Some Practical Aspects of Ship

Launching
Part 1. Starting and Checking
BY .HARRISON B. ANDREWS, I~/~EMBER2
In the first part of this paper it is proposed to
dwell briefly upon two aspects of launching--the
engineering and preparations required to assure,
as far as possible, an unaided start and the engineering and set-up required for checking the travel,
without regard to studies of pivoting, way-end
pressure, stability, etc.
1 Paper presented before the October 1945 meeting of the New
England Section of The Society of Naval Architects and Marine
Engineers.
2 Engineer, Bethlehem Steel Company, Shipbuilding Division,
Quincy, Mass.

STARTING

Before any new ship construction may be

started, several launching problems present themselves for consideration. The decisions reached
at that early stage tend to allay the fears or to
increase the anxiety on launching day of the persons responsible for the launching operations.
The two most important figures to be settled
for the successful start of a vessel at launching are
the slope of the ground ways and the pressure on

3.2

3.'0
I

2.8

2.6

A~rcraff

Carrier

Z Bafflesh~p
3 Cruiser
4 Pass.torso Ship
5 Cargo Ship

6"
to 2.4
o)
O.

2.2
~%- 2 0 ~ ' ~

IO Trawler

0:2.0
u)
to
I-

o.

1.8

....

1.6

-Z22ZL~....~......_

-6
c

1.4

,W ~-=-

1.2

1.0
0.8
0.45

0.50

0.55

0.60

0.65
0.-/0
0.'/5
O.BO
5lope o4 Ways, S~ In. Per" Ft-.

Fro. 1
424

0.B5

0.90

0.95

LO0

SHIP LAUNCHING

425

The figures of grease pressure mentioned are not

th~ grease. The slope of the ground ways affects
the elevation of the vessel on the slip during con- hard and fast limits, as m a n y launehings have
struction and consequently it must be determined been performed successfully with pressures both
before construction has started. Since it is ad- above 2.25 and below 1.75 tons per square foot.
vantageous for the construction departments to Since in all cases the object is to set up the launchbuild a vessel close to the ground, there is always ing arrangements for a good start, and the pubpresent, in the early stages, the tendency to sacri- lished figures of pressure and slope in connection
fice starting slope for lower elevations.
with slow starts are obviously few, it is difficult to
Good launching practice indicates a fairly defi- obtain data for the lower limits of pressure.
nite relation between the slope of the ways and the
L a b o r a t o r y test curves are available from the
pressure on the grease. This relationship is il-- various grease manufacturers, and also from tests
lustrated broadly in Fig. 1. All of the points at some shipyards, showing coefficients of sliding
shown are actual figures of typical vessels launched friction plotted against grease pressure b u t it is
under the observation of the author. The dotted
difficult to translate the results from the smalllines indicate the upper and lower limits of typical scale test equipment into the much larger and
launching practice and were chosen primarily to rougher actual launching set-up.
define a band encompassing most of the past sucA ship's launching justifiably m a y be classed
cessful launehings observed b y the author. The equally as an art as a science. Despite the acband is wide and appears to allow considerable counting for all known factors and identical conspread in pressure for a fixed slope. I t is none too ditions for two launehings, marked differences in
wide, however, to allow for the variation in es- performance often occur. This is notably true in
timated pressure before launching. Present ship starting. For example, instances can be cited
construction methods develop large assembly sec- when, with all conditions constant except tempertions and the difference in launching grease pres- ature, the launching taking placed in the lower
sure caused b y one or two large superstructure as- temperature will be normal while t h a t at the
semblies being on or off the ship can be quite no- higher temperature m a y be sluggish in starting or
ticeable.
the ship even m a y refuse to move. This does not
I t would be quite simple to solve the starting discredit the t r e a t m e n t of the problem as a sciquestion in a n y launching b y setting up the ground entific one but simply urges the addition of a
ways with a high slope and letting the pressure grain of salt to the result and makes the final succome what it may. This obviously is not the solu- cess of the launching more savory.
tion in the case of vessels of the length of our presAdditional considerations of the starting probent airplane carriers, where the bow of the vessel lem ar~ largely in the hands of the carpenters who
would be out of sight on foggy days. On the construct the launching cradle. The ground ways
other hand, a trawler could not be set up sensibly should be alined closely to the required declivity;
on a low slope, relying upon the pressure for a good the grease should be applied carefully; the cradle
start, since the width of ways would be too narrow should be constructed so as to o b t a i n the best
for reasonable cradle construction.
possible distribution of load over the grease; and
The shipyards of this country have accepted the wedging up and removal of shoring and blockmineral base greases almost unanimously with ing should be carried out according to schedule, so
verv few exceptions such as the one in the South as to lower the cradle as evenly as possible on to
which is quoted as using bananas. All of the vari- the grease surface.
ous companies handling launching greases have
The care taken to aline the ground ways propcontributed to the i m p r o v e m e n t of the product so erly becomes particularly i m p o r t a n t when four
t h a t the shipbuilder now can rely on the published launching ways are used. Although the use of two
data without hesitation. N o t too m a n y years launching ways is almost universal, even in the
ago in this country, and still t o d a y in some Euro- cases of the Queen Mary and Normandie where the
pean yards, the carpenter has formulated the launching weights, including cradles, were 36,700
launching greases and, as with all good chefs, tons and 27,660 tons, respectively, several fouradded his own special touch or words of sorcery to way launchings have been carried out in recent
the brew.
years. Among these were the recent launehings
T o d a y we accept the fact t h a t the base coat of our new Iowa and South Dakota classes of batshould give a hard smooth surface capable of tleships, when four ways were used because of the
withstanding the pressures at pivoting, and t h a t
high launching weight concentrated in a relatively
the slip coat should give constant low coefficients short length. The custom in the Netherlands has
of friction, without squeezing out, under pressures been to concentrate the load on a centerline way
of 1.75 to 2.25 tons per square foot.
with secondary ways outboard for stability.

426

SHIP LAUNCHING

With the accomplishment of wedging-up and the

removal of shoring and blocking, the weight of the
vessel is transferred to the sliding ways through
the cradle and poppets. Referring to Fig. 2, the
components F1 of the total weight of the ship and

movement, while at the other end rope lashir]gs

made it fast by applying force in the direction of
the force applied by the ship's weight down the
ways. Another similar method is employed at
present in the side-launchings on the Great

4,

FIG. 2

Slldi~g Way(

)Sole ?lae

sliding structure down the ways is opposed, upon

release, only by the force of friction F~. Now
F1 = W s i n 0
FIG. 4

and
Fa = f t V c o s O

where f is the coefficient of friction.

For the vessel to move, W sin 0 must be greater
than f W cos 0; t h a t is, tan O, the starting slope,
must be greater than f, the coefficient of friction.
Before release, this net force down the ways of
W sin 0 -- f W cos 0 is neutralized b y the triggers or
other holding and releasing mechanism, of which
there are several types.

Ground Wa
Fulcrum Pile 9
-~----Timber Lever
Deo4man

FI~. 3

Among the earliest and simplest holding and

releasing devices are those using a few pieces of
heavy timber and rope lashing. One device,
shown in Fig. 3, was a type of dog shore and consisted of a large beam laid at right angles to the
ways on each side of the ship. Some distance
along the beam, on the water side, was a pile
driven into the ground for a fulcrum. One end
was applied to the sliding way to prevent its

Lakes. Here, again, use is made of a beam as a

lever with the outboard end lashed to a deadman.
In both eases the release of the ship is effected by
simultaneously cutting all the rope lashings. In
some instances this is accomplished b y men with
axes. In others a guillotine arrangement is used.
T h a t used in the Great Lakes side-launchings
consists of pneumatically operated knives, all connetted to the same air line.
A second method of holding, still in use for
launchings of moderate weight, is that which employs a wooden "saw piece" or a steel "sole plate"
to tie the forward ends of the sliding ways to the
stationary ground ways. This is illustrated in
Fig. 4. The release is brought about b y cutting at
equal rates the two "saw pieces" or "sole plates."
Ultimately these ties should break and the area
through the breaks provides a straightforward
means of measuring the net starting force down
the ways. However, high tensile stresses are set
up in the forward ends of the sliding ways and
the loading cannot be distributed evenly throughout the length of the sliding ways nor over more
than two such holding units.
In universal use, at present, for launchings of
any large magnitude is the system employing
groupings of hydraulic or mechanical triggers arranged in such numbers, and so spaced, as to limit
the local loadings in the sliding ways or on the
triggers. Both types of trigger have been used in
recent years, the hydraulic (Fig. 5) now having
been replaced by the mechanical in many yards,
although overall usage indicates about equal favor
for both. The holding effect of the hydraulic trigger is brought about usually through the appliea-

SHIP LAUNCHING
tion of force from an hydraulic ram to one end of a
lever whose other end prevents movement of the
sliding ways. Removal of hydraulic pressure
through the handling of one valve at a central
control station releases all triggers together.
The adoption of the use of mechanical triggers
at the Quincy Yard of the Bethlehem Steel Company was brought about when new contracts required the building of additional means of holding
5hoe

'1

~/./K./,.///,,.

I1,11I]

Preventer

ka'fch
FIG. 5

and releasing vessels up to 28,000 tons launching

weight. At t h a t time, consideration was given to
the possibility of designing a trigger radically different from the hydraulic type which was being
used, in forms generally similar and varying only
in detail, in all of the large shipyards in the United
States.
Since the shipbuilding program then indicated a
great m a n y vessels to be built, and of varying
types, it was thought advisable to design the triggers so t h a t they could be used in pairs, one pair
for vessels up to 10,000 tons launching weight,
two pairs for vessels between 10,000 and 20,000
tons, and three pairs for vessels from 20,000 to 30,000 tons launching weight. This system also
would provide the advantage of distributing the
holding power along the length of the cradle instead of concentrating it at the one trigger location, as had been the practice previously.
The design problem then became one to produce
a workable trigger, simple in installation, as foolproof as possible, to allow release in a limited
vertical clearance, to be applicable to various
types of vessels and, when used in groups, to make
possible a simultaneous release. These conditions
led to the adoption of mechanical triggers capable
of release electrically or manually. The details of
the design are deseribed in the second part of this
paper.

427
CHECKING

While sliding down the ways and until entry

into the water, the two previously mentioned
forces are acting upon the ship and sliding structure: (1) Its weight, bringing about a component
of force acting down the ways, and (2), the force of
friction between the sliding and ground ways acting in a reverse direction. As soon as the cradle
enters the water, a third force, that of water resistance, is introduced which also acts to retard
the ship's velocity and to limit its travel. T h e
third, and final, phase of the ship's travel begins
as the vessel floats or drops off the way ends.
From this point on, there is no component of force
down the ways in effect and the ship comes to rest
when its kinetic energy is entirely dissipated b y the
water resistance alone or, if space is limited, it becomes necessary to add a further resistance so
that the travel, over which work must be done by
the resisting forces, becomes less.
There are, of course, m a n y yards so advantageously situated alongside a clear expanse of deep
water that, once released, the vessels are allowed
to run free until picked up by tugs. Some of the
very largest ships, however, have been launched
in very restricted waterways, such as, for example,
the River Clyde in Scotland. In such cases, unusual precautions must be taken for the ship to remain clear of the opposite shore. In the case of
the Clyde, the slipways are laid out at an angle of
about 30 degrees to the shoreline, considerabIe
dredging is necessary and checking devices are
used.
Checking devices have taken several forms and
have been used in m a n y combinations. The simplest type is a mask hung at the after end of the
ship so as to present a plane surface at right angles
to the line of motion and thereby increase the
water resistance. Only a relatively small gain is
added, however, and this is not easily estimated
without model experiments. Furthermore, water
resistance varies with the square of the speed and
does not remain a substantial, effective force.
Theoretically, then, the ship will keep moving
into infinity unless other forces are brought to
bear.
Where a more positive method of control is desired and space is still not severely limited, anchors may be dropped when the vessel is water
borne.
Among the devices used in Germany on the
Kaiser Wilhelm II, some time prior to 1903, was a
wedge friction brake consisting of a wedge-shaped
log made fast to the ship b y lines so that the log
was pulled against friction through a tightly fitting trough of steel bands. No other instance of its
use is known to the author.

428

SHIP LAUNCHING

There is, further, the device which employs

rope stops. Here a length of chain is made fast
to a stationary point on land and another is made
fast to the ship. Between them run rope lashings
which are spaced so as.to break at intervals in the
travel of the one chain relative to the other. The
travel of the ship during launching brings this
about and the breaking of the rope lashings serves
to decrease the ship's speed and to restrict its
travel.
At the launching of a battleship in 1919, hydraulic rope brakes were used for checking. These
were made up of a mated pair of steel shoes
grooved to take a wire-rope cable fixed to the forward end of the ship. T h e y were mounted on the
ground, under the bow of the ship, and so fitted
with a hydraulic ram as to make it possible to vary
the pressure between the shoes and thus vary the
frictional force on the wire rope running between
them. Five such brakes were used on each side.
The disadvantage of this system of checking is
that it is difficult to gage the pressure to apply and
the method requires manual control during a period of excitement and strain when improper action
is more apt to be taken and signals are more apt
to be confused.
The most common checking methods in use today for launching vessels in restricted waters are
those which lend themselves most readily to direct
calculation. The prevalence of the moment is on
the side of some form of the chain drag system.
When the initial checking calculations indicate
that the length of the launching basin is not sufficient to allow a free run but is long enough to allow
the water resistance to act for a relatively long
time, slewing methods are generally the most practical, since the amount of chain required can be
kept to a minimum. In general, any method of
slewing involves placing of chain piles or other
weights in the water, clear of the building slip,
and making them fast to the stern of the vessel
with wire drag ropes of predetermined length.
When the vessel is waterborne during launching,
the resistance of the chain pulling along the bottom of the launching basin slews the vessel from
its initial line of motion.
The method of slewing, using chain drags, has
been used successfully in recent years in the
launching of some of the heaviest vessels constructed in this country. Previous to the launching of one of the Iowa-class battleships, model experiments were conducted to ascertain the slewing
effect of the chain drags with several coefficients
of drag friction. The use of models in this connection proved to be a valuable aid in predicting
the stopped position of the vessel and in determining the amount and distribution of chain used.

The other common method of checking at

launching, using chain drags, consists of placing
the chain on the slipway alongside the vessel to
offer additional resistance while the ship moves in
a straight-line path, and of making the piles fast
to the vessel with wire drag ropes attached near
the bow of the vessel.
The factors involved in the solution of a checking p r o b l e m u s i n g either the slewing or the
straight-line system are the same. These factors
include the weight of the ship and cradle, the depth
of water over the way-ends, the coefficient of friction of the grease, the water resistance coefficients
and the coefficient of friction of chain on the drag
surface.
The launching calculations performed for producing the routine launching curves provide the
figures for the first two factors enumerated. The
three coefficients of resistance must be determined
experimentally or estimated from the analysis of
actual launehings.
Coefficients of friction of launching grease will
vary somewhat, depending upon the type of
grease selected. The results of several hundred
launchings within the author's experience have
shown, however, that the variation in coefficient
of grease friction is small. The figures commonly
used will vary from 0.0175 to 0.025.
Data obtained from experiment for coefficients
of chain drag friction have been questioned, since
it is difficult to conduct the tests with high enough
velocities to compare with the actual launching
conditions. Experimental data have been used,
however, and the overall results of the launchings
have indicated that these data are well within the
range of accuracy of the launching calculations.
Typical data for coefficients of drag friction for
.various surfaces are:
Hard-packed dirt
Smooth concrete
Sand on concrete

0.65-0.75
0.35-0.45
0.65-0.70

Although some model experimental data are

available for water resistance coefficients, most
shipyards must rely on the data obtained from actual launching analysis.
ANALYSIS

The analysis of results of a launching is carried

out as a check on the estimates made in the design
stage and equally to determine the values of water
resistance and coefficients of grease and drag friction.
The basis for such an analysis is an accurate
record of time and distance obtained b y any one of
several timing methods or an accurate record of
acceleration as measured b y means of an accelerometer. Since acceleration is the key figure in

SHIP LAUNCHING

429

300

200

~00
o
t'-

+
)
0

0,,o

I00

ZOO

900

800

700

600

500
400
T r a v e l - Fe~+
FIG. 6

the launching analysis, the aceelerometer method

eliminates the necessity for successive differentiations of the time-distance curve.
The force diagram (Fig. 6) shows graphically
the relative amounts of work done b y each and
their interrelation. T h e work done b y the ship's
weight during the launching is represented b y the
area under the circumscribing trapezium. T h e
formula for the boundary curve is W sin 0, where
W is the total weight of ship and cradle and 0 is
the angle of slope of the ways. The resisting
force attributed to buoyancy is B sin 0, so t h a t the
actual force acting down the ways is (W -- B) sin
0, B being the buoyancy.
Grease friction is expressed in the form f (W -B) cos 0. The accurate value of W is determined
after the launching from a reading of the ship's
drafts, while f, the coefficient of friction, is found
in the following way.
With the acceleration and the weight of the ship
and cradle known, the resultant'force F can be determined from the formula
F = ~a

Wa
= g

Referring to Fig. 6, for the portion of the ship's

travel from the start until the cradle enters the
water

300

~_00

I00

Fa= F I - F

and the coeffcientof grease friction f is

f

F~
(W - B) cos 0

or

f = tan 0 - a
g cos 0
The value of the coefficient of drag friction can
be found in a similar manner b y utilizing the end
spot of the force diagram. At t h a t point the
forces of grease resistance and water resistance
have become zero and
~=F
then
F5 = f ' w

where f ' is the coefficient of drag friction, and w

is the weight of drags.
The last of the unknowns, and the most imp o r t a n t from the standpoint of valuable data, is
the water resistance coefficient.
The value of the force of water resistance #'4 can
be found from the formula
F = F,--F2-Fa--F4--F6

450

SHIP LAUNCHING

1400

1200

cargo

-----

Pass.Sh{p

Mrcra~ C~rrier

10O0

/,~.
,-,

Desfroy~r
Cruiser

.....

-l~

/ " ~

NN

800
2

~I

o 600

1"/

. ~"

/'"

40o

"/

~-'i:

/F
0
o

Io

Z0

ao

go

50

Per Cen~rof

60

To~olBuoyancy

70

8o

g0

Ioo

FIG. 7
since at this stage of the analysis all of the other
forces are known.
When the force of water resistance is denoted as
F4 = K v ~

it has been shown t h a t the coefficient K may be

expressed as
B2/,
C

where C is a value varying with the type of vessel

and the buoyancy.

Part 2.

Curves showing the value of C for different

types of vessels taken from the analysis of actual
launchings are shown in Fig. 7 plotted against per
cent of total buoyancy.
In the use of the curves of the coefficient C, the
assumption can be made that C remains constant
between float-off and the stopped position. The
analysis of many launching results has shown that
the error in this assumption is small and will have
relatively little effect on the overall results.

Mechanical Triggers

BY ARCHER /V[. NICKERSON, IR., ASSOCIATE ~[EiViBER 3

Mechanical triggers of the multiple-lever type

have been used in foreign shipyards for m a n y
years. Their use in American yards for end3 Engineer, Bethlehem Steel Company, Shipbuilding Division,
Quincy, Mass.

launchings has not been customary; {n fact, at

the time of the trigger developments to be described, which began in the spring of 1940, there
was virtually nothing similar in American practice for end-launchings of larger vessels with the

SHIP LAUNCHING
exception of the lever and hydraulic r a m systems
known as hydraulic triggers, mentioned in P a r t 1
of this paper and one type of which is illustrated
in Fig. 5. I t is believed therefore t h a t an account
of the development and use of three successful
mechanical trigger designs, with which the authors
are familiar, m a y be of general interest. I t is
elected t h a t this be an outline of what was done
rather than a treatise on how such triggers should
be designed, for while the principles m a y be the
same the launching conditions will v a r y between
shipyards and for this reason no standard trigger
type can be recommended. Figs. 8 to [5 inclusive
are diagrammatic. No details are given as their
development followed accepted engineering practice.
Mechanical and hydraulic launching triggers
offer a number of advantages over a n y other type
of holding device. A multiplicity of such triggers
can be arranged readily for simultaneous release
and therefore can be used for safely holding the
largest vessels. The fastenings between wood and
the steel trigger frames can be made dependable.
The force down the ways can be well distributed
over the cradle and ground ways and in large
measure can be transmitted to the wood in compression. B y comparison, the holding of a hull
at the forward end of the cradle concentrates
loads on members least able to withstand them.
When successive launchings of the same type of
ship from one slipway are required b y a quantity
building program, the state of fastenings transmitting loads in shear and in tension from the forward end of the sliding ways to the ground ways
is sometimes in doubt, because of a tendency toward gradual failure of the wooden members with
repeated loadings.
The use of the hydraulic trigger is subject to
some disadvantages which can be serious. Piping joints m a y fail as pressure is developed in the
hydraulic system. The r a m packing m a y start to
leak at an embarrassing time in the schedule of
prelaunching events. Exposed piping and valves
are subject to possible damage while shores and
blocking material are being removed from the
hull. In theory, the hydraulic system m a y be
filled and its pressure, as the load comes on the
triggers, m a y be taken as a measure of the total
load down the ways. In practice, the systems
are often p u m p e d to a pressure well in excess of
the a m o u n t anticipated, in order t h a t there will be
no question about holding. Hydraulic triggers are
sometimes slow in releasing, especially when a
group of triggers m u s t be served b y one relatively
small releasing valve. In some cases this has resulted in scuffing of the cradle b y failure of the

451

trigger load lever to clear the sliding ways

p r o m p t l y upon release; in others, the trigger design has been complicated b y the application of
supplementary air-operated booster rams to force
the main levers away from the cradle. Finally, the
hydraulic fluid must be of an anti-freezing type for
winter-time launchings and there m u s t be a sufficient margin of volume on hand to allow for an
amount of leakage difficult to predict. After
weighing these facts against the probable advantages and possible deficiencies of mechanical
trigger operation, a decision was made to proceed
with the mechanical type in a design exceeding the
capacity of anything at hand in the shipyard at
the time.
A precedent for this design was the type of trigger used in launching the British steamship Queen
Firsl L e v e r ~
Tricjger

..

Frome

i ~\'<

f-'!-'"

', ~Second Lever

I
i

],

}
__~Releasing

I
i

6ear

"

We~cJh

O-uadran~

1Wooa Founda'Hon 1
FIG.

8.--LAUNCHING

~Hgger

TRIGGER--SET

POSITION

""

/XSec0nd \

\/--gaelec~ing 6eor

Wood "-"-L~/7 I
Foundv,~ion ~ ', '
We, h~ ~'~'~'

/
,"
/

FIG. 9.--LAUNCHING TRIGGER--RELEASED POSITION

Mary.

This is shown in Figs. 8 and 9 and consists

of a train of levers, the load being applied to the
"main lever" and reduced b y the overall mechanical advantage of the system to an a m o u n t which
can be controlled b y the safety quadrant illustrated. T h e trigger is released electrically b y the
action of a solenoid. In the "set" position, a weight
in the form of a h a m m e r is suspended b y the solenoid latch. The h a m m e r "handle" rides in a slot in
the quadrant are and, bearing on the upper end of

432

SHIP LAUNCHING

Sa~efy Clip

Hammer I

.Releose Lever
/ SolenoTd
-FirsfLever

,/
i

~I

~I/_/~u~/Yoke

.9"l_

'--[i__.

I .~;~',,X~,~.-.~ .L/l r Ind~cafor \ ,/ ,'~',

", \ _ _ 2 1

,~-.q lO.580Lb. F

Trigger
Frame
'

Or~ L'nk ~ B u ~ ; : ; n [

Lev'r

6 12
Inches

2
/

~
0

". . . .

Topof Ground W~

6 IZ
Inches

FIG. 10.--170-TON LAUNCHING TRIGGER--SET POSITION

FIG. l l . - - 1 7 0 - T o N LAUNCHING TRIGGER--RELEASED

POSITION

this slot, holds the quadrant in positive engagem e n t with the long a r m of the last lever in the
train, thus preventing its inadvertent release.
When the solenoid latch is tripped, the h a m m e r
swings freely until its handle strikes the other end
of the q u a d r a n t slot. T h e impact turns the quadr a n t on its axis, permitting the last lever in the
train to fly upward and thereby release the other
two, the main lever falling away from the structural bearing m e m b e r in the cradle and releasing
the ship. Three pairs of these triggers were used
in this launching, the solenoids being wired in a
common electrical circuit. For all practical purposes this arrangement assures simultaneous trigger release.
T h e release b y gravity of the Queen Mary triggers is questionable because the third lever must
rise. T h e releasing quadrant is mounted on a separate structure a p a r t from the ground ways. I t
was felt t h a t this design could be improved upon
b y the elimination of these two features, since
tile rising lever offers additional resistance to be
overcome in the event of a sluggish launch, and
the separate support for the holding and release
of the last lever is subject to possible damage during removal of shores and cribbing. The new design therefore was based on a lever and releasing
system self-contained in the trigger frame, each
lever of which was to fall freely upon release.
T h e resulting arrangement is indicated in Figs.
10 and 11. These triggers were planned specifically for the launching of U.S.S. 3Iassachusetts and
three pairs were to be used in the set-up. The
original load calculation per trigger, based on uniform distribution, was 170 long tons. I t was decided t h a t the trigger elements should be stressed
to the elastic limit at 515 tong tons load. The solenoids of each pair were wired in series and the
pairs in parallel, for it was desired t h a t in the event

of last-minute failure of electrical wiring the triggers should at least be "hung-up" in pairs, two of
which would hold until released b y hand or one of
which might hold, depending on actual loading, or
be forced away from the cradle without jeopardizing the success of the launch.
A test set-up was made in which each finished
trigger was bolted to simulated ground ways,
given a check for " g r a v i t y " release and load by
increments to a total of 340 long tons or 100 per
cent over the average load anticipated. Other
special tests included separate overloading of the
outside or third lever, simulating trigger load up
to the elastic limit of the levers, and checks of the
releasing mechanisms at this load, wiping contact
surfaces between the third lever and its supporting
yoke dry and finally abrading t h e m in a demonstration of adequate h a m m e r energy. The strain
gage shown in Fig. 10 and used for giving a measure of load on the trigger in terms of the deflection
of the last lever was also calibrated during tile trigger tests. The triggers were released under load
onto the eccentric safety pins and reset after rotating the pins b y means of the hand lever provided as an integral p a r t of each pin. Following
this, they were released under load as in actual
service.
T h e first trigger test demonstrated the need of
rugged b u t simple wooden buffers shaped to the
general contour of the levers for good impact
load distribution. These had been fitted for the
main and second levers. At no load, the release
averaged 2 seconds time from the solenoid action
to clearance of the main lever. Under load, the
release was quite like an explosion and it was impossible to distinguish the "one-two-three" order
of dropping the levers. The outside lever flew
around at great speed and b u t for an intervening
plank would have carried away the links and le-

SHIP LAUNCHING
Top of Ground Wmy

.e"

TOp of Ground Way

C ~/
~

433

41..__!_~~ "tons
Lb.

Trigger~~
Frc~rnet ~

~ . ~,
c~ ~.

Trigg

Frome ~1

t]

"'-2-

SecondL e v e r /

0(5.

Safe+ypin/ "~,~jxF
~o, r~hLever ~Thlrd Lever
\.%
tb

"%7~.

Sccde

0
0

6
]nches

~2

Hook S~ripF

-~

FIG. 12.---40-TON LAUNCHING TRIGGER--NET POSITION

vers of the releasing mechanism. Under a hull the

stored energy produces much the same effect but,
as the total resilience is less than t h a t of the trigger and test frame combination, the lever impact
is not as severe. The trigger resilience contributes
to a snappy release and a buffer was found essential for arresting the motion of the last lever as
well as t h a t of the other two.
The strain gage, eccentric safety pin and release
lever latch were additions made to the design
after testing the first trigger. The strain gage was
the result of a suggestion t h a t it would be desirable to measure the total load down the ways and
determine its general distribution on the individual triggers. These gages were not expected to
be extremely accurate, b u t they have served well
as an indication of the percentage load distribution. A jacking arrangement was requested to enable a full test of the releasing mechanism up to the
m o m e n t of launch if desired. The eccentric safety
pins worked out very nicely. T h e y are set to be
clear of the last lever b y }~6 inch when the hand
lever is down and to raise the lever }{ 6 inch clear
of its supporting yoke when the hand lever is
turned 180 degrees. T h e y require about 20
pounds on the hand lever to reset the trigger under
full test load. The release lever latch was applied
as a safety measure "to hold the supporting yoke
clear of the last lever and to prevent automatic
resetting of the trigger in the event of a sluggish
release.
Several wiring arrangements are possible for
the releasing solenoids. The fundamental circuit
adopted is described in the foregoing. To this
were added a test circuit and switches permitting
the application of low voltage to each solenoid in
order to check the voltage drops in the individual

Inche~

12

FIG. 13.--40-TONLAUNCHINGTRIGGI~R--RI~LBASBD
POSITION
circuits at a n y time up to the m o m e n t of launch.
This voltage was tapped off from the same storage batteries as used as a current source for the
actual release. In addition, signal lamps were
arranged to glow at each trigger when its solenoid
is energized and at the control panel when each
trigger is released, the latter actuated b y normally closed switches which are held open in the
"set" position of the levers. Pushbuttons were
provided for testing these lamps. The main control panel contains the test switches, pushbuttons,
voltmeters, "released" signal lamps and a twopole, single-throw master switch with a positive
safety lock in the open position.
In use, a "trigger m a n " is assigned to each trigger for taking strain gage measurements and for
removing the h a m m e r safety clip and outside
lever safety pin. An electrician is also on d u t y at
each trigger for transmitting readings and instructions between the triggGr man and launching log
desk b y telephone. Upon removal of all supporting material under the hull, the trigger men are
instructed to remove strain gages, h a m m e r clips
and safety pins in t h a t order and to stow t h e m in
a rack provided for the purpose adjacent to the
log desk. All triggers and these removable parts
are numbered and the latter have their assigned
places in the rack. When the rack is filled, those
in charge know t h a t the triggers are clear. After
a warning signal to the sponsor, the triggers are
released b y the single act of closing the master
switch. When the "release" signal lamps glow,
the master switch is again opened to de-energize
the trigger solenoids.
After a "trial trip" in which one pair of triggers
was used to launch a tanker, the three original
pairs were set up under U.S.S. ~llassachusetts,
which was launched on September 23, 1941 Since

454

SHIP LAUNCHING
~./~/

,Topof Ground WQv

j , ~ ? p o4 Ground W~

Trigger
Fourfh Lever

Sa{efyPin

Th[r Lever

Trigger Frame

Scole
0

6
tZ
~nches

_FIG..14.--110-TON LAUNCHINGTRIGGI~R--SETPOSITION

Scale

,that date the same triggers have been used with

*complete success for numerous launchings, using
~)ne, two or three trigger pairs, the number depending upon the size and launching weight of
each hull. In addition, two pairs were made without change from the same plans for successive
launchings of two ship classes at other shipyards.
T h e a d v a n t a g e of the mechanical trigger for a
succession of launehings recommended the development of a type for holding smaller ~essels on
the order of destroyers and LSTs. Whereas the
larger triggers were somewhat complex in t h a t
their levers were forgings, the shafts were carefully
m o u n t e d in bronze bushings fitted for pressure
grease lubrication and the releasing mechanisms
were actuated electrically, the intent of the new
specification was simplicity. One pair of triggers was to be used per hull and released manually.
At the time, the procurement of forgings was out
of the question and elements of structural steel
were suggested. H e a d r o o m was at a 15remium,
however. T h e requiremerLt was t h a t 80 tons per
pair be held in a total working headroom of 2 feet
for the dropping of levers, yet the gravity release
feature of the first design was to be retained. This
was accomplished as outlined in Figs. 12 and 13.
T h e main lever was cut from 1-inch plate, three
sections of which were welded together for a total
thickness of about 3 inches. T h e second lever
was made a channel-shaped weldment to cradle
the first. T h e third and fourth levers were hung
from the first and second, as illustrated. All
elements were cut from plate with the acetylene
torch and rough ground as required. While this
system has the appearance of being held up b y its
bootstraps, it provides a great mechanical advantage. The smaller levers swing clear with the
larger and thus headroom is conserved as no
clearances are necessary to prevent interference of
the larger levers with the smaller lever pins.
This development was so urgent at the time

Iz

Inches
FIG. 15.--110-TON LAUNCHINC TRIOGER--RELEASED
POSITION

t h a t the first pair of triggers was in the test frame

before a decision had been made on the arrangem e n t of the releasing device. T h e test was completed b y hauling the last levers clear of their
safety pins, using a simple rope bridle and block
and fall. T h e pins were then removed and the
release effected b y letting go the tackle. The test
load was established at 80 tons per trigger or
double the m a x i m u m designed working load.
The releasing mechanism was arranged as a
through-shaft athwartships between the ways (2inch standard pipe). This was held in simple clips
"bushed" with 2 ~ - i n c h pipe. A hook was welded
in line with each trigger and each was fitted with a
"keeper and stripper" for the eye of a turnbuckle
leading through a wire rope connection to the last
lever. One hook was welded to the pipe shaft,
alined with turnbuckle and stripper on t h a t side
and held while the second hook was located similarly on the other side and welded in place, to assure simultaneous release. A hand lever was
welded to an extension of the shaft, b y which the
latter could be rotated for stripping the turnbuckles off of the hooks, thus releasing the triggers. T h e hand lever was fitted with a final safety
pin preventing its motion until desired.
These trigger levers were not bushed on their
pins nor has the need of bushings been indicated
in literally hundreds of launehings using this
trigger type. No strain gages were considered
necessary for the triggers in single pairs per
launch. In preparation for launching, each trigger man maintains sufficient tension on the turnbuckle to keep the fourth lever clear of the trigger
safety pin. This has been done generally b y
hand, no wrench being required. Upon signal he
removes the safety pin and the two pins are
brought to the official in charge at the launching

SHIP LAUNCHING
log desk. The sponsor is warned and the person
who is to release the hull is sent to the releasing
station. This m a y be an honorary function, in
which case he (or she) is accompanied by an attendant who is versed in the art and can lend his
practical and moral support. In a special event
of this kind the final safety pin is polished and
engraved as a souvenir of the occasion. This is
pulled and the hand lever is thrown to release the
triggers and the hull. On the average, about 10
pounds effort is required on the hand lever which
is about 30 inches long. This force is more dependent upon the alinement of the pipe shaft than
upon the load applied to the releasing hooks,
which is very moderate.
The third design was inspired b y the success of
the second and b y the needs of successvie launchings for the Victory ship program. The arrangem e n t of levers is illustrated in Figs. 14 and 15.
Here one pair of triggers was used per ship and the
last lever was carried aft in the same direction as
the third lever to permit the use of simple yokes
on the release shaft, which was turned b y a hand
lever to effect release. This trigger was designed
for 110 long tons load (220 tons per pair). The
procurement of forgings was also a problem at the
time of this design and the proportions of the
main and second levers were considered to be too
large for their easy fabrication from plates as
weldments. Steel castings were used therefore
for these items and they were inspected carefully
and proved under test at twice the designed working load. In addition, electrical strain gages of
the wire grid type (Baldwin-Southwark SR-4
gages) were applied to the main levers as a check
on stress and as an indication of load on the triggers when holding during launching preparations.
The construction details were kept as simple as
possible and the levers were not bushed on their
pins. Outside of more complex but convenient
electrical signaling arrangements, the procedure
for releasing these triggers is quite similar to
that used for the 40-ton triggers.
In conclusion and in general review of the foregoing, some notes in detail m a y be of interest.
The basis for determining design lo~ds was subject to some discussion. Previously an allowance
for friction usually had been made as aiding the
holding device. The use of the full down-ways
gravitational component was now advocated by us
on the premise t h a t the triggers would be subjected
to a certain a m o u n t of dynamic loading as blocks,
shores and cribbing were removed from beneath
the hull and, further, t h a t because of trigger resilience the apparent static load on triggers might
exceed the down-ways component because of
frictional resistance to trigger deflections a t t e m p t -

455

ing to push the cradle up the ways after such

loading impacts. This has been demonstrated in
fact in a n u m b e r of launchings. The a m o u n t of
overloading is not sufficient to warrant an addition to the theoretical down-ways component
considering the safety margins built into these
triggers.
The nominal factor of safety used in the designs was predicated on stresses not exceeding 30
per cent of material yield strengths based on
the down-ways load without reduction for grease
friction. Ordinary materials were used as there
was no strong point in favor of weight saving and
it was believed t h a t materials having good ductility were safer for this application than more
brittle materials of higher strength. I t is believed
also t h a t the provision of h e a v y wooden buffers
and their fitting to the outline of the levers in
the released position are important details in preventing the overstressing of the levers and their
pins when the levers are dropped.
Actually, the triggers m a y be massive and yet
remain within scale to the eye, for they are
dwarfed b y the size of cradle and hull. Their
m a x i m u m size can best be established b y the headroom available from the line of the ground ways
to the ground or trigger pit. The digging of pits
is not always practicable; for example, a pair of
triggers m a y be in a location covered b y several
inches of water at high tide ; also, the larger levers
m u s t not be so h e a v y t h a t they cannot be "set"
using simple tackle.
The levers are carried in frames assembled as
weldments from h e a v y plates and bolted to the
ground ways.
Cross members in the larger
trigger frames (170-ton and l l0-ton sizes) provide
large areas to which the ground-way timbers are
fitted for assuming loads in compression. The
larger triggers work through apertures in the
ground ways and bear on frames which are bolted
to the sliding ways. The latter assure adequate
load distribution to the sliding-way members.
The 40-ton trigger frames are bolted in recesses
in the outboard sides of the ground ways. The
sliding-way frames or pressure members for these
triggers are simple channels bolted to the sides of
the cradle. Adequate bolting is provided in their
length and no cross members are required.
In general, the mechanical triggers a p p e a r to
assure more promptness in the starting of the
launch than when other methods of holding are
used. While the time interval is short, a release
effected b y sawing logs, burning sole plates or
using sluggish hydraulic triggers usually is accompanied b y a certain amount of creeping or
dragging as the final resistance of the holding device is overcome. In comparison, the mechanical

456

SHIP LAUNCHING

trigger permits the full down-ways component of the burrs are chipped off and the surfaces are built
gravitational force to be applied almost instanta- up b y welding, laying on and grinding flush a
neously to the acceleration of the hull and cradle layer of harder metal, which is an i m p r o v e m e n t
mass and to the overcoming of "starting friction." as well as a repair.
While the load applied to the triggers is eccenAfter the triggers are set up in the ground ways
tric with respect to the ground ways, the resulting they are checked b y an experienced mechanic who
couple is readily handled by the distribution of the assembles the releasing mechanisms and makes
forces through the bolting of the trigger frame and any necessary adjustments. T h e externals are
by the weight of the hull and cradle which prevent painted red and green for port and starboard sets
the tilting of the whole assembly. The principal and otherwise kept bright and clean.
load is applied to the ground ways in compression.
Due to the small headroom available, the turnThis is appreciated especially when a launching buckle and wire rope combination was selected as
must be held up for any reason, when most or all a simple and flexible holding arrangement for the
of the hull weight is on the cradle. For example, fourth lever in the 40-ton trigger systems. These
in a t a n d e m launching a floating barrier was cast do not lend themselves to protection b y means of
adrift b y the wave motion due to the first hull buffers however and the turnbuckle screws usually
launched and was carried across the end of the take a beating. As their loading is v e r y conservaslipway b y the river current. For 20 minutes of tive, they can often be straightened and if they
the time required to clear this obstruction the sec- m u s t be replaced their cost is insignificant.
ond hull was held b y the triggers alone. In anReverting to basic design features, it is posother case a rivet hole in the shell plating was un- sible to improve the mechanical advantage of
covered upon removing the last bilge cribbing un- trigger lever systems b y the use of toggles or b y
der a vessel and nearly as much time was con- carrying the points of applied loads close to the
sumed in welding, grinding and painting the spot. lever pin centers throughout the train. The use of
The creeping of the hull under such circumstances, toggles implies some research to determin ~_ the
when held b y the forward end of the cradle, is effects of friction on the extent to which one m a y
nerve wracking, whereas the mechanical trigger go in taking advantage of the toggle action. Meinspires confidence b y relative freedom from creep chanical friction varies with the condition of the
after the loads are fully established on the triggers. trigger and the state of the weather. The simple
Observations of the triggers released b y sole- lever systems were feasible and were felt to be
noid-controlled h a m m e r s have been made b y sight- more dependable than toggles. T h e shortening
ing across individual pairs and by watching the of the "short" levers and lengthening of "long"
bank of "released" signal lamps. As nearly as the levers to apply loads close to the pin centers, in the
eye can tell, the release of all triggers is simultane- train following the main lever, requires the use of
trunnions rather than through pins. Also, the efous.
As an ultimate precaution, a crew has been fects of pin and lever bearing friction in resisting
trained to trip the solenoid latches b y hand upon release are more pronounced with greater mechania given signal. T h e men take p a r t in the trigger cal advantage per lever. An added lever is not a
rehearsals prior to the start of launching events serious complication. Through pins and a nomiand they are adept to the extent t h a t there is no nal mechanical advantage per lever were given
apparent difference in the signal lamp response preference therefore in these designs.
In setting up the sliding ways, the sections forwhether the release is effected manually or electriward of trigger locations are jacked aft into concally.
No mechanism is perfect b u t thus far the trou- tact with the triggers and with each successive
bles with these triggers have been few and con- section in order to assure the transmittal of load
cerned chiefly with maintenance. The larger to each trigger as uniformly as possible and with
triggers are bulky and in moving t h e m to the the least amount of creeping. With a single pair
slipways or to storage they m a y suffer collisions of triggers, those sections aft of the trigger locawith other objects. The resulting damage has tions are jacked aft until clearance or lost motion
been minor and no complete replacement of any in their connecting links is taken up, thus enapart has been required. T h e lever contact sur- bling these links to transmit load in tension from
faces could not be designed to eliminate all faults the after sections to the triggers. When two or
and so were left perfectly flat. In release they more pairs of triggers are used, intermediate
are subject to rolling and sliding and the loads sliding-way sections between trigger sets are
are transferred to a small area at the ends of the jacked aft into contact and additional clearance is
long levers with the result t h a t these ends be- provided in their connecting links immediately aft
come turned over and burred. After some use, of the forward triggers. To summarize, all sec-

SHIP LAUNCHING
tions forward and intermediate are loaded in compression with no load on the connecting links.
The links of sections aft of the aftermost trigger
pair are stressed in tension. The actual loads on
the triggers are predicated partly on this method of
handling the sections and partly on the general
distribution of hull weight on the cradle. As a
rule, the after triggers assume more of the total
load down the ways than do the forward sets.
At the time of these trigger developments there
was practically no comprehensive literature on t h e
subject. Recently a paper on "Launching Triggers" was presented before the N o r t h - E a s t Coast
Institution of Engineers and Shipbuilders, b y lXlr.
H. B. Robin Rowell, a Vice-President of t h a t Institution. The date on the advance copy of this
paper was 26 J a n u a r y 1945 and it was published in
the Transactions of the N o r t h - E a s t Coast Institution :;f Engineers and Shipbuilders, Volume 61,
1944-1945. I t was also published in two issues of
the magazine Shipbuilding and Shipping Record,
dated February 1,5 and March 1, 1945. I t is of

457

interest in reflecting British practice and it is

thorough in scope on the design of a mechanical
trigger quite similar in arrangement to the 170-ton
solenoid-operated trigger previously described.
ACKNOWLEDGMENT
As is usual in any development which takes
place over a number of years, the thoughts and
ideas of m a n y people form a background for the
subject and successful practice generally will be
found to embody much of what has been proved
previously. In this paper acknowledgment is
made to the m a n y sources from which basic information has been drawn, and particularly to
Professor Keith's chapter on "La'anehing" in the
textbook on "Principles of N a v a l Architecture,"
published b y the Society and also to the paper b y
Mr. J. 2yr. YleNeill on the "Launch of the Quadruple-Screw Turbine Steamer Queen Mary," published in the Transactions of the Institution of
Naval Architects for 193,5.