY uRPHG FOLYORMPARSE MROSAOL .

R
§ &
PROTACTEDR

¢ b
A% BER

UHITS

4 PNREUNMATIC TOST INHRR CYL.{ PAETS (2,101 § el ) WiTW WITROGRN @
HOLD #OR 1 HouRe

o0 Peia ¢

WLELII W, - WAC
BlALL DR LR

G! AFTER LeAK

¥: OVE:CHECK VALPDS OF INNLR Cvi, ARTER ERebaymEl’ & LEAK, CHESK
P b A e ool ’

 

BNL =
LTRER  eTALLATION® AT OB WL’

2! RLTER SLethENT MAY BE HANOLEED  ONLY IE BELEMENT %
hGAIMNRT HOIWTLURE , O, HAND, &TC,

NG PARTS

NOTE

6 2 {BDIOCTYL PUIMALATA 1 GEFICIENCY TRAT ORN
BAEEICIRNCY WRALL Bbh
AR #OW AVERAGE PARTICLE WIZE OF O.8 MCZONS.

SEPORT 44T "TTRRT OF W EFRICIENCY FLTERS

Wb, (el 1y 118 & PARTSR MG, 13T, 10
o WE GWEN' T DA TELT BLASEE PROCERDING
gt ST Caee

WITHEJT Any PRESSLURE DROWE,

M LRAK CUECK INNRE C¥L. LBAK Y0 Uk \WSIOW
AN LKID"S CC/ent. OF WhLW M.

CLECK CUT OFE TuUBL (PART iG] & PLLG WRLD, N

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. MICROFILMED

 

Link 110 GBALT FILTER AwSnmbry [E-49636
5 R Rmenat Lioenameny
MA D W
} Brivn Carne Corpamation
O TREONEE
LR )
o : | M Ll e LTER Ay ETALLS
.-: - 2 Lde res l.' o 'l .
1.4 - M-u.t...h....\.s..]
r-cs sflet | rh ) [enno-49038 ] I

 

 

 

MPOTOARNOSBE T

 

 

 

 

 
 

 

 

Ao Sl

Vs -

T il

 
 

 

 

» » L

\O 00— VT o IO 1

10.

o
w o
e . ®

1k,

)
T
.

16.

18,
19.
20.
21,
22,
23.
2k,
25.
26.
- a7.
28.

H
-3

29.

30.
31.
32,
33.
34-39.

Lo,

3:?1?'Uifliwlaffltfi:fitdzfi!bistfiC)b>¢*b‘tfig

L.
F.
E.

S.

G.
B.

REHDWDGHH W

E:Ulfiifififltdtfitfl HEoOyg

. . - - * -*
* . - » -

- . » -
- . * * -

PiUikiUiCi?lF'EI?I?!FI&:F’?flEfiIfi'fi

8L,
- 85-86.

7 88-102 .
103.

ORNL-TM-2478

Internal Distribution

 

Anderson
Baes
Beall

Bender

Bettis

Blumberg

Bohlmann
Briggs
Collins
Cottrell

. Crovley
. Culler

Ditto
Eatherly
Engel
Ferguson

. Perris

Franzreb
Fraas
Gabbard
Gallsher
Grimes
Grindell
Guymon
Harley
Haubenreich
Helms
Hoffman -
Hudson
Kesten
Kedl
Kirslis
Krakoviak
Lindeauer
Lundin

by,

Lo,
L3,
Ly,
L5,

- W6-51.

52.
53.
5k,
55.
56.
5T
58.
29.
60.
61.
62.
63.
6k,
65.
66.

67.
68.
69.
T0.
T1.
T2.
T3.
Th.
15
T6-T77.
78-790

80-82.

83.

. Lyon |

R. N

H. G. MacPherson
R. E. MacPherson
H. E. McCoy

H. C. McCurdy

C. K, McGlothlen
L. E. McNeese

J. R. McWherter
A. J. Miller

R. L. Moore

E. L. Nicholson
A. M. Perry

R. C. Robertson
M. W. Rosenthal
A, W, Savolainen
Dunlap Scott

J. H, Shaffer

M. J. Skinner

A, N, Smith

P, G. Smith

I. Spiewak.

D. A, Sundberg
R. E. Thoma

D. B. Trauger

J. R. Weir

M. E. Whatley

J. C. White

G. D, Whitman
L, V. Wilson :
Gale Young l

Central Research Library (CRL)
Document Reference Section (DRS)

- Leboratory Records (LRD)

Leboratory Records (LRD-RC)

. External Distribution

C. B. Deering, AEC-OSR
T. W. McIntosh, AEC Washington

- H, M. Roth, AEC -ORO
Division of Technical Information Extension
Laboratory and University Division, 0RO