a seawater sampling bottle (rli fjarlie)
DESCRIPTION
Bottle requirement for seawater samplingTRANSCRIPT
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 1/23
FISHERIES RESEARCH BOARD
OF CANADA
477
A Seawater Sampling Bottle
R.L.I . Fjarl ie
1952
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 2/23
dl j J )ITt!. '1 1- \ t ,
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 3/23
PACIFIC OCEANOORAPHIC
Nanaimo, B.C.
A SEAWATER SAMPLING BO'1T.LE
R.L.I. Fjarlle
P.O.G. File: N 7-7
August 15, 1952.
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 4/23
A SEAWATER SAMPLING BOTTLE
by
R.L.I . Fjarlie1
The ear l ies t seawater sampling bottles such as used by the f i r s t
Challenger expedition consisted of a simple tube, through which the
water flowed more or le ss f re ely (flushed) while i t was lowered to any
desired depth on a sounding l ine. When in position, i t was closed by
striking a t r igger mechanism with a small weight (messenger) sl id down
the l ine. The bott le and the sample were recovered by hau li ng t he l ine
aboard'ship. With the advent of reversing thermometers in 1902, a re-
versing mechanism was added.
The most successful bottles are the Nansen (figure 1) , Knudsen
(figure 2) and Ekman (figure 3) in which the thermometers a re a tt ached to
the tube which r ever se s ( tu rn s over) and closes when the t r igger is
actuated. In the f i r s t two, both ends of the bottles are attached to the
sounding l ine while being lowered into position, but when the messenger
st r ikes the release mechanism the top end fal ls away fran the l ine by
i t s own weight, and so reverses. The main difference between the two is
in the tube closing mechanism. The Nansen bot t le has a stopcock a t each
end closed by the weight of the bottle actuating a lever as it fal ls over.
The Knudsen bottle has rubber gaske tedl ids
which are s pring loaded and
are tripped individually by the messenger, as a separate act ion af te r
the top of the bottle is released from the wire. Both of these bottles
are l ight in weight and easy to handle. They have a cOllllllon objection in
that the end of the sampling tube is not perfect ly seaJ.eU when the
1. Now with the Pacific Naval Laboratory, Defense Research Board.
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 5/23
o o ~ t l e is r e v e r s e d . As a resul t , contamination o f the sample wh ile i n th e
s e a and l o s s by leakage when it is brought on deck ca n occur. T h e i r
flus hing c h a r a c t e r i s t i c s a r c poor and th e y must be allowed t o remain a t
sampling depth f o r some time t o enable them t o o b t a i n a t r u e sample.
Once the b o t t l e s a re t ri pp ed , they a r e fastened only a t on e end, an d
tend t o gyra te around th e l ine , and t h e r e is a p o s s i b i l i t y o f damage
a g a i n s t the s i d e o f th e s h i p , p a r t i c u l a r l y i n ba d we a th e r.
In th e :El' ~ b o t t l e th e thermometer and sample tube assembly i s
mounted on pivots i n a frame t h a t ca n be a tta c he d to th e sounding wire
a t both ends. The l i d s a re held t o the ends o f th e tu b e by an i n -
t e r n a l spring and a re pushed open by e x t e r n a l cams, as the tub e w ith
thermometers attached is cocked b ef or e l ow er in g. This b o t t l e is
rugged, b u t v ery heavy and t i r ing to h a n d le . I t has r e l a t i v e l y good
f l u s h i n g c h a r a c t e r i s t i c s , and with proper c a r e the l i d s ca n be
maintained leakproof. The method of locking th e tube af te r r e v e r s a l
is p o o r. an d u s u a l l y does n ot f u n c t i o n . I n t h i s case t h e b o t t l e
is f r e e to r o t a t e through 110 degrees and t h e r e is a r e a l p o s s i b i l i t y
t h a t the thermometer readings could be a l t e r e d , an d t h e l ids could be
opened.
The i d e a l water sampler must obta in and preserve a true
sample of th e water a t sampling depth, and b rin g i t on shipboard
intact . This r e q u i r e s good flus hing c h a r a c t e r i s t i c s when open, and
a p o s i t i v e s e a l when c l o s e d . Furthermore, once th e thermometers have
reversed th e y should be locked i n t h i s p o s i t i o n to s a fe g u a rd th e tem-
p e r a t u r e re a dings . The u n i t should be f a s t e n e d to the l ine a t both
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 6/23
-3-
enps to prevent damage aga in st th e side of the ship in rough weather.
The thermometer cases should be readily removable so that the instruments
may be stored properly(3)(5) between observations a t sea. In addition,
it should be l ight in weight, corrosion resis tant , easy to manipulate,
and must be dependable.
With the requirements of an ideal water sampler in view.
a new bottle has been designed (figure 4) to incorporate as many of
t he se f ea tu re s as possible. The design was commenced in 1949 and
the prototype of th e present bott le was f i r s t used in January 1950.
Since then the design has been modified six t imes. I t is presented
here as a unit which has been proven under trying circumstances a t sea
and has been found dependable, easy to handle, and capable of obtaining
t rue samples and temperatures.
This sea sampling bott le consists of a l ight weight metal tube
with hinged l ids on each end tha t are co-ordinated with a rotat ing frame
holding three reversing thermometers. The tube i s clamped to the
sounding l ine a t the top, and hooked a t the bottom end similar to the
Ekman bot t le . A device for releasing a messenger below the bott le when
it closes allows any number of units to be used in series on the l ine .
The major features which have been designed into the bottle allow unres-
t r ic ted now of water through i t when open, a positive seal when the
l ids are closed, and canplete mechanical re l lab i l i ty .
The free flow of water through the bottle bas been faci l i ta ted
by placing the operating spr ings on the outside am providing large
openings a t the ends of th e tube, in ....hich the only restr ic t ion to
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 7/23
-4-
flow is a 1/4 in::h diameter axle passing transversely across i t .
suppo rts the thermometer reversing frame. The springs which hold the l ids
closed have been coiled around the hinge on which they are mounted, in a
manner s im ila r to screen door hinge eonstruction.
The l id is carried on the hinge b olt (fig ure 5) which is bent in
such a way that the l id comes normal to the axis of the tube in closing.
The hinge bolt passes through the center of a pressed metal l id fi t ted
with a rubber gasket, which is tigh t enough to p revent l eak age , but flexible
enough to allow the l id to move s l ight ly . Consequently, if the hinge
bolt i s ac cid en tly ben t, th e l id can adapt i t se l f to the end of the
bot t le and prevent possible leaks. The metal rim which forms the seal
with the gasket has been recessed sl ight ly inside the ends of the bott le
to protect i t from accidental burrs that would destroy the seal .
The linkage between the l ids and the reversing frame is a f lexible
steel cable pass ing through a small hole in the axle of the frame, and
attached at each end to a pulley on the hinge. When the frame is rotated
to r ever se the thermometers, the cable is wound around the axle and pulls
ehe l ids open against the action of the springs . With this arrange-
ment, the thermometer frame is urged throughout the entire 180 degrees of
reversal and ,;hen the l ids are closed. a spring loaded plunger locks the
frame to prevent oscil lat ion and so safeguards the sample and the
thermometer readings.
The removable thermometer case f i t s in a rack in the rotating
f rame where it is locked with two spring catches. The thermometers are
held within thei r cases by a screwed-on l id , which covers a l l the tub es .
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 8/23
-5-
A hole is provided a t one end of each tube to faci l i ta te pushing the
thermometers out of the cases.
The seawater sample in the bottle i s in contact only with
monel metal, which i s used for the tube, l ids , reversing thermometer
frame, am axle . This minimizes the possibi l i ty of contamination of the
sample with corrosion product's. The castings on both ends of the
bottle are manganese bronze, chosen for durabili ty in service.
'rhe weight of the new bott le (7.3 lbs.) compares favorab13
with that of the Nansen bottle , 7.7 lbs . , and i s the same as the
Knudsen bott le . I t is much l ighter than the 17 pound Elanan bot t le .
In use, the bottle is f i r s t attached to the sounding l ine
by a spi ral hook a t the lower end (figur€ 6A) and then clamped
a t the top (figure6B). The plunger locking the thermometer
fra'lle is released and the l ids are opened by rotating the frame 180
degrees unt i l it engages the t r igger (figure6C). The messenger
is t hen a tt ached to a spring loaded plunger bearing against the
lower hinge bar ( figu re6D) and the messenger is then f i t ted to the
l ine (figure 7 ) .
Every attempt has been made to reduce th e number of motions
required. The hook a t the lower end is positioned so that the
movemsnts used in attaching and detaching from the l ine are those which
would naturall3 be used in moving th e bottle towards or away from
the wire. The bottle i s arranged for holding in the l e f t hand,
while the right hand performs the more complicated motion of screwing
up the clamp, which has an extra large wing nut to ensure that it can
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 9/23
-6-
be securely attached to the l ine, with reasonable effor t .
opened and the thermometers reversed in a single motion. The messenger is
attached after this action i s complet ed , so tha t it is not neces sa ry for
t he ope rato r to try and co-ordinate two moving parts.
When the messenger hits the t r igger , it releases the thermometer
frame which is now free to rotate under th e t en sion of the springs on the
hinge bar, towards the reversed posi t ion, where it is locked. At the same
time, both l ids close and the messenger for the next bottle on the l ine i s
The bott le is provided with an eye welded to the tl.\be at each end,
so it can be hllng upright on a hook before the thermometers are read,
and hung upside down af ter the sample is drawn and temperatures recorded.
In this way, th e r ever si ng thermometers can be kept in the approved storage
position (3) (5) when they are s t i l l on the bottles, i f it is not convenient
to transfer them to a separate rack. In pract ice, a wooden case is provided
with hooks, which serves as a bottle rack on deck, as well as a case for
shipping.
The messenger shown in f igure 7 is a "spli t" type, cast on two
pieces of manganese bronze, which are held together under tension by a
heavy piece of rubber. The messenger is pushed onto the l ine , and pulled
off , but has no tendency to fa l l off or to become jammed.
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 10/23
-7-
l lushing Efficiency
A Comparison with Other Seawater Sampling Devices
The speed a t which the various water sampling bott les take
up the water typical a t thei r positions in the sea i s of considerable
in teres t and importance.
The relative efficiency of f ou r ty pe s of bott les currently in use
has been investigated in one experiment in water having a moderate sal ini ty
gradient. Stl'ings of Nansen. Knudsen, Ekman and these bottles were hung side by
side from a bar suspended from ·the main cargo boom of H.M.C.S. Cedarwood.
There were four of each type of bottle on a str ing a t in te rv als of two
meters, and the end of the cargo boom was high enough above the sea to
allow a l l of the bott les to be' in the a ir a t once. With the bott les
open. and messengers attached, they were lowered into the wate r, and at a
measured time from immersion of the uppermost bott le , messengers were
dropped on each l ine simultaneously. The bott les were then hauled uP.
and the samples. were drawn. This procedure was repeated to obtain
samples after 0, 1/2, 1, 1 1/2. 2, 3, 4, 5, 6. 9, 11 and 15 minutes
submersion time, the d if ferent in te rval s being observed in random order
(Table I ) .
With this method of sampling i t is believed possible to estimate
the time required for a b ottle to flush completely, by comparing the
sal ini ty of the samples taken. Ideally, the time required for complete
flushing of a l l bottles is the time which must elapse af ter submersion.
before a l l bottles a t the same depth have identical samples. In th i s ex-
periment an arbitrary reference salini ty for each cast was assumed by
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 11/23
choosing the maximum sal ini ty a t each depth, regardless o f the type of bott le
from which it was drawn. I t was a ssum3d tha t if the sa l ini ty as shown by
one bottle was less than the reference a t the same depth, it was because
some water of a lower salini ty had been trapped in the bottle , and carried
down with it to the sampling posit ion.
To evaluate the relat ive efficiency of the different bott les , the
area between the reference sa l ini ty curve, and the curve shown by each type
of bottle was measured for the various submersion times. I t was a ssumed
tha t as this area approached zero with longer " soak ing", flushing became
more complete. ThUS, the area between the curves and the time required
for i t to approach zero, i s the measure of efficiency., For d iscussion pur -
poses, the units of area have been cal led "Sal in it y Meters" (S.M.).
The data are shown in Table I . The sal ini ty profi les for a sub-
mersion time of 0 minutes are shown in figure 8" together with the re-
ferenceprofi le .
The relat ive efficiency of the four types of bott les is shown in
f igure 9. The variat ion in the data i s about 0 .25 S.M. and becomes approxi-
mately constant after 9 minutes sul::mersion time. The data for the 6 and
15 minute intervals are not consistent with the remainder of the series
and cannot be explained from th e records of the experiment. Despite these
anomalies the relative efficiencies are clearly indicated. In these waters
where the sal ini ty (and consequently the density) gradient was large, the
sample in the new bottle apparently was in equilibrium with i t s surroUD-
dings immediately, the Ekman bott le samples were in equilibrium after one
half m...nute, the Knudsen samples af ter two minutes, but the Nansen bott le
r " ' ' 1 ~ . . r e d five minutes "soaking" to obtain a true sample.
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 12/23
-9-
When a bottle is being lowered flushing occurs due to the velocity
head i n t he f l u i da t t h eendopen i ng , anda f t e r i t i s i npos i t i on t he
l ighter water entrapped and carried down in the lowering. wil l f loat out
and be replaced by the denser surrounding water. Small end openings and
in ternal obstructions cause a flow resistance which is most effective in
retarding th e flu sh ing while lowering when the velocity is greatest , but
is a lso e ff ec tive in retarding th e hydrostat ic f lushing when the bottle
is in posit ion.
The data in figure 9 show that t he e ff ic iency is proportional to
the effective size of end openings and freedom of f low through the
bot t les . The end caps on th e EIanan bott les must retard free flow, but
equilibrium is quickly attained (1/2 minute) because th e openings are
large. and the tube is unobstructed. The end opening of the Knudsen
bottle is of a fa i r s ize. but is part ial ly obstructed by the internal
spring, which also occupies a large proportion of the cross s ec ti on a re a
of this narrow bottle. The stopcocks on the Nansen bo tt le const ri ct the
end openings so much that they require more th an twic e the time of any other
bott le for flushing.
The bottle described here approached the postulated requirements
more closely than any of t hese s tandard des igns . and i t is belieTed to be
an acceptable improvement.
Copies of the plans and specifications may be obtained a t cost
from the Oceanographer in Charge. Pacific Oceanographic Group,
Nanaimo, B.C.
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 13/23
1. Anonymous. Deutsche Atlantische Exped. 'Meteor ' , Vol. IV, Part I , p 21
f f .1925-27.
2. Knudsen, M. A frameless reversing water bott le . Journal du Conseil ,
Vol. IV, No.2 , P 192, 1929.
3. Ekman, V.W. On the use of insulated water bott les and reversing
the rmomete rs. Pub!. de Circ. No. 23, April , 1905.
4. Knudsen. M. Some oceanographical instruments. Publ. de Circ. No. 77,
April , 1923.
5. Richter, N. Uber die funktion der umkipp thermometer sowle die beseitigung
von storungen. Richter and Wiese. Berlin, Germany, 1951.
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 14/23
The salinity of samples taken simultaneously in four types of seasampling bottles after various intervals of immersion.
Typesof Bottles
Depth~ ~ ~ ~ !!!!!!a
Salinity (S ·1 •• )o }linutes "Immersion"
23.35 23.39 23.33 23.3123.42 23.33 23.35 23.37
24.43 23.57 ;13.64 24.3024.69 24.78 25.48
1/2 Minutes "Immersion"
23.93 23.90 23.93 23.9024.16 24.00 23.90 23.95
24.36 24.42 24.7825.62 24.90 25.55 25.64
1 Minute "Immersion"
24.40 24.2724.29 24.29
24.29 24.30 24.30 24.2925.30 24.74 25.01 25.2825.68 24.69 25.52 25.68
1 1/2 Minutes "Immersion"
23.98 24.50 23.98 24.0224.20 24.14 24.16 24.1624.81 24.61 24.69 24.8125.81 25.52 25.73 25.80
2 Minutes "Immersion"
24.16 24.38 24.40 24.4024.5:1. 24.45 23.87 24.4725.35 25.16 25.39 25.41
25.59 25.50 25.59 25.57
1 Minutes "Immersion"
24.02 24.00 24.02 24.0024.00 24.04 24.00 24.0424.92 24.61 24.92 24.9425.75 25.53 25.73 25.72
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 15/23
TABLE I Continued.
Types of Bottles
Depth
~ ~ ~ ~ ~Salinity (8 ·1 ••)
4 Minutes "Immersion"
23.60 23.69 23.71 23.69
23.73 23.71 23.73 23.74
24.34 24.05 24.18 24.34
25.59 25.48 25.57 25.57
5 Minutes "Immersion"
24.14 24.13 24.11 24.05
24.11 24.13 24.11 24.14
24.47 24.38 24.31 24.47
25.73 25.64 25.72 25.72
6 Minutes "Immersion"
24.11 24.18 24.11 24.11
24.22 24.27 24.27
24.76 25.16 24.69 25.10
25.82 25.84 25.82 25.84
9 Minutes "Immersion"
23.59 23.66 23.62 23.57
23.69 23.62 23.71 23.68
24.45 24.54 24.61 24.50
25.55 25.61 25.62 25.62
11 Minutes "Immersion"
24.18 24.07 24.11 24.11
24.22 24.25 24.27 24.27
24.58 24.54 24.52 24.6025.81 25.81 25.80
15 Minutes "Inunersion"
23.48 23.44 23.38 23.44
23.48 23.51 23.50 23.55
24.09 25.10 25.37 25.3925.64 25.72 25.62
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 16/23
I n pos it io n Reversing
The Nansen deep sea reversing
irater S8JIlpling bot t le .
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 17/23
In pos ition Reversing
The Knudsen deep sea reversingwater slillpling bottle.
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 18/23
In position Reversed
The Ekman deep sea reversing
water sampl ing bottle.
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 19/23
·The seawater sampling bot t le .
Detail of l id construction onthe seawate r sampling bot t le .
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 20/23
Attaching the new bottleto the Bounding l ine .
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 21/23
Pushing the sp l i t messenger
onto the sounding l ine .
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 22/23
The sal ini ty of samples taken simultaneously in the
four types of bottles af te r zero minutes sul:mersion.
Deviation of the sal ini ty in each type of sample bottle from thereferen ce v alu e. E fficien cy varies inversely with the deviation.
7/16/2019 A Seawater Sampling Bottle (RLI Fjarlie)
http://slidepdf.com/reader/full/a-seawater-sampling-bottle-rli-fjarlie 23/23