Preview

INTERNATIONAL LABOUR OFFICE

STUDIES AND REPORTS
Series F, Second Section (Safety) No. 3

HYDRO-EXTRACTORS
THEIR SAFE CONSTRUCTION
AND EQUIPMENT

GENEVA
1929

Published in the United Kingdom
For the INTERNATIONAL LABOUR O F F I C E ( L E A G U E OF N A T I O N S )

By P . S. K I N G & SON, Ltd.
Orcliard House, 14 Great Smith Street, Westminster, London. S.W. 1

PREFACE
The present monograph is the second of a large series of
studies that the International Labour Office proposes to publish
on individual problems in accident prevention. The first,
Safety in the Use of Chains, appeared in Volume IV, No. 2,
of the Industrial Safety Survey, and is also obtainable as a
reprint.
These studies are the outcome of international co-operation,
which is organised in the following way. The Governing Body
of the International Labour Office has instituted a Correspondence Committee on the Prevention of Accidents to work
in conjunction with the Service of the Office that deals with
accident prevention and the inspection of labour. At the
present time the undermentioned gentlemen compose the
Committee:
Sir Gerald BELLHOUSE, Chief Inspector of Factories, London.
Mr. J. BOCQUET, Director of the Accident Prevention
Association of Normandy.
Mr. BOHUSZEWICZ, Labour Inspector, Warsaw.
Mr. P. BOULIN, Divisional Labour Inspector, Lille.
Mr. L. DELADRIÈRE, Director of the Belgian Employers'
Association for the Prevention of Accidents.
Dr. FISCHER, former President of the Senate, Berlin.
Mr. M. FROIS, Labour Inspector, Paris.
Mr. R. A. GORTER, Director of the Amsterdam Safety
Museum.
Dr. LeYMANN, former Ministerial Councillor, Berlin.
Mr. MASSARELLI, Director-General of the Italian National
Association for Accident Prevention.
Mr. R. B. MORLEY, General Manager of the Industrial
Accident Prevention Associations, Toronto.
Lt.-Col. J. A. A. PICKARD, National Safety First Association,
London.
Mr. Qnni PYYKKÖ, Senior Inspector of Labour, Helsingfors.
Major Henry A. RENINGER, Allentown, Pennsylvania.

Mr. H. J. SCHÖLTE, Senior Factory Inspector, The Hague.
Dr. A. TZAUT, Director of the Swiss National Accident
Insurance Office.
Mr. VAN DE WEYER, Inspector-General of Labour, Brussels.
Mr. Shunzo YOSHISAKA, Chief of the Industrial Inspection
Service, Tokyo.
The Committee advises the Office on the choice of the problems to be dealt with in the monographs, and also proposes the
experts — who need not necessarily be members of the Committee — for the various subjects. The expert draws up the
plan on which his monograph will be based, and the plan is
sent to all the members of the Committee, who are at the same
time requested to make any proposals for amending or completing
it which they may think desirable, and to procure the material
available in their country on the subject. This material is then
sifted by the Safety Service of the International Labour Office,
completed where necessary, and transmitted to the expert, who
thereupon prepares a preliminary draft of the monograph
entrusted to him. Next the draft is scrutinised by the Safety
Service, translated into different languages and distributed to
the members of the Committee for their opinion, which may be
ascertained either by correspondence or, occasionally, orally
in the course of a meeting. The expert must take the views
of the members of the Committee into account in the final
drafting of the monograph. Should differences of opinion arise
on technical points, the Safety Service of the International
Labour Office is empowered to decide them, if necessary after
consultation with special experts. The International Labour
Office is also empowered to determine the final form of the
publication and is responsible for its contents.
At the present time the monographs mentioned below are
being prepared under this procedure:
Protective Devices for Presses.
Expert Collaborator :
Mr. FROIS.

The Prevention of Accidents in the Manufacture and Use of
Celluloid, including Cinematograph Films.
Expert
Collaborator: Dr. FISCHER.
Protective Devices for Wood-Working Machines. Expert
Collaborator: The Swiss National Accident Insurance
Office.

—

The Prevention of Accidents in the Manufacture and Use of
Acetylene. Expert Collaborator: Dr. ULRICHS, Ministerial
Councillor in the Prussian Ministry of Trade and Industry.
The monograph on Safely in the Use of Chains, which, as
has been said, has already appeared, was prepared by Mr. DELADRIÈRE. The present monograph on hydro-extractors was
prepared by Mr. MASSARELLI, assisted by Mr. SCOTTI. These
two surveys were considered and approved by the Safety
Committee at its Second Session, held on 5-7 November 1927,
in Geneva.
We hope that by means of these publications we are doing
good work in the campaign for the prevention of industrial
accidents in all countries, and we avail ourselves of this opportunity to thank our collaborators warmly for their valuable
co-operation.

CONTENTS
Page
in

PREFACE
INTRODUCTION

1
CHAPTER

General Remarks and Classification of Hydro-Extractors
§ 1. Introduction
§ 2. W h a t is a Centrifugal Hydro-Extractor ?
§ 3. Classification of Hydro-Extractors
First Group: Suspended Machines
Second Group: Machines Driven from Below
(Oscillating Drum)
Third Group: Machines with Drum between Two
Bearings
CHAPTER

Danger
§
§
§

7
8
9
10
11

III

Calculations Concerning Centrifugal Hydro-Extractors
§ 1. Centrifugal Force
§ 2. Equations of Equilibrium for Besistance
§ 3. Kinetic Energy
§ 4. Torsion of the Casing
§ 5. Riveting
§ 6. The Shaft
§ 7. Necessary Power
CHAPTER

of Accidents in the Working of Hydro-Extractors
. . . .
1. Inherent Sources of Danger
2. Incidental Sources of Danger
3. Working Conditions of Hydro-Extractors. — Quality
of the Material dealt with
CHAPTER

3
3
3
5
5

14
II
15
17
18
18
19
19

Details of Construction:
Materials Used and Methods of Joining
•considered chiefly from the Safety Standpoint
§ 1. Casing and Strengthening Rings
Copper Casing
Iron Casing
Perforation of the Drums
Circular Perforation
Oblong Perforation
Attaching the Casing to the Bottom of the Drum .
The Bottom of t h e Drum
Riveting the Bottom
Balancing Rings

21
21
21
25
27
27
28
:S0
31
35
37

— vin —
Page

§ 2. Outer Shell
Cast-Iron Shell
Sheet-Iron Shell
§ 3. Shafts ; Supporting and Adjusting Bearings
§ 4. B r a k e s : Different Types; Their. Advantages and Disadvantages
Shoe Brakes
Band Brakes
§ 5. Systems of Drive
Suspended Machines (Pendulum System)
Belt Drive
Hydraulic Drive (Pelton Wheel)
Drive by Electric Motor
Steam Turbine Drive
Machines Driven from Below (Oscillating Drum) .
Machines with the Drum between Two Bearings .
CHAPTER

44
44
46
48
48
49
49
50
50
50
52

The Safety of Hydro-Extractors in Practice
§ 1. General Considerations
§ 2. Systems of Locking the Cover
CHAPTER

37
38
38
41

55
55
55

Conclusions
§ 1. Studies and Publications
§ 2. General View of Legislation and Practice in various
countries
§ 3. Relations between the Constructors and the Manufacturers
who Employ these Machines. —• Guarantees Required
from the Constructors. — Periodical Inspections . .
§ 4. Accidents
BIBLIOGRAPHY

A P P E N D I X : Some Accidents caused by Centrifugal Hydro-Extractors
First Case
Second Case
Third Case
Fourth Case

80
80
82
87
90
92

93
93
94
99
100

INTRODUCTION

A visit to industrial undertakings can give one an idea of the
extent to which centrifugal hydro-extractors have spread at
the present day, replacing older systems of drying or being used
for the ejection of substances in modern methods of manufacture.
Most of these machines are of modern construction, manufactured by firms who have made this type of work their
speciality. Many, however, are built empirically with a neglect
of detail which cannot escape the eye of the expert and makes
him wonder whether the machines are really solid. There are,
moreover, good and bad machines and old machines often bought
second-hand and worn out, with the metal casing damaged, the
soldering defective, and the moving parts not perfectly balanced.
In certain undertakings work has been carried on for years
with hydro-extractors whose gearing or friction wheels could
not be regulated, which had no brakes, and on which the drum,
without reinforcing rings, had in several places been corroded
by acid substances. These machines were often placed in the
most dangerous parts of the room from the point of view of
workers on neighbouring machines.
Quite frequently an accident resulting from the explosion
of a hydro-extractor attracts public attention, and if there are
fatal results an enquiry is instituted to establish the responsibility. It is, however, difficult to prove that the manufacturer
was responsible for defects in the machine or that the owner of
the factory was responsible for neglecting its upkeep; in general,
all that the expert can do is to conclude that the worker (who
is generally the victim of the accident) was responsible for the
bad working of the machine, or else that no one was responsible
because the accident was due to wear and tear or to flaws resulting
from long use and invisible to the worker. If there are no
serious consequences, the enquiry is dropped.
Just as with steam boilers, hydro-extractors are subject to
internal pressure (generally not less than 3 kg. per cm.2), but
there are good reasons for considering them more dangerous

— 2 —
than steam boilers, not only from the point of view of the
probability of accidents, but also with regard to the disastrous
effects of such accidents. The hydro-extractor is not provided
with gauges and warning signals such as are usually found on a
steam boiler, and in many cases it is not worked by a specially
trained staff.
Without entering into details of the very numerous accidents
resulting from the method in which these machines are employed,
it seems desirable to include in the Appendix details of a few
characteristic accidents resulting from the explosion of hydroextractors and serving to show why so much space has been
devoted in this essay to the construction of the machine.

CHAPTER I
GENERAL REMARKS AND CLASSIFICATION
OF HYDRO-EXTRACTORS

§ 1. — Introduction
The great diffusion of centrifugal hydro-extractQrs in
various branches of industry, especially during recent years,
has increased the probability of accidents and made it necessary
to correlate all research carried out for the elimination as far
as possible, or at least the reduction to a minimum, of the causes
of these serious accidents, both as regards the constructors and
the users of the machines.
The present monograph, based on information collected from
various sources, aims at making known the dangers involved
in the working of hydro-extractors as well as the protective
regulations and devices suggested by long experience.
Before touching the main subject of this monograph, however, it is well to mention certain general ideas regarding hydroextractors and their classification.
§ 2. — What is a Centrifugal Hydro-Extractor?
The centrifugal hydro-extractor consists essentially of a
drum A, formed of a sheet metal cylinder, generally perforated
and revolving round a vertical axis (fig. 1).
Under the action of centrifugal force developed during
rotation, the liquid content of. the substance contained in the
drum (the load to be dried by centrifugal action) is ejected
through the holes in the drum. All the other parts of the
machine are accessories helping to ensure that the centrifugal
action takes place under the best possible conditions.
Thus the shaft B acts as an intermediary between the motive
power and the drum.

— 4 —

Fin. 1

Ú
Fig. 2

•u
Fis. 3

— 5 —
The outer shell C contains the drum and prevents the liquid
which is ejected from flying out, guiding it instead towards the
waste-pipe; it also acts as a safety device.
The brake F enables the revolving parts to be stopped either
when the operation is completed or if the load loses its balance
and the machine has to be stopped in order to prevent dangerously large oscillations.
Finally, the frame D supports and connects all the parts of
the centrifugal machine, as is clearly shown in fig. 1.

§ 3. — Classification of Hydro-Extractors
There are naturally various types of hydro-extractors which
may be grouped as follows :
(1) Machines suspended from, a single bearing (supporting
and adjustable ball bearing) like a pendulum, the bearing disc
being attached to the upper end of the shaft which carries the
drum (fig. 2).
(2) Machines with oscillating drum (driven from below) with
two bearings under the drum or, in exceptional cases, a single
bearing (fig. 1).
(3) Machines with drum between two bearings (driven from
above or below) (fig. 3).
In the following paragraphs the essential particulars for
each type of these machines are given. Together with the
foregoing figures, they will afford to the reader clear and exact
indications respecting these useful and widespread machines.

FIRST GROUP : SUSPENDED MACHINES (Fig.

Drive :
(a) Hydraulic or electric motor directly driving the shaft
of the machine.
(ò) Belt running from the main shaft with fast and loose
pulley or friction clutch or, in rare cases, belt drive from a
countershaft on the ground with electric motor.

— 6 —

Drums :
Sheet copper or iron casing with or without strengthening
rings, always perforated and sometimes coated with ebonite.
Diameter from 800 to 1,300 mm., generally 1,000 or 1,200.
Spheres of Use:
Employed almost entirely in sugar refineries and chemical
industries, more rarely in wool-spinning works.
Brake :
Band brake on a pulley on the drum shaft slightly under the
point of suspension and worked by a wheel or lever.
SECOND GROUP: MACHINES DRIVEN FROM BELOW
(OSCILLATING DRUM) (Fig. 1)

The machines of the second group (with oscillating drum)
are the most numerous and most widely used in industry.
Drive and Characteristics:
Two bearings under the drum, the upper one being either
rigid or elastic, while the lower one is always rigid.
(a) Direct motor drive on the drum shaft;
(b) Belt and pulley direct from motor or from a countershaft
on the ground, with fast and loose pulley;
(c) Conical wheel gearing or friction cones.
Drums and Spheres of Use:
(a) Drums with copper casing, with or without iron or copper
strengthening rings (used in numerous industries,
especially chemical works and dyeworks).
(b) Drums with iron casing, with or without iron or copper
strengthening rings (chemical factories, dyeworks, clothprinting works and dairies).
(c) Drums with iron or copper case, with or without strengthening rings, and coated with ebonite, galvanite or
some similar substance (in chemical factories and dyeworks using liquids or materials with strong corrosive
action).

7—

Type (c) may be subdivided into two groups:
(1) Thin casing (5 to 7 mm.), with strengthening rings;
(2) Thicker casing (12 to 15 mm.), with or with out
strengthening rings, diameters from 600 to 1,200 mm.,
normally 1,000 and in exceptional cases under 660
or over 1,200 mm.
(d) Special non-perforated drums, used in dairies (separators)
and starch factories (for separating oils and fats from
the washing water by difference of density).
Brake :
Generally band or shoe brake working on a, pulley on the
driving shaft and controlled by a hand or foot lever.

THIRD GROUP : MACHINES WITH DRUM BETWEEN
Two BEARINGS (Fig. 3)

These machines are not common and are generally of an old
type.
Drive :
(a) Conical friction wheels and steam engine.
(b) Belt drive and countershaft.
(c) In rare cases electric motor with direct drive on the shaft
of the machine.
Drums :
As under letters (a) and (£>) of the second group.
Brake :
As in the first group in general.
Sphere of Application :
Chiefly dyeworks.

CHAPTER II
DANGER OF ACCIDENTS IN THE WORKING
OF HYDRO-EXTRACTORS

The preceding chapter has given us a clear idea of the
construction and working of a centrifugal hydro-extractor.
We may therefore broach the main subject and examine the
danger of accidents inherent in or incidental to the working
of these machines.
§ 1. — Inherent Sources of Danger
The most important and most obvious source of danger is the
bursting of the drum. This accident occurs under the action
of centrifugal force because of the weakness of the material of
which the drum is made (casing and strengthening rings, if any).
This weakness may be relative or absolute.
Absolute weakness is found when the machine is used at a
normal speed and with a normal load (e.g. those stated by the
manufacturer), but, as a result of long use or of the influence of
corrosive substances, has been reduced to such a state of
weakness, especially where joints occur, that it can no longer
bear the pressure for which it was built.
In a later chapter dealing with the calculation of forces
it will be seen what forces come into play and what their
importance is.
The relative weakness of the drum is shown in the case of a
drum which may be in very good condition but which, through
error or ignorance, is made to work at a speed or with a load
greater than that for which it was built.
In connection with relative weakness, it is not unimportant
to recall that the increase in speed to which a machine is
subjected may result from accidental causes and not from any
mistake. In the first place, such increase in speed will be due

— 9 —
to the irregularity of many kinds of motive power. Such irregularities may be expected, when the centrifugal machine is driven
by a steam engine, either directly (which is very rare to-day)
or indirectly (very frequently in sugar factories), or by a steam
turbine or by a direct current electric motor (series or compound
excitation). With steam engines an unforeseen increase in
speed may arise from various causes: (a) faulty working of the
speed regulator; (b) increase of pressure in the boiler; (c) sudden
increase of pressure in the feed pipes through the stoppage
of a large number of the machines driven by the steam. The
trouble is much greater if two or three of these causes act at
the same time. The same remarks apply also to small steam
turbines (impulse wheels with two or three speeds) directly
driving the hydro-extractor.
In hydro-extractors driven by a direct current electric motor
(series or compound excitation) the increase in speed may be
due to a sudden weakening of the induction field as a result of
disturbances in the windings. The tendency to-day is to replace
such direct current motors by three-phase alternating current
motors, fitted with appliances for gradually increasing the speed.

The other inherent sources of danger must now be dealt with.
(ö) Excessively wide oscillations. — These can occur only in
suspended machines or in machines with an oscillating drum
where the upper bearing is elastic, and they appear whenever the
machine is badly balanced either owing to its construction or
because the load is irregularly arranged or is not homogeneous.
The first case (bad arrangement of the load) occurs chiefly in the
manufacture of stuffs or of yarns, and the second case in chemical
factories and sugar refineries.
The most frequent result of excessive oscillations not
restricted in time by the action of the brake is that the drum
strikes against the outside covering of the machine, possibly
involving serious damage to both parts and even breaking the
shaft at the point where the nave is inserted. This latter
accident happens always when the load is very badly arranged
in hydro-extractors with rigid bearings, because the latter sharply
oppose any irregular movement of the drum. The breaking
of the shaft does not always merely mean the stopping of the
2

— 10 —
machine or the falling of the drum to the bottom of the outer
shell. It frequently happens, especially when the accident
occurs at full speed, that the drum is hurled outwards, involving
serious danger for the persons working the machine.
(b) Breaking of the shaft and the nave may also take place as
a result of too sudden application of the brake; this naturally
occurs when the brake pulley is fixed on the shaft, which thus
acts as an intermediary between the brake and the loaded drum
which has to be stopped.
These sources of accidents will be examined in more detail
in the discussion of the principles for calculation given later.
§ 2. — Incidental Sources of Danger
As is shown by
are those resulting
machine, but arising
the working parts or
the workers.

this title, the dangers referred to here
from causes outside the action of the
either from lack of sufficient protection of
from negligence or rashness on the part of

(1) Falling objects. — The first source of danger is the falling
of heavy objects, even small ones, into the drum while it is in
motion. The object which falls (any screw or nail) immediately
moves to the wall of the drum and, unless any exceptional
circumstances take effect (e.g. a load such as sugar which can
absorb the object which has fallen), it flies off at a tangent with
a speed of approximately thirty to fifty metres per second. Or
under more favourable circumstances, if the object is caught
by the edge of the drum it makes the drum oscillate excessively
and the results may be serious.
It is not necessary to insist on the effect which even a small
object moving at the speed mentioned above can have if it
strikes a person who happens to be in the neighbourhood of the
centrifugal machine.
(2) Another cause of serious accidents is to be found when
inexperienced workers, ignorant of the danger, imagine that they
can improve the balance of the load without stopping the machine, while the latter is revolving at a high velocity. Frequently the shovel is torn from the hands of the worker, or, in
more serious cases, which occur during the drying of stuffs or of

— 11 —
yarns, the hand remains caught in the load and the worker may
lose his arm or even be dragged bodily into the rotating drum.
The attempt to increase or decrease the speed by applying
the hand to the rim of the drum may give rise to accidents of the
same type as those mentioned above.
It is also possible for a worker to fall into the drum for
several reasons. For instance, the machine may be fitted up
in a pit and may work, as is nearly always the case, without any
cover. Under these circumstances it may easily happen that
a worker who leans over the machine for any purpose, loses his
balance, and falls into the rotating drum.
3. Less serious but much mo-re frequent accidents are caused
by the accessory parts of the hydro-extractors:
(a) Belts and pulleys of the transmission machinery.
(b) In the case of electric drive, the possibility of- short
circuits or of contact with metallic apparatus suddenly
subject to tension.
(c) Chemically dangerous substances which in the centrifugal
action give off poisonous gases (e.g. aniline salts and
nitro-derivatives in general).
In a later chapter the various protective fittings which have
been adopted will be discussed. At present it is merely sought
to show the dangers occurring during the employment of hydroextractors.
§ 3. — Working Conditions of Hydro-Extractors —
Quality of the Material Dealt With
The quality of the material which is being handled may have
an important influence on the working of the machines. Thus
hydro-extractors for dyeing and bleaching which have to deal
with stuffs or yarns of a weight generally under 40 or 50 kg.
(frequently considerably less) may be very much lighter than
those in a sugar factory or refinery which have to deal with
cooked masses always weighing more than a hundredweight
and even as much as three or four hundredweights. Passing
from the question of mass to the quality of the substance dealt
with, it will be found that this has an influence not only on the

— 12 —
power of the machine but also on the quality of the materials
which compose it. In dyeworks and laundries, copper drums
are chiefly used, as these have the great advantage, when no
corrosive liquids are employed, of not dirtying or damaging the
stuffs which are being dried. The light iron drums which are
much less frequently used are generally coated in the inside.
Under these conditions the life of the drums is very long, and
even after several years they offer a maximum of safety.
It frequently happens, however, that the substances handled
give off acid liquids which have a corrosive effect on the metal.
In such a case, iron must almost always be excluded; copper
may be employed but is liable to very rapid corrosion. One
may take, as an example, the separation of lactose from whey
(treatment of by-products of milk) and in the chemical industry
the drying of naphthol and copper sulphate. In both cases the
copper drums (the only metal which can be employed) are
attacked so rapidly that after eight months or a year the metal
has been reduced to half its original thickness. This example
is sufficient to show the dangers which exist if the condition
of the drums is not carefully examined at regular intervals.
It is becoming more and more common to apply a coating
of ebonite or galvanite (hard rubber) to the drums, strengthening
rings and the bottom of the machines which are to be used with
corrosive matter. The framework of the drum can thus be
made of iron, which offers a great advantage on account of its
mechanical resistance to centrifugal force. A further advantage
is the longer life of the drums, but the system also shows a
certain danger. As a result of the cracking of the ebonite
coating (usually from 2 to 3 mm. thick) or even by a break in the
latter, the corrosive liquid may penetrate between the ebonite
and the metal, producing corrosions which are all the more
serious and dangerous because they remain hidden. Particularly close supervision must therefore be given to coated
drums. It is rarer for drums to be completely plated or to be
' of aluminium. In the latter case they should be employed at a
reduced speed and with a light load, in view of the low specific
resistance of the metal.
The load, apart from its chemical composition and its weight,
may, however, also have an influence on the safety of the hydroextractor as a result of its lack of homogeneity. A single example
may be taken from the sugar industry, where — especially

— 13 —
in dealing with the cooked masses of the second product
(raw sugar) — the equilibrium of the machines may be considerably disturbed because the viscosity of the cooked masses
is not uniform. This disturbance of the equilibrium results
in extensive oscillations of the drum with the consequences
mentioned above.
We have already seen the dangers resulting from a bad
distribution of the load. It may be recalled simply that while
slight disturbances may in the majority of cases pass unnoticed
and have no appreciable effects, this is due to the self-centring
action of the drums, which takes effect automatically at a
given angular velocity. The well-known system of working
of the spindle moulder sufficiently exemplifies this phenomenon.
This would be the place to mention the critical velocity, but
the character of this monograph — which aims chiefly at
studying the dangers in the use of hydro-extractors and their
prevention — makes it necessary to restrict the calculations
concerning these machines to the unavoidable minimum.
In concluding this chapter it may be recalled that the working
conditions are also frequently affected by the accessory parts.
Soft or moist transmission belts, badly fitted countershafts,
unsuitable or neglected brakes, twisted or badly regulated shafts,
may all result in bad working of the hydro-extractor and consequently in danger to the workers.

CHAPTER III
CALCULATIONS CONCERNING CENTRIFUGAL
HYDRO-EXTRACTORS

In view of the aim of this monograph, reference will be made
in this chapter merely to the principles and elements entering
into the calculations of the chief parts of a hydro-extractor.

§ 1. — Centrifugal Force
When a centrifugal hydro-extractor is at work the rotating
masses are made up of numerous concentric rings, differing in
section, but having all the same axis of rotation A-B (fig. 4).
Fig. 4 shows diagramatically the drum of a hydro-extractor
formed of the casing, strengthening rings, the bottom and the
load (shaded part), which has attained its position of equilibrium.
Let:
a = coefficient of loss of area of the drum resulting from
the perforation of the metal.
h = interior height of the casing.
s = thickness of the metal of the casing. .
•
a = number of strengthening rings.
co = cross section of one of the rings.
d = density of the material of the casing.
d' = density of the material of the strengthening rings.
The general expression for the centrifugal force of a mass
M :F =
, applied to all the elements forming the drum,
after all the recessary substitutions and reductions have been
carried out, gives the final expressions
(1)

F = i 7 ' + F" where

(2) F'

(centrifugal force of the casing and the rings)
0.65 y2 (a h s d + a d') and

F" (centrifugal force of the load) = D« • 0.043 ~ •

Tj^jtfi

or after simplification
F = i>« (0.65 (a h s d + a d') + 0.043 ^

• jgl^j)-

§ 2. — Equations of Equilibrium for Resistance
The parts of the drum which resist the internal pressure
determined by the centrifugal force are the metal of the casing
and the rings (if any). There are:
(a) The internal surface of the casing = 2n Rh (cm2)
F
(5) The specific internal pressure / = ^—rr
(in kg.
2 it Rh

per cm 2 ).
This last expression has more value in practice with reference
to supervision of the machine (determining the centrifugal stress)
than from the point of view of construction, for which it is
necessary to know the stress on the metal casing and the
strengthening rings. The force which seeks to make the casing
burst on a diametral plane passing through the axis is therefore
given by the expression : / h 2 R = —. And for the resistance
the following equation of equilibrium will be found:
(3)

^=2(shyk

+ amk%

where k and k' are the coefficients of resistance (safe loading
stress) of the materials which form the casing and the rings, and
y is the coefficient allowing for the weakness resulting from the
perforations and corresponding to the ratio between the effective
section and the complete assumed section along a generatrix of
the cylinder passing through the centre of a vertical line of
holes.
After having obtained from (3) the value of k and k', it will
be seen whether they are == those employed in good constructive
practice, and the thickness of the casing and the section of the
rings will be increased or decreased as is necessary.

— 16

î : itNv-

Fig. 5

i.sM

Fig. 6

§ 3. — Kinetic Energy
In calculations concerning a hydro-extractor the kinetic
energy is of extreme importance for the calculation of thebrake, the shaft, and the motor power required.
The kinetic energy possessed by a machine in rotation is the
sum of the kinetic energy of each of its elements. Each of these
forces may be expressed by the basic formula :
w
W -

—

** " a 7l d (R* — R *)
1800 g

-"i >•

This formula applies to a cylinder of radius R and 2?ls
height h, density d, and revolving at a velocity of n revolutions
per minute (fig. 5).
The total kinetic energy W having been obtained, it is easy
to determine the brake, which is capable of stopping the machine
within a given time. If we call
;• = the radius of the brake pulley,
P j = the tangential force of friction applied to the periphery
of the pulley,
P 2 = the tangential force resulting from the moment of
friction of the bearings and reduced to the arm r,
S

= the distance traversed by a given point of the periphery of the pulley from the moment when the brake
begins to act until the machine is stopped completely,

we will have as the basic equation for the kinetic energy:

= (Pi + ft) S.

Since the movement of the drum may be considered as being
retarded at a uniform rate and the action of the brake lasts t
seconds while the initial velocity
v = —TTT;— (metres per second) and S = —¿^
it follows that -y = (P1 + P2) •
and consequently
=

W130_
ir nrt

^-/,

P 2

^ '»

18 —

We shall employ this formula in calculating the factors
of the brake.
§ 4. — Torsion of the Casing
The force of torsion to which the casing is subjected under
the double action of inertia from the load and the weight of
the casing, on the one hand, and the action of the brake on the
other, is also of great importance, and although this action
usually involves only a small stress, it may sometimes, if a
powerful brake is suddenly applied, have serious consequences.
It will be sufficient, therefore, to give the final expression
for the stress on the casing as a function of the tangential force:
0

- ^

where Mt = r (Pi + P%) and the internal and external diameters
of the casing are Dz and D.
Q2
Df
Kt

= 0.392

( D 2 4 — D*)Q

where ß is a coefficient expressing the weakness due to the
presence of the perforations.

§ 5. — Riveting
With reference to the riveting of the casing and the bottom,,
it is sufficient to consider the total section of the rivets and the
tangential force which tends to cut them. If Z is the number
of rivets and í their diameter, the total section will be equal to
Z • ~ , and the tangential force being, as we have seen above,
equal to (Px + P2) . ~, the stress will be

(4)

r -

-JfJx4

It must be noted that Kr must not exceed two-thirds of
the stress corresponding to the tension obtained from (4); the
total minimum surface has to be divided between the given
number Z of rivets.
It may also be noted here that, in general, riveting carried
out according to the regulations of good constructive practice

— 19 —
gives a resisting section much greater than that given by
equation (4). Ordinary riveting, therefore, is more than
sufficient for resisting the tendency to cut the rivets.
§ 6. — The Shaft
The shaft is also of great importance in hydro-extractors
because it supports, the drum and is subject to all the stress
imposed on it by the latter. The drum is subject to stress :
(a) from flexion (tension of the belt);
(b) from compression (weight of the drum and the load);
(c) from torsion (application of the brake).
The stress due to flexion and compression is generally
negligible in comparison with that resulting from torsion. On
that account, mention will be made here only of the calculations
concerning torsion.
The moment of torsion, which is determined, as we have seen,
by the opposing action of inertia and of the brake, may be
expressed as follows:
M, - (Pi + P 2 ) /' = 0.196 d3 Kr
where Pi and Pi are known tangential forces, r the radius of
the brake pulley, d the diameter of the shaft, and Kr the stress
of torsion, then

•K

(fi + Pur

0.196 d3 '

These stresses may be quite extensive, rising to two or three
kg. per sq. mm. with very powerful brakes.
§ 7. — Necessary Power
In order to fix with sufficient precision the power needed by
the machine, the data of the problem must be clearly stated.
We know the kinetic energy and we know the maximum velocity.
We know, although only approximately, the passive resistance,
or, at least, we can estimate it as a certain fraction (generally
one-tenth) of the kinetic energy.
It is then necessary to determine the time for putting the
machine in motion according to the data given when dealing with

— 20 —

the water leaving the drum, taking, for example, a minimum
(say, 60 seconds) for substances and drums with which drying
takes place easily or where the velocity is not very great and
arriving at double or even three times that figure for substances
which are more difficult to dry or drums having a greater
velocity.
It is evident that the maximum power absorbed by putting
the machine in motion will be greater than that required to keep
the machine moving at its normal speed when it is only
necessary to overcome the passive resistance.
Experience shows us that it is sufficient to increase the
normal power by 50 per cent, in order to obtain the maximum
necessary power.
We may therefore lay down the following expression equating
the motive power and the power absorbed by the machine.
•x (1.5 Ne + Ne) 75 / = -~- + passive resistance,
where Ne is the power required expressed in horse power
(cf. diagram, fig. 6).
The expression may be further simplified by making the
passive resistance equal to j» W. We then get
| (1.5 Ne + Ne) 75 t = W g- + ¿ ) .
and the end formula becomes :
Ne

_J^W_ '
2.5 • 75 • t'

which we may take as being the final expression.
It will be seen that the power required decreases as the time
required for putting the machine in motion increases (as might
easily be foreseen), and this allows the constructor and the
manufacturer a large margin in deciding what system of drive
is necessary for use with the hydro-extractor.

CHAPTER IV
DETAILS OF CONSTRUCTION: MATERIALS USED AND
METHODS OF JOINING CONSIDERED CHIEFLY
FROM THE SAFETY STANDPOINT

§ 1. — Casing and Strengthening Rings
The casing is generally of sheet copper or iron, and perforated.
COPPER CASING

Copper is generally used for light machines employed in
drying stuffs and yarn (wool spinning works, dyeworks, laundries) or in special industries (chemical and similar industries).
The sheet metal is rolled into a cylinder (fig. 7), and the edges
are joined by different means. The most common is brazing
with a dovetail joint (fig. 8). This system is widely employed
because, in most cases, it gives excellent results. A short
examination is sufficient to show its advantages. The brazing,
which is the most sensible and the strongest method of welding
sheets of copper, is reinforced by the dovetailing of the two edges
of the metal, and thus it is much more difficult for these edges
to be forced apart by centrifugal action. In certain cases,
however, this system of joining the metal does not give sufficient
resistance because in many industries (chemical industry, wool
spinning, wool carbonising works, etc.) the mother liquids of
the substances to be dealt with are so acid that they rapidly
corrode not only the copper of the drums, but also, and more
particularly, the joints. The result is that after a certain period
(a few months are often sufficient), the joint is completely
open, and the casing is kept in place, if at all, only by the
strengthening rings.

- 22

Fig. 8

Fig. 7

m
i^
Fig. 9

Fig. 11

Fig. 10

«—«

Fig. 12

— 23 —
The same remarks apply naturally also to these rings,
although with copper rings the brazing is often supplemented
by rivets which are usually placed as in fig. 9. Riveting,
however, has the disadvantage of weakening the ring to a noticeable extent by reducing its cross section.
Experience shows that in special industries using acid liquids,
the joining of the casing and the strengthening rings must be
made the subject of special care.
The seam of the casing should be dovetailed and strengthened
by butt-straps or better still, by double butt-straps, because
riveting (fig. 10) weakens the metal which, for the efficient
working of the machine, must be perforated in the part covered
by the butt-joints to the same extent as on the rest of its
surface. It is also necessary to fit to the other side of the
casing a counterpoise to re-establish the equilibrium of forces
round the axis of rotation (fig. 11).
For strengthening rings, still greater security is obtained
by the system of double butt-joints shown in fig. 12, in which
the brazing is done away with (being useless after a short period
of working) and the parts are joined by a double butt-joint whose
total cross section is equal to that of the ring.
In this case also, it is necessary to balance the rings, and this
balance is obtained by arranging the joints at angles of 180° or
120° from each other according to whether there are two or
three rings. These butt-joints must be sufficiently broad to
take riveting whose total effective sections are equal to the
effective section of the ring (figs. 13 and 13a).
In dealing with copper, it must be kept in mind that the
rivet acts only at the surface, and any slight resistance which
it may have as a result of friction between the ring and the buttjoint must be neglected. This system of joining ensures a good
joint both for the casing and for the rings, but it has the disadvantage of making the drum heavier and of inducing forces
which, although symmetrical with reference to the axis of
rotation, bring about an irregular distribution of the internal
stress, so that in practice recourse is generally had to a more
simple, but rather less safe, system.
The metal of the casing is made to overlap slightly and is
riveted along the generatrix of the seam (fig. 14) or the system
shown in fig. 9 is used.

— 24

Fia.. 13 a

Fig. 13

0V<

Fis. 14

Fig. 15

Fig. 16

It is obvious that it is well to supplement the join by brazing
the edges, and this has at least the effect of retarding the
corrosive action of any acid between the two metal parts and on
the rivets.
This intermediate solution has also other advantages, among
which may be mentioned that of allowing the perforations to be
arranged according to a regular system, and of making the
internal surface smooth, without hollows or projections, which
might damage the material handled (yarn or stuffs).

IRON CASING

The method of joining generally employed, especially with
casing of a certain thickness, is autogenous welding which, in
well-constructed machines, is often made invisible by turning.
This system is employed more particularly in very heavy
machines, say, in sugar factories, where they revolve at a high
speed and must be very carefully balanced. The welding must
naturally be made with the greatest care because the breaking
of the drum or of one of the rings would have very serious
consequences.
In those heavy machines which have no strengthening rings,
the drum is sometimes not welded, but made out of one piece
of pressed and forged steel.
In drums where the resistance to centrifugal pressure is
largely offered by the strengthening rings, the riveted butt-.
joint with partial overlapping of the metal is widely used
(fig. 15). The riveting of sheet iron offers, of course, much
more resistance than that of sheet copper.
In these types of machines the rings are generally welded,
and if turning is carried out it does not always hide the welding.
In order to fit the rings without weakening the edges of the
metal sheeting excessively, the joint of the casing is often
constructed as shown in fig. 16. In this connection it must
not be forgotten that the strengthening rings should be
shrunk on under a slight tension, so that once they are
fitted a good part of the stress may be borne by them when
centrifugal action begins, and thus the strain on the casing is
lessened. For this purpose the heads of the rivets, at least on
the external surface of the drum, must be left projecting only

— 26

Fig. 17

Fig. 18

4 4 4" • f 4
(*
«• 4 4 4 4 4
4- <!>- 4 4 4
*
4
4 4 4
\ *4
-è- -* 4 4 4
* 4 4 4- 4
*
• ^

4
4 4-

4 44- 44 44- 4Fig. 20

Fig. 19

7
n

•

i .

Fig. 21

— 27 —

a very short distance and, if the thickness of the metal permits
it, it is well to countersink the rivet and make it level with the
surface of the casing (fig. 17).
In drums of a certain thickness, 8, 10, and 12 mm. without
strengthening rings, a very common and successful method
is to use a butt-joint or, better still, double butt-joint with
balancing weight (figs. 10, 11 and 18). This method has,
however, the same disadvantages as have been mentioned
above in connection with copper drums: increase in the
weight of the machine and irregular distribution of forces
and stresses.
In lighter drums employed for small loads (generally in
dye works and laundries), autogenous welding has frequently
been adopted in recent years, and this gives a great guarantee
of safety, provided it is carried out in regulation fashion by
a skilful specialised worker.

PERFORATION

OF T H E D R U M S

Circular Perforation
The perforations are generally circular, whether in iron or
copper drums; the holes, which take up *•/&• or 1 /io or even less
of the surface of the casing, are arranged along the generatrix
of the casing in parallel rows (fig. 19).
In order that the perforations may not weaken the resistance
of the metal any more than is necessary, the quincuncial arrangement is frequently adopted (fig. 20). By this method the dead
vertical section (which oners no resistance to stress) is reduced
by one-half for any given filtrating surface.
It is only in rare cases that the holes are distributed uniformly
over the whole surface of the casing. They must indeed be
suppressed along all the horizontal parts occupied by the
reinforcing rings, either because the rings would block the
opening, or because, if the holes were placed too near the
rings, they would cause the latter to wear down too soon under
the action of the liquid which is ejected.
It is also well to reduce the number of holes or even to
have none at all in the row along the join between the drum

— 28 —
and the bottom of the machine, seeing that this part is always
liable to wear out more quickly 1.
In general the number of holes is less in the neighbourhood of the double butt-joint and the counterpoise, because
of the presence of rivets.
This fact increases the irregular distribution of the internal
stresses of the drum as has been seen above 2, and as a result
the use of double butt-joints, especially with a counterpoise,
is not common in spite of being safer.
Finally, the holes are less numerous (one vertical row having
been left out) along the line of the welding, so as not to weaken
the drum excessively at a point of less resistance.

Oblong Perforation
The system of oblong perforation which is frequently met
with in the drums must also be arranged in rectangular or
oblique rows.
It can indeed be easily understood that for a given perforated surface (as compared with circular perforation) the
cross section resisting the action of centrifugal force will be
increased (fig. 21). This section is obtained along a vertical
plane passing through the axis of the cylinder.
The horizontal section, on the other hand, which resists
the torsion of the casing is decreased. The force of torsion is,
in fact, unimportant seeing that the stress which it causes is
generally much weaker than the centrifugal stress.
Another advantage is that for any given section of resistance
to centrifugal action the ratio of the perforated to the imperforated part is much greater, and therefore the load can
be much more quickly dried. The necessity which sometimes
arises for drying substances as rapidly as possible has led to
the adoption of drums in which the proportion of perforations
sometimes reaches or even exceeds 50 per cent. This has been
done by using drums with a casing of iron or copper wire twisted
1
The authors have observed this in numerous machines examined
by them. The fact is perhaps due to the action of the liquid eddying
in the corners.
2
Where the number of holes is less, the liquid cannot escape so
easily, and the static pressure on the wall of the drum increases.

— 29 —
^

,\._
L
-x
V \

/
'

/ \
A

/ \

\
\

I
/
I I

f
./
• i
X
Ì ../

•

/ —r
11

\
\

7
7-1
•1—¿.

/•

/ I
T
ì Z2Z

T^~7

\ *v /

Ò-

Fig. 22

WH^T
Fig. 23

iss)
Fie. 24

Fig. 25

4x
Fig. 27

— 30 —
spirally and kept rigid by cross pieces fixed to the top and
bottom (fig. 22).
These very light drums are used almost exclusively for
cloth (dyeworks and laundries) and the work at a low speed
(400-500 revolutions) and with small loads (80-100 kg.).
In conclusion, a few words may be said on the methods
employed for making the perforations. The quickest and cheapest
system is punching; the most expensive is boring. The first
method, however, has the disadvantage of straining the sheet
iron.
The sudden action of the punch generally, indeed, affects
not only the hole but the adjacent part (fig. 23), which is
frequently found to have small cracks radiating from the
hole. In this case the co-efficients of reduction used in calculating
the power of resistance must be kept very low, for it would
be rash to consider the area thus affected as having the same
resistance as the rest of the metal.
The punching of copper does not show the same disadvantages on account of the greater malleability of this metal, and
the slight decrease in resisting power which is shown in the
area round the holes is made up for by the absence of any
sharp edges on the inside of the metal in contact with the stuns
or yarn which must be protected (fig. 24). The lighter the metal
the more suitable is punching.

ATTACHING THE CASING TO THE BOTTOM OF THE DRUM

The methods of attaching the casing to the bottom of
the drum are very numerous. Mention will be made here
only of the most common, and an effort will be made to
indicate the advantages and defects which in our opinion
are shown by each method.
One of the commonest systems is that shown in figure 25.
The metal casing is fixed to the bottom by means of a riveted
angle iron; an outer iron'ring helps to keep the whole rigid
and strong. In some cases the iron ring is left out of this
type.
This system is very often used in light hydro-extractors
with a copper casing or more rarely with a casing of light
sheet iron (in laundries, dye works, small chemical works, etc.).

— 31 —
The same system is adopted with slight modifications in
riveting (countersunk rivets in the outer ring) for heavier
hydro-extractors with a sheet iron casing (Daneck, Rudolph)
employed in sugar factories (fig. 26).
Countersunk rivets are used for two reasons: the drums
are very often turned (the rings always are) and the strengthening ring is frequently used as a brake pulley.
The whole machine is very strong and rigid. The rivets
are easily fixed. Repairs and maintenance are very simple
(changing the rivets and cleaning). But there are also disadvantages the chief of which is- certainly the fact that the
head of the rivet wears more or less quickly but particularly
in that part which corresponds to the direction in which the
drum moves (fig. 27).
Another disadvantage, less evident but more dangerous,
is the possibility of liquid getting in between the casing and
the bottom plate (fig. 28). If the liquid is acid this may result
in internal corrosion between the casing and the angle piece,
between the casing and the bottom plate and, finally, between
the latter and the angle piece. Such corrosion is all the more
dangerous because it is invisible and may seriously endanger
the joints of the drum before anyone has noticed its existence.
The most usual method of avoiding this disadvantage is the
tinning of the parts exposed (fig. 29).
This tin-plating is sometimes carried out on the whole
inner surface of the casing if the latter is of iron, and also on
the bottom if it is not already covered by a thin sheet of copper
(fig. 30). In drums with copper casing this coating is usually
restricted to the bottom and to the nave (if any). A system
of fixing which tends to do away with the danger of liquid
getting into the joints is that shown on fig. 31. It is largely
used on light drums with the modifications shown in figs. 32
and 33. The method is to have the bottom flanged and riveted
directly to the lower part of the casing with or without strengthening angle irons. It is, of course, well to plate the points
marked by the arrows (figs. 31, 32 and 33).
T H E BOTTOM OF THE DRUM

The bottom of the drum may be of many different types,
according to the purpose for which the machine is to be used

— 32

sss ^\\\\\\\\v\\v\\\\v\^^^^
Fig. 29

Fig. 28

\s
Fig. 30

m^msw
WM/MM/Sm.
Fig. 31

l?
\\\\VV\VvVV^ ^

Fig. 32

^m^\m^^^^
Fig. 33

— 33 —
and according to the type of the machine. In this connection
it is well to remember the classification adopted above so that
the same order may be kept.
In the first class (suspended machines with overhead drive),
the classical type of which is the Weston, we find a double
tapering end with an outlet funnel (fig. 34). A light sheet
metal cap (shaded in the diagram) prevents the material from
being discharged during centrifugal action 1.
Returning to fig. 34, this type gives rise to the following
considerations. Its usefulness for emptying has already been
seen. It is much easier for the worker to push the substance
which has been dried towards the bottom of the drum with a
shovel than to lift it out from the top, because the load is
generally from 2 to 3 or even 5 to 6 cwt.
As regards fitting and safety, it may be noted that the
riveting of the bottom to the nave is subject to great stress
in these machines in the neighbourhood of the axis 2, more
especially because the brake is fixed to the shaft and its action
is nearly always violent 3 . The nave is also subject to considerable stress if the brake is applied suddenly, so that it sometimes
breaks and has to be replaced by a steel one or strengthened
by means of steel rings. This disadvantage is naturally increased
by the inertia of the rotating mass which may be as much as
500,000 kgm.
In the second class (machines driven from below) with
fixed upper and lower bearings, the bottom is usually flat near
the edges (and fixed to the drum as is shown in figs. 29 and 33),
and rises in conical shape towards the centre (fig. 35). The bottom
is often made of a single piece of cast iron fixed directly to the
casing (fig. 36). There are no special remarks to be made with
regard to its safety beyond those already mentioned with reference to the fixing of the casing.
In the other classes of machines driven from below there is
nothing special to be noted differentiating them from the ones
1
It will be observed t h a t emptying through the bottom of the drum,
as is done in sugar factories and with certain chemical products, will
also be found exceptional in the class of machines which are driven
from below (Rudolph), b u t is usually arranged in another way, as will
be seen later.
2
The tangential force which tends to cut through the rivets varies
inversely
with the distance from the axis of rotation.
3
In hydro-extractors which were examined by the authors, stresses
more than ten times those imposed on the rivets in the bottom of other
machines were observed.

34 —

Fig. 34

Fig. 35

Qp
Q
Fig. 36

Fig. 38

Fig. 37

Fig. 39

— 35 —
discussed above. In all these machines the bottom may be flat
or may be conical towards the centre.
The flat bottom is largely used for light machines with an
elastic upper bearing and generally have the appearance shown
in fig. 37. The bottom with a conical centre may have very
different forms as in the two extremes shown in figs. 38 and 39.
In fig. 38 the central cone is reduced to the minimum
required for giving the necessary support to the shaft and
allowing the loading capacity of the drum to be used to a
maximum. This type may easily be used with the upper
elastic bearing which is so common at the present day.
In fig. 39 the fixed upper bearing entering as far as possible
into the space offered by the central cone increases the solidity
of the whole, and in particular reduces to a minimum the
harmful effects which any disturbance in the equilibrium of
the load and the consequent oscillation of the drum may have
on the shaft, since this action may break a shaft at the point
where it leaves the bearing. It can easily be understood that
the rigid bearing makes any stress on the shaft more sudden,
and hence makes it more likely to break at the spot mentioned.
The present tendency in the construction of all these
machines may therefore be summed up as follows:
For machines with an elastic upper bearing, a flat
bottom or a central cone reduced to a minimum is preferred.
In machines with a fixed upper bearing, the central
cone is raised considerably, and the upper bearing is placed
as high as possible.

RIVETING THE BOTTOM

This subject has already been mentioned in connection
with the joining of the bottom to the drum.
When the bottom is in two parts (the real bottom being of
sheet iron and the central cone or nave of cast iron) the two parts
are riveted together (fig. 34, 35 and 38). The rivets for this
purpose are generally larger than those used for the rim (fig. 40),
and in many cases bolts with countersunk heads are used.
The usual method of two rows of rivets is safer. The countersunk head of the rivet is covered by the coating of sheet copper

Fig. 42

Fig. 43

— 37 —
if such exists (lig. 41). As regards the stress which this row
or double row of rivets has to bear, the reader is referred to
what has been said about the riveting of the Weston machines
(fig. 34, suspended machines). It must, however, be noted that
in view of the type of brake employed these stresses are not
nearly so powerful as those found in the Weston machines,
and the riveting mentioned is always more than sufficient to
ensure the safety of the whole.
BALANCING RINGS

In conclusion, a few words must be said about the balancing
rings, which are done away with in most modern machines but
are found on some of an older type still in use (fig. 42). As these
rings, are not fixed and have a diameter slightly greater than
that of the drum, they rotate rather more slowly than the drum,
and by the force of their inertia prevent, or at least lessen, its
oscillations. In certain types their diameter is less than that
of the drum, in which case they are enclosed in a central shell
round the shaft (fig. 43). The balancing rings do not always
fulfil the purpose for which they are fitted.
They are less and less employed because they hinder the
action of the machine (one or two rows of holes being put out
of action) and they are liable to break under the continual
impacts to which they are subject or as a result of contact with
the liquid which is ejected irom the machine.
It has been mentioned that balancing rings are employed
very rarely in modern machines where perfect equilibrium of
the drum is obtained by the methods described above and by
very careful tests in the workshop before delivery of the machine.
A certain amount of oscillation is inevitable when the machine
is set in motion and an attempt is made to prevent any serious
consequences by adopting the upper elastic bearing, which
lessens the sudden stresses caused by these oscillations until
such time as the self-centring action (resulting from the high
velocity of rotation) takes effect and brings the machine back
to perfect equilibrium.
§ 2. — Outer Shell
That part of the machine which supports and encloses the
revolving part is of great importance in the construction of

— 38 —
hydro-extractors. It is composed of the stand in the strict
sense of the term and the outer shell.
As regards the stand, it is sufficient to note that it is generally
of cast metal and includes the supporting and adjusting bearings
of the shaft. It also forms the base of the outer shell. The
latter is generally of sheet iron or, more rarely, of cast iron.

CAST-IRON SHELL

This type is sometimes found in older machines, either of
the second class (oscillating drum) (fig. 39), or more particularly
in the third class (drum between two bearings) (fig. 44).
This type of shell has certain advantages on account of its
rigidity and strength, but has also serious disadvantages.
There is, in the first place, the difficulty of assembling the
machine on account of its weight and shape, so that if it is at all
possible it is preferred to take down the drum and lift it out of
the shell. In machines with the drum between two bearings
(third class); the work of taking it down is long and tiring so
that it is never performed unless absolutely necessary, and this
naturally does not tend to improve the condition of the machine.
But a much more serious disadvantage, which is of the
greatest importance from our point of view, arises from the
nature of cast iron, which, under a sudden shock, breaks into
innumerable pieces which may be hurled a long distance, and
thus any accident which takes place becomes much more serious.
For these reasons this type of shell is almost completely
abandoned at the present day, except in machines of small
diameter, in which the advantages of obtaining the shell and
the stand in a single piece outweigh the disadvantages which
have just been mentioned.
SHEET-IRON SHELL

Most shells then are constructed of sheet iron. The seam
may be made by a riveted butt-joint or by autogenous welding,
which is the commonest system to-day. The upper closing
ring, however, is generally of cast iron. Its purpose is to give
rigidity to the shell and also to close the upper part of the space
between the drum and the shell, thus preventing the liquid

3ü

Fig 44

•

^^^^^^^H^v JI^^^H

•
m

^^^•iSiH

1
•
•

1
m
'il

^^HÉli

3JF
^^^^^^^^^^^^HHP^S^^^P W

" ' ^ ¿ T " ^ ^ ^ " ^^^ES^k^^^^EmF^^i \

Fig. 45

-ïk %
T?

— 40 —
which is ejected from flying out of the machine and making it
easier to load the machine with the substance which has to
be dried. When it is of sheet iron it is almost always fixed to the
shell by autogenous welding.
If a sheet-iron shell bursts, it naturally is not so dangerous
as the fragile cast iron. But there is another risk, because these
shells are generally constructed in a spirit of economy rather
than according to mathematical calculations, as should be the
case; in the majority of cases indeed, the metal forming the
shell is only between 3 and 5 mm. thick. Under these circumstances, it can be easily understood that this sheet metal
cannot have any very considerable effect in modifying the
dangers of a possible bursting of the drum.
Recently, however, a certain tendency has been noticed
among constructors of these machines to make the shells thicker,
but it is doubtful whether this is due to humanitarian considerations or to strictly technical reasons. It is indeed easy to
understand that the greater the volume of the shell in comparison
with that of the drum the more easily will the oscillations of the
latter be restricted or prevented.
This is observed more particularly in machines of the second
class (oscillating drum) in which the whole machine (drum,
outer shell and stand) forms a rigid whole suspended elastically
from three columns fixed to the ground at an angle of 120° from
each other (fig. 45), so that the oscillations, which are of equal
amplitude for all parts of the machine, are only very slight and
can quickly be subdued. This type of hydro-extractor has the
further advantage of not putting an excessive stress on the shaft
or the ball bearings, while it offers the minimum of passive
resistance by friction to the working of the machine. It is
driven by a co-axial motor keyed on underneath or by intermediate belt gearing from a countershaft, or a motor placed
high up on a bracket or on the wall.
Brief mention must also be made of the fact that for dealing
with certain types of materials these shells are coated with lead
on the inside. This coating, generally not less than 5 mm.
thick, is necessary to protect the sheet metal against the corrosive
action of the acids contained in the liquids which are ejected
from the drum. It has advantages and disadvantages, the latter
being fortunately unimportant as they consist only in making
it more difficult to take down and inspect the machine. The

— 41 —

advantages arise from the fact that the coating of lead considerably contributes to greater safety in case of the bursting
of the drum. This fact is all the more important because the
drums of hydro-extractors which require to be coated with lead
are generally covered with ebonite, and being very difficult to
supervise, are less safe. The dangers arising from lack of
supervision are, to a certain extent, eliminated when these
drums are coated with lead.

§ 3. — Shafts; Supporting and Adjusting Bearings
We refer again to the classification made at the beginning
of this monograph.
(a) In suspended machines (first group) the shaft, like
the shaft of a pendulum, bears the drum to which it is rigidly
fixed, while its upper end is suspended from the frame by
means of a double series of adjustable ball bearings; the
system of bearings is generally reversible.
In order to make the suspension elastic and permit the drum
to oscillate freely, the case which holds the bearings is fitted
into a conical rubber ring, which in its turn bears on an iron
box forming part of the suspension frame. Fig. 46 gives
a sufficient idea of this system.
The shaft is of necessity thick and strong (80 to 110 mm.
in diameter) because it must resist any tendency to flexion caused
by violent oscillations. The attachment to the bottom of the
drum is made by means of a simple nave of cast iron or hard
steel in the form of a wheel which allows the drum to be emptied
between its spokes.
(b) In machines of the second class (oscillating drum)
the shaft is vertical and supports the drum at its upper end ; it
is therefore provided with two bearings, a lower step bearing
and an upper adjusting bearing. These may be ball bearings
(a type which is being more and more widely employed), especially
in the case of rigid supports, or they may be roller bearings or
made of anti-friction metal. This last system is employed
particularly for the upper elastic bearings, which are not
enclosed in a rigid support but are connected to a certain
number of radiating rods, on the outer ends of which are rubber

— 42

Fig. 46

— 43

pads or strong springs which absorb the oscillations of the drum
(fig..47). When the upper bearing is elastic the lower bearing
must naturally be a swivel bearing.
The advantages and disadvantages of the two systems have
already been briefly dealt with.
It may be added that the elastic bearing system (oscillating
below) is preferable when the load is of stuffs, yarns or some
other substance which is not exceedingly heavy but which it
is difficult to keep in perfect equilibrium. The fixed bearings,
on the other hand, are to be preferred with heavy loads in
which the material is so fluid that it automatically brings
about a uniform distribution and perfect equilibrium of the
machine.
The rigid system, as has already been seen, also allows the
upper bearing to be placed higher up in the machine so that
the shaft, being subjected to a less amount of flexion stress,
may be smaller without losing any of its strength.
In practice, however, no clear distinction is made between
the two systems. In the sugar industry, for example, where
the loads to be dealt with are often very heavy (3 to 4 cwt.)
and far from homogeneous (uneven viscosity of the cooked
masses), a very rigid system of construction is found, but the
upper bearing is made elastic by means of very powerful springs
which, even when the equilibrium is disturbed up to a certain
degree, allow only relatively slight oscillations.
The shafts of machines of the second class are generally
from 60 to 70 mm., or in exceptional cases for very large and
heavy machines 90 to 100 mm.
(c) In machines of the third class (drum between two supports) the bearings are always rigid (fig. 48), generally with
rollers or anti-friction metal. Some larger machines of recent
construction, however, have ball bearings.
The distance between the supports varies from a minimum
of 60 to 80 cm. to a maximum of 130 to 150 cm., and the
thickness of the shafts varies proportionally from 70 to 80 or
100 mm.
The drum is slipped on to the shaft up to a wider part of the
latter, to which it is fixed by means of small bolts which serve
to keep it in position, or in exceptional cases by means of a nut
which is screwed on after the drum, the shaft of the machine
being threaded for this purpose.

— 44

§ 4. — Brakes: Different Types; Their Advantages
and Disadvantages
(a) The brake has very important functions in the centrifugal
hydro-extractor. It is, first of all, a very important working
part (and it is almost exclusively from this point of view that
the manufacturers consider it) because it serves to stop the
revolving drum quickly and therefore means a considerable
saving of time, which may be estimated at a quarter or a
half of the time which would otherwise be necessary for the
complete process.
But from our point of view the brake has another important
function connected with the safety of the machinery. The
machine oscillates more and more under the influence of synchronism, and when these oscillations are likely to become dangerous,
they are rapidly moderated and brought back to normal limits
by the judicious use of the brake which must, of course, be
preceded by the stopping of the motor.
A single application of the brake is generally sufficient,
but if the machine when started again should begin to oscillate
dangerously before the self-centring action (gyroscopic effect),
which accompanies angular acceleration, has had time to
moderate the oscillations, it is prudent to apply the brake again
until the machine stops and to rearrange the load before restarting the machine.
(b) Types of brake. — The brakes employed in centrifugal
hydro-extractors may be divided into two main groups, shoe
brakes and band brakes. The various types which may be found
can all be included in one or other of these groups.

SHOE BRAKES

These are the simplest. They consist of an iron fork A
turning round two'pivots B, having at one end the controlling
lever C, and at the other end two wooden blocks L, which,
when the brake is applied, rub against the lower edge of the
driving pulley which acts as a brake drum. To the action Q
one has the corresponding reaction 01 = Q • - (fig. 49).

Fig. 49

— 46 —

The force Q cannot exceed the weight of the revolving parts
of the machines (drum, shaft, pulley, etc.) without raising them
up. This serves as a safety factor for the brake by restricting
the torsion of the shaft and the shearing of the sheet-iron casing.
To bring the brake back to the rest position, two counterpoises G are generally used (shown in dotted line in the diagram)
or a spring on the underside of the brake lever.

BAND BRAKES

These brakes are naturally more widely used and of very
different types. The simplest type is that shown in fig. 50.
Round a pulley of radius R, which is generally integral to
the driving pulley, but of greater diameter, there runs a steel
band coated on the inside with leather or some similar material
to increase friction. One end of the band is fixed at A, and the
other is connected at B to the end of a lever controlled by the
worker.
The tangential force P which is produced on the pulley in
this type of brake by the force Q exerted on the handle is given
by the well-known expression

P = 0 • £ (<^a - i).
which shows that this brake is more effective than the preceding
type. Indeed the ratio — may be increased at will by increasing
the transmissions ; or better still, if great strength is desired,
by moving the point B by means of a screw acting in a rigid
female screw (fig. 51).
In certain types of centrifugal machines which require a
powerful brake the latter may always be controlled by a hand
wheel and a screw with a movable female screw and a fixed
bolt (fig. 52). The two screws are threaded in opposite
directions so that a few turns of the wheel suffice to apply
the brake quickly and forcibly. This type, which is largely
employed in sugar factories, is applied either to a pulley fixed
to the shaft (Weston type) or to a pulley fixed to the drum
(type driven from below). In view of what has been said above,
it will be seen that this latter type of brake is stronger than the
shoe brake, but consequently also more dangerous. We have in
many cases observed that the time of applying the brake did not

— 47 —

Fig. 53

Fig. 54.

— 48 —

exceed four or five seconds, which implies a strain which may
not be dangerous in the case of brakes working on the drum,
but may become dangerous with brakes working on the shaft,
because the latter, especially when the pulleys are of large
diameter, is subjected to torsion stress which may sometimes
rise to 4 or 5 kg. per sq. mm. (Weston).
All other types of band brakes which are used lie between
these two extremes.
There are also other brakes on a principle which may be
considered intermediate between band brakes and shoe brakes
such as that shown in fig. 53.
Two wooden blocks diametrically opposed to each other
are fixed to a metal bracket and worked on the periphery
of the pulley. The brake action is controlled by a lever L
fixed at C and moved by the rod T attached to the handle.
Still simpler is the method of working of the brake shown
in fig. 54. In this case the brake is frequently worked by a
pedal. Another widely employed and very effective type is
that shown in fig. 52 Avhere the simple steel band is replaced
by a ring of wooden blocks which give more friction on the
brake pulley.
Another type which comes nearer to the shoe brake and
which is very commonly used with small centrifugal machines
is that shown in fig. 55, which works directly on the belt
pulley. An overstraining of the bearings is avoided by the
reaction of the driving belt.

§ 5. — Systems of Drive
A hydro-extractor may naturally be driven by any mechanical system of rotation. Mention will be made here merely
of the methods which are found in practical use and an effort
will be made to point out their characteristics and their advantages and disadvantages, particularly from the point of
view of safety.
SUSPENDED MACHINES (PENDULUM SYSTEM)

Three types of drive are found to be characteristic of these
machines. The least common is the belt drive acting on a pulley

— 49 —
fixed to the shaft below the point of suspension and the brake
pulley (fig. 46). The belt very rarely comes directly from the
motor but generally from a transmission, and is often guided
by sheath pulleys, while the machine is put in motion by a
friction clutch.
A more common system is the small Pelton hydraulic
turbine with an impeller of about 500 mm. in diameter fitted
above the point of suspension and co-axial with the shaft of
the centrifugal machine.
The third system, which is found quite frequently and is
spreading rapidly, consists of a co-axial electric motor placed
above the point of suspension, with a starting rheostat.
Recently there has been a considerable extension in the
use of small steam turbines fitted in the same way as the
hydraulic turbines. They are also worked by means of a single
lever which prevents the brake being applied until the steam
has been shut off.
In the following paragraphs we briefly describe the advantages and disadvantages of the various systems.
Bell Drive
This is the cheapest system and is most easily fitted. It is
especially suitable for sets of machines with a single transmission.
In the case of machines which empty below and are placed
on a pedestal, it is better for the drive to be on the opposite side
from that on which the workers are placed so that the latter may
not be near the belts. If the belts should have to pass above
the workers, the latter should be protected by means of strong
iron gratings.
Hydraulic Drive (Pelton Wheel)
This is the safest method and it is perhaps to this fact,
besides its special employment in sugar factories 1, that it
1

The use of hydraulic drive for hydro-extractors in sugar factories
is due to the fact t h a t in these establishments the use of water pumps
worked by steam is easy and convenient because of the abundance
of water and of steam. Further, the presence of a certain amount of
water in circulation moderates the high temperature which generally
exists in the rooms.

— 50 —
owes its wide diffusion. Indeed, none of the working parts
are uncovered, the pressure of the water which is transformed
entirely into kinetic energy on leaving the pipes cannot cause
the turbine casing to burst, and even a burst in the feed pipes
would scarcely involve danger for the workers.
Drive by Electric Motor
This is one of the simplest methods to fit up as it needs
little space and can be very simply worked. It involves,
however, a certain danger by bringing an electric current
into rooms which are often very damp and into the presence
of large masses of metal (sugar factories). It frequently happens
that liquid matter (waste products from manufacture) penetrate,
for example, into the rheostat, and lead to the formation of
dangerous arcs which may put the rheostat and even the motor
out of action.
Steam. Turbine Drive
This is the most expensive system, but it does not present
any grave danger if the turbine is fitted with the safety appliances
approved for these machines (speed controller).

MACHINES DRIVEN FROM BELOW {OSCILLATING DRUM)

The. most common system is a belt drive from a general
transmission or directly from an electric motor. In neither
case, apart from certain exceptions, is the transmission direct,
but generally from a countershaft on the ground with a fast
and loose pulley (with fork) or a friction clutch.
This system of drive is entirely similar to that generally
employed for other rotating machines (moulding machines,
casks, washing machines, etc.) In more rare cases there is
direct drive.
When the machine is driven by an electric motor the belt
(half-cross belt) must be very long and must pass over guiding
sheaves. These belts should be carefully encased and if possible
should be fitted under the floor. The motor should be put

— 51 —

Fig. 55

J^,
1

Fig. 56

—

— 52 —
into action slowly, so that, especially with heavy machines,
too sudden jerks can be avoided.
Another system which is widely employed to-day in sugar
factories for a set of machines is the hydraulic system. Here
again it is the Pelton wheel which is used, as has been seen
already with suspended machines, with one difference, namely,
that it is fitted underneath the hydro-extractor, but the method
of driving the machine is the same, as well as the closing of
one of the pipes when the machine has reached its full regulation
speed.
The system of drive by a co-axial electric motor connected
with the drum from underneath is rarer, but at the same time
is tending to spread at present. The same remarks as were
made before apply to this type. Examples are also found of
electric drive not by a three-phase induction motor but by a
direct current motor (series or compound excitation) (fig. 56).
All direct current motors are characterised by a variable
speed depending on the intensity of the field, and consequently
in certain cases able to attain a very high level, while alternating current motors can never exceed a certain maximum
speed (synchronous speed) and are therefore much safer.
If one considers that centrifugal force varies according
to the square of the velocity, one can easily understand the
grave danger of a hydro-extractor with a compound or series
motor when for any reason an unforeseen change occurs in
the magnetic field.
We have so far omitted to mention the direct drive by
a steam engine, which is very rare at present, especially in
Italy, because of the following disadvantages: the possibility
of excessive speed; the fact that the feed-pipe is frequently
very long and condensation can easily take place in it, especially
if it is badly isolated; the necessity for a high speed (more
especially if the crank works the centrifugal machine directly);
the difficulty of isolating the moving parts; finally, the very
high cost.
MACHINES WITH THE DRUM BETWEEN Two

BEARINGS

These machines are almost always driven from above
by friction coupling : a cast-iron conical wheel on a horizontal
axis and a pinion of fibre or compressed cardboard coaxial

Fi«. 59

— 54 —
with the centrifugal machine (fig. 57). An adjustable plate
spring gives a more or less strong pressure between the wheel
and the pinion and allows the machine to be put into gear
gradually. Apart from this, it is almost a matter of indifference
which system is used for putting the drum in motion.
If the machine is driven from a general transmission by
a belt the horizontal intermediate shaft will have at one end
a pulley which may be fast or loose with a disengaging fork.
This system, however, is not to be recommended because
the continuous pressure (when the machine is stopped) between
the wheel and the pinion easily tends to flatten the surface
of the latter with the result that when the machine is working
noisy and dangerous vibrations take place. We have noted
this disadvantage in several machines driven from below on
the friction system (fig. 58) 1.
Returning to machines of the third category, it may be
noted that after the belt drive, but much less frequent, there
comes the drive by an electric motor fixed directly on to the
framework of the machine (fig. 59). In this case the hydroextractors are generally small and the conical driving wheel
has a diameter almost equal to that of the pinion or even less
because the speed of the motor generally exceeds that of the
centrifugal machine.
On this machine it is rare to find an electric co-axial motor
placed above the framework with rheostat starter or an automatic starter.
No space need be devoted here to steam engines, which
are hardly ever met with.

1
In connection with vibrations mention m a y also be made of the
drive by cone gearing which has been given up in modern machines,
but is still found in older machines driven from below.

CHAPTER V
THE SAFETY OF HYDRO-EXTRACTORS IN PRACTICE

§ 1. — General Considerations
The preceding chapters show clearly the importance, from
the point of view of the safety of hydro-extractors, not merely
of the rules followed by the constructors of the machines in
their calculations and method of building, but also of all the
legal provisions aiming at laying down and enforcing safety
regulations and devices (locking the cover, etc.) intended to
make the machines less dangerous.
The authors possess and have examined carefully a large
number of documents coming from all the principal European
countries. The impression gained from this study is far from
satisfactory, not because the available mechanical safety
devices are unsuitable, for they are numerous, varied and, in
general, quite practica1, but because the legislative provisions,
where such exist, are at present not sufficiently radical or
definite to deal effectively with these important questions.
First of all, certain of the most practical systems of locking
the covers will be examined. Attention will then be directed
to studies which have been carried out on this subject for a
long time past, especially in Italy, and to the extent (very
limited at present) to which protective devices have spread
in industry and the reasons which prevent their adoption.
Finally, a few remarks will be made concerning such special
legislation as exists.
§ 2. — Systems of Locking the Cover
An interesting device is provided by the firm Rudolph
for machines of the first class (suspended machines) (fig. 60,
60a, and 61)

56 —

Fig. 60 a

- 57 —

Fig. 61

Fia. 62

60 —

The machine is driven by a general transmission with a friction clutch controlled hydraulically by means of a three-way
cock. A Lube is fixed to the sliding cover of the centrifugal
machine and, fitting into another tube leading to the cock,
does not allow pressure to be transmitted to the piston controlling the friction clutch unless the cover is closed. In addition
to this, even when the friction clutch is disengaged it is not
possible to open the cover unless the machine is at rest, because
of the action of a centrifugal regulator placed on the shaft of
the turbine and acting by means of a small lever on the valve
gear regulating the water under pressure.
Fig. 62 represents another safety device for tin suspended
machine based on quite a different principle: it simply consists
in covering all the moving parts with wire gauze.
The systems of locking the covers of machines of the second
and third class are naturally much more numerous and varied.
We give, in the first place, a few diagrams (figs. 63 to 74).
As can be seen in figs. 63, 63a, 64, and 65, we have a system
which prevents the belt from being transferred from the loose
pulley to the fast pulley as long as the cover is open, because
the lower end of the rod b enters into a hole made in the horizontal arm of the belt fork. In order that the cover may not
be opened if the drum is still moving the device shown on the
left in figs. 63, 64, and 65 is employed. The handle h which
passes over the cover is fixed to the hooked rod i by the vertical
rod k and cannot be displaced to release the cover unless the
tooth I fixed to the driving pulley (and also the drum) is
stopped.
Naturally, the hydro-extractor cannot be put into action
unless the rod i is turned into the position shaded in the diagram, and this only occurs if the handle h is closed over the
cover when the latter has been lowered on to the machine.
This system has been described in detail because the same
principle with small variations will, be found in several other
systems which are apparently different.
Figs. 70 and 71, for example, show a system for locking
the cover when the intermediate shaft is directly fixed to the
machine. When the fork e (fig. 71) is on the fast pulley the
projection of the rod of the belt shifter prevents the cover
from turning on its hinges and consequently from opening.
It cannot be opened, either, if the fork has been placed on the

— 61

<.:»"ç

Fig. 75a

Flg. 76

9
avV
Fig. 76a

— d'à —

loose pulley, because the lever p fixed to the rod r is kept
vertical as long as the hydro-extractor is in rotation because
of the centrifugal device p which controls the rod r.
It is not necessary to make any comments on figs. 63 to
75a, which are self-explanatory.
Figs. 79 and 80 show other locking systems partly similar
to the preceding ones and partly differing from them.
In figs. 76 and 76a, for example, we find the first specimen
of a locking and stopping device supplemented by an arm F
pivoted at d and contained in the box G. As a result of the action
of air forced out from the drum by centrifugal force during
rotation this arm leaves the vertical position and the hook H
enters into the ring O of the cover, thus preventing the latter
from opening as long as the drum is in motion even if the belt
has been shifted to the loose pulley.
Figs.. 77, 77a, and 77b show another good system for sliding
covers. The rod S falls when the lever P brings the belt on to
the loose pulley, but the rod Si cannot fall unless the horizontal
rod P j is moved towards the right so as to bring the catch H
in contact with the projecting tooth which is fixed to the rod
of the intermediate shaft, and for this purpose that tooth
must be motionless. The device shown in fig. 79 is employed
for covers which move horizontally but pivot round a vertical
axis. The rod S cannot enter the hole X of the rod of the belt
shifter unless the belt is on the loose pulley, but the cover
which actuates this rod cannot be moved unless the regulator
which actuates the rod Si allows it to fall as a result of
diminished speed.
Figs. 78 and 81-83 require no explanation.
Leaving these theoretical diagrams and passing to the
practical application of locking systems, it will be found that
the following device showing the application of the system of
a rod entering into the belt shifter combined with a hooked
arm for closing the cover is often met with. This is the
typical system in machines driven from below with an intermediate shaft on the ground (fig. 84).
The firm Gebr. Poensgen A.-G., of Dusseldorf-Rath, employs
a system of locking the cover which also depends for its action
upon the position of the belt shifter (fig. 85). The cover can
only be opened when the driving belt is on the loose pulley and
the oil pump m has been stopped and the piston Z of the

Fig. 776

$—pH=B«ïï

Fig. 78

66 —

Fig. 79

Fig. 80

Fig. 82

— 68 -

Fig. 83

Fig. 84

— 69 —
cylinder k (which receives the oil under pressure) falls, thus
allowing the ring to descend and the tooth and lever d to be
disengaged.
Another locking system, still for machines of the second
class with an intermediate shaft on the ground and on the
same principle as was illustrated before, is as follows : the cover
cannot be opened even when the belt is on the loose pulley
unless the centrifugal regulator fixed to the intermediate shaft
stops, and thus allows the horizontal rod to slide into the tube
which contains it, thus making it possible for the vertical rod
to descend (fig. 86).
We now come to the application of a device which is slightly
different as regards the second part of the locking machinery
(the first being always with the rod which prevents the belt
shifter from moving unless the cover is closed) (fig. 87).
Here the arm for closing the cover is not moved by centrifugal air from the drum but by air forced through a rubber
tube by a small co-axial ventilator.
Considerable importance may be attached to the following
safety device in which the locking rod, instead of fitting into
the belt shifter, enters into a hole made for this purpose in the
fast pulley, which is thus prevented from revolving. The device
is very simple and allows the usual pneumatic closing of the
cover to be done away with (fig. 88). The same device can be
used for a cover which rotates on a horizontal plane.
For hydro-extractors with an intermediate shaft fixed to
the machine and with friction cones there is a locking device
in existence, by means of which the cover cannot be opened
unless the end of a small rod connected to it can enter into
a hole provided in the protective covering of the friction pulley
and also enter one of the numerous holes made to correspond
with it on the rim of a disc which is firmly fixed to this
pulley.
The following device for closing the cover of a hydro-extractor
of the third class (drum between two bearings) driven by friction cones is shown in figs. 89 and 90. According to this device
the curved part of the rod G keyed on to the cover prevents
the hand wheel H, which regulates the pressure of the conical
wheel on the cone of the centrifugal machine, from turning
forwards, so that the machine cannot be set in motion as long
as the cover is open.

Fig. 85

Fig. 86

Fig. 87

/.l
7

Fig. 88

— 72 -

Fig. 89

Figs. 91 and 92 show a simpler system for machines of the
same group. According as the toothed segment lever is moved
to the left or to the right, the vertical pivot connected by an
arm to one of the covers can or cannot enter a hole provided
for this purpose in the horizontal part and thus does not allow
the covers to be raised simultaneously.
Among the locking systems which we have not yet studied
is the system of the "Ateliers Raxhon Theux" (figs. 93 and
93a).
Very similar to the preceding one is the "Kamlok" system
adopted in Belgian industries for machines of the same type
(fig. 94). In this system the machine cannot work unless the
cover is closed, because the end of the rod of the belt shifter
partly runs across the cover. Further, the cover cannot be
opened unless one presses on the handle in front, which cannot
be lowered unless a horizontal rod worked by the vertical one
to which the handle is fixed has entered into a hole made for
this purpose in the shaft of the machine. Obviously this cannot
occur unless the drum has stopped.
The patented device "Broadbent" seems to us quite original
and practical (fig. 95). It is, of course, employed only in centrifugal machines with a co-axial electric motor, as is shown in

— 73

Fig. 90

74 —

pup "

^" *

fi*"j^Hj^H

^Hn Ü8B». i

..ñau....

—

'Iff

H ^ l
•

j^^H

rapÄ£ ~-j¡fi¡§|AfcijH

IM!
1 1
•^R|

Fig. 91

• "'.-^^^ .. iiia!

Fig. 92

- 75 —

Fis- 93

Fig. 93 a

76-

Fig. 94

Fig. 95

-V-

I'ig. 96
•

•

WWTlÊ^msm
j£JsSi

tÊÊÊÊÊ'^ ' """K

Inf /gay r PI« \íg ;
E u r i LaÄSeiA.
(few

IB«
i

li,

"HWOSSLÍ,« *

^ ^

^SL "«»lift
'•li**!

-•• . * • ,

•* * ' '

''^B

Fig. 07

— 78 the diagram. "When the cover is raised, the interruptor of the
motor, which is seen on the right in the diagram, is open, and
cannot be closed unless the cover is lowered. In this latter
position a pendulum device connected with the head of the
shaft passes through a hole in the cover, and, as it turns
with the shaft, swings out and prevents the cover from being
raised.

Fig. 98

Mention should be made of the device for locking the machine
driven from above by friction cones, which is patented by
Johnson and Russell (figs. 96 and 97). The small hand wheel
cannot make the fork move from the loose to the fast pulley
unless the rod A, leaving the hole into which it is fitted (which
is impossible unless the cover is opened), allows the part B to
move freely towards the right. To open the cover, on the
other hand, it is necessary not only to bring the fork on to the
loose pulley, but also to make the part B, to which is fixed
the forked lever C and the muff D, move completely to the left,
which cannot take place unless the muñ D is disengaged from
the rotating shaft — that is to say, until the shaft has come
to rest.

7« -

Finally, we wish to mention the "Gardloc" locking device,
which is notable on account of its simplicity (fig. 98). The
belt-shifting rod cannot move into the working position unless
the cover is closed. On the other hand, the two discs which
pass through the centre of the cover remain eccentric as long
as the drum is in rotation, and do not allow the cover to open,

CHAPTER VI
CONCLUSIONS

§ 1. — Studies and Publications
In Italy great attention has been paid to the question of
safe working with centrifugal hydro-extractors. As far back
as 1905, indeed, under the auspices of the Association of Italian
Industrialists for the Prevention of Industrial Accidents,
there appeared an important work on this subject entitled
Gli Idroestrattori a Forza Centrifuga e i Pericoli d'Infortunio,
by Mr. Massarelli, an inspecting engineer, whose work was al
the time the most complete on this subject, because it contained
not only a thorough theoretical study of the question, but aiso
the results of practical experience acquired by the author
as manager of a factory for the construction and repair of these
machines. Mr. Massarelli was already protesting against the
empirical methods used in the construction of hydro-extractors
(and still being used), and he completed his theoretical study
of the machine by a practical example dealt with in great
detail. The description of the protective devices was followed
by a series of safety rules officially adopted by the Association
of Italian Industrialists, and prescribed for use by all members
of the Association.
In 1912, at the First International Technical Congress for
the Prevention of Industrial Accidents, held at Milan from
27 to 31 May, Mr. Massarel i, Chief Inspector of the Association,
and General Secretary of the Congress, put forward on behalt
of Italy a monograph entitled Construction et conditions aptes
à garantir la sûreté de fonctionnement des hydroextracteurs à force
centrifuge l.
1

Cf. Comptes rendus du 1er Congrès technique international de Prévention des Accidents et d'Hygiène industrielle, tenu à Milan, 1912.

— 81 At the same Congress reports on the subject were also put
forward by Mr. H de Heu, as representative of Belgium, and by
Mr. Padovani for Italy (on elastic brakes for centrifugal hydroextractors).
The war, which broke out two years later, brought great
changes in the sphere of activity of the Association of Italian
Industrialists, and it was only after the war that it was possible
to take any more definite and effective steps in this matter.
In 1920 a system of inspection of hydro-extractors was inaugurated, and has developed to such an extent that at the present
day approximately 1,300 machines are inscribed, half of which
belong to sugar factories. The most important points in the
inspection are:
(a) a scrupulously careful examination of the material of
which the machine is constructed, particularly the
rotating parts;
(b) the measurement of every part of the machine as far
as is necessary for verifying the calculations ;
(c) tests of the working of the machine, empty and loaded,
in order to examine its method of working, its speed,
any oscillations or vibrations which may occur, and the
efficiency of the accessory parts (brakes, bearings, belts,
motors, transmissions, etc.);
(d) verification of the calculation and the stresses;
(e) introduction of an inspection book for each machine,
showing all the data as a result of which the machine
was approved or was refused permission to work in view
of the stress involved; in certain cases merely a reduction of the speed or of the load, or even of both, may be
prescribed.
Every hydro-extractor which is examined is given a registered
number corresponding to that of the book, and every year
(or every six months in certain industries) the tests mentioned
under (c) above are repeated.
At present membership of this service is voluntary, but
the large number of machines inscribed shows that the
inspection is appreciated at its true value. In the following

— 82 —
paragraphs we shall see, according to the documents consulted,
how much has been done in other countries up to the present in
this respect.

§ 2. — General View oí Legislation and Practice in Various
Countries
We have already examined a certain number of systems
for locking the cover, most of which seem to us practical and
sensible. It must be regretted, however, that we have only
very rarely seen these devices applied in the numerous centrifugal machines which we have visited in actual practice. The
manufacturer is ignorant of their existence and their usefulness,
or, still worse, fears a loss in production and prefers to run the
risk of an accident, the serious consequences of which he generally fails to recognise. The worker, on the other hand, either
from ignorance or because he is generally hostile to all novelties
— especially if he believes, rightly or wrongly, that they may
hinder his work — agrees with the manufacturer, and objects
to seeing a hydro-extractor fitted with a device which, in his
opinion, merely complicates the working of the machine. The
result of this state of affairs is, as we have already said, that we
have observed very few machines at work which are provided
with a cover, and scarcely any which have a safety method of
locking the cover. When such a method exists, it is due rather
to the necessities of work than to humanitarian considerations.
The hydro-extractors in sugar refineries, for instance, are provided with a cover because the sugar must be covered by a layer
of steam during the centrifugal process. The same is the case
in certain machines employed for chemical products; these
machines are covered because the substances dealt with give
off very poisonous vapours (aniline salts, benzol, etc.). This
observation is, moreover, confirmed by reports from abroad.
In fact, in studying documents from various countries, we
find that in general there is little confidence shown in the
practical advantages of the locking systems. In Germany, for
example, the mutual insurance association of the chemical
industries makes the necessity of providing the hydroextractors with a cover conditional upon the quality of the
load.

83-

On the other hand, 'the mutual insurance association of the
clothing trade (Bekleidungsindustrie-Berufsgenossenschaft) lay
down precise regulations in which it is stated, for instance
(section 147) :
Every hydro-extractor, mechanically driven from below, must be
provided with a cover connected with the engaging and declutching
device, so t h a t the cover cannot be opened except when the drum is at
rest, and, moreover, so t h a t the cover must be closed before the machine
can be p u t into motion.
On mechanically-driven hydro-extractors of the suspended type
(overhead drive) the upper part of the moving drum must be effectively
protected b y a loading hopper, 10 cm. high, and solidly fixed to the
outer shell of the machine.

It is often stated that the locking systems whose parts
are worked by a current of air produced by the ventilators
worked by the turbine have given bad results, as also those
which are worked by the oil pump controlled by the centrifugal
machine, because their working is hindered by all the liquids
ejected from the machine (acids, etc.), which make the working
parts rusty and corrode the tubes. We have already studied
certain examples of these systems.
The Factory Inspector of Annaberg, in a letter to the
Departmental Directorate at Chemnitz, points oui that the
locking system employed in that district is that of the firm of
Haubold. But he adds that the workers do not care to employ
it, and prefer to work without any cover on the machine.
Mr. Igel, inspector at the Departmental Directorate of Chemnitz, in his report of 17 September 1925 uses the following
words :
In the course of twenty-six years' experience as a factory inspector,
I have noted t h a t the automatic-locking systems for the covers of
centrifugal machines are unfavourably received in industrial practice.
These devices, indeed, do not give good results unless they prevent
the machine from being set in motion while the cover is open,
and prevent the cover from opening until the movement of the drum
has ceased to be at all dangerous. In practice, this last condition for
their perfect working involves a considerable loss of time, and for
that reason the worker and the manufacturer are both tempted to save
time by doing away with this locking device.

The English report on this subject is still more explicit.
A memorandum drawn up by the Technical Office of the Ministry
of Labour in London states plainly:
The use of automatically locking covers has been pressed by inspectors
in laundries and textile works, and most makers of centrifugal machines

— 84

are prepared to fit them, b u t the successful application of t h e m has
been disappointing. No difficulty appears to arise in the laundries of
hospitals and similar institutions where rapid output is not an essential
feature, but in commercial undertakings the operators, presumably to
save time, often retrain from using the device when installed.

Manufacturers and workers are therefore opposed to the
adoption of these safety devices. In order to overcome this
opposition, it seems to us necessary not only to make the
safety devices better known, and prove their usefulness and
ease in handling, but also for the legislative bodies in all
countries to make the employment of such devices compulsory,
and to institute severe penalties for any contravention of
these regulations.
But, the reader will ask, is it possible that in countries
where hydro-extractors are employed there are at present
no legal regulations concerning them ? Such regulations certainly exist, but they are insufficient, and are limited to certain
countries.
In other countries it is customary to carry out inspections,
but these are arbitrary rather than at regular intervals. It
often happens that laws were formerly promulgated on this
subject, and for no apparent reason have later been repealed
or have fallen into disuse.
Let us pass, however, to documents which will give us a
clearer idea on this subject.
A. In Germany the different regulations regarding the working of hydro-extractors contained in the prescriptions applicable
to the food and chemical industries, etc., have recently been
replaced by new provisions now embodied in the Code of Safety
Regulations for Centrifugal Hydro-Extractors prepared by the
Federation of German Mutual Insurance Associations. The
text of these Safety Regulations reads as follows:
1. Hydro-extractors must rotate from left to right.
2. The outside casing and the cover closing the space between
the outside casing and the rotating drum must be constructed of stout
material (e.g. wrought iron, copper, steel). Where hydro-extractors
with cast-iron casing are still in use, the casing must be reinforced
with wrought-iron rings.
3. All centrifugal hydro-extractors must have protective lids ;
they must be so constructed t h a t they cannot be set in motion before
the lid has been securely fastened. The lid must be so constructed t h a t
it can be opened only when the drum has ceased to revolve. If the nature

85 -

of the industrial process makes it necessary for the material in the drum
to be treated while it is rotating, it is permissible to use protective
lids t h a t can be removed shortly before the drum comes to a standstill.
4. All hydro-extractors must be furnished with a brake, unless its
use might entail the risk of igniting inflammable material.
5. All. hydro-extractors must be fitted with a strongly riveted
plate indicating the name of the constructor, the year of construction,
the workshop number, the maximum number of revolutions per minute
allowable, the material of which the machine is made, the strength
of the drum cover, and the maximum load allowable. By weight of
the load is meant the weight at the moment of loading the drum.
6. Precautions must be taken to ensure t h a t t h e maximum number
of revolutions per minute and the maximum load allowable are not
exceeded.
7. Hydro-extractors must be examined at least once a year by an
experienced man appointed by the employer for the purpose. The
examination must include the' safety devices, the condition of the
bearings and the pivot journal and shall be conducted more especially
with a view to discovering whether the drum has received any damage
or has worn thin at any spot. A report shall be made on the examination, and the results shall be recorded in a Maintenance Register, which
must be preserved, and shown to the technical officials of the Factory
Inspectorate at their request.
8. Milk separators, oil refiners, and conical centrifugal oil extractors
with felt covers serving as filters are excluded from the scope o.f the
above regulations.

The Chemnitz Order, which is often quoted, dated 18 October
1887, is a similar document, of which we give the complete text:
Communal

Order Concerning the Employment
Centrifugal
Machines

The Municipal Council has drawn up the following regulations for
centrifugal hydro-extractors, whether employed by the owners or hired
out to third parties, with a view to preventing as far as possible any
accidents with such machines :
Each of these machines must possess the following safety devices:
(1) All projecting parts of the rotating pieces of the machine, as
well as t h e toothed wheels, should be entirely covered.
(2) The hydro-extractor should be provided with a brake or a similar
device, which must always be in good working order.
(3) The cranks m u s t be constructed in such a way t h a t they will
be stopped automatically when disconnected from the prime mover.
(4) A t the point where the upper edge of the drum is not covered
by the t o p of the shell, a cover should be applied to prevent the worker's
hands getting caught in this space.
Infringements of these regulations will be punished by a fine not
exceeding 300 marks, or by arrest.
The Council further recommends the adoption of all safety devices,
even for hydro-extractors employed for household purposes.
T h i s O r d i n a n c e is still in force.

— 86 —
B. In Belgium this matter is regulated by the Royal
Decree of 20 November 1927, of which the text follows:
Royal Decree Regulating the Use of Centrifugal Hydro-Extractors
in
Undertakings Listed as Dangerous, Unhealthy or Disagreeable
1. In undertakings listed as dangerous, unhealthy or disagreeable,
and subject to the supervision of the Ministry of Industry, Labour and
Social Welfare, and without prejudice to any special conditions which
the competent authorities have the power to prescribe in each particular
case, the employment of centrifugal hydro-extractors is subject to the
strict observation of the provisions specified below.
2. Every centrifugal
and durably with:
(1)
(2)
(3)
(4)

hydro-extractor

must

be marked

clearly

its origin and manufacturer's n u m b e r ;
the date on which it was p u t into use;
the maximum load and velocity;
the minimum time to be used for putting the machine in motion
and stopping it.

3. Every centrifugal hydro-extractor must be provided with an
alarm, which takes effect whenever the maximum speed is exceeded.
It must also, if it is at all possible, be provided with a cover, actuated
by the mechanism which puts the machine into gear, so t h a t it cannot
be opened until the machine has stopped, and the machine cannot
begin to revolve until the cover is closed.
4. Every centrifugal hydro-extractor must be provided with an
effective brake appliance.
5. The protective shells must be constructed of steel, and be
sufficiently strong to resist any explosion of the drum and prevent
the substances being handled from being thrown out.
6. Persons employing centrifugal hydro-extractors shall have
t h e m tested first of all before commencing to use them, and periodically
thereafter, by some body or agent duly appointed by the Central Factory
Inspectorate.
These institutions shall not appoint as agents any persons who do
not offer all the desirable guarantees as regards character, freedom
from bias and skill in discovering and estimating the gravity of any
defects in the machines and appliances inspected.
The inspecting agent shall not be the owner, user, constructor or
retailer of the machines and appliances inspected, nor a person in their
service or employed as their commercial representative.
He must not have performed, nor at any time perform, any repair
to part or all of the machines and appliances inspected.
7.
(1)
(2)

The inspector must carry out :
at least one visit per year t o all hydro-extractors which are
working less than twelve hours per day, and are not dealing
with corrosive substances.
At least two visits per year to all hydro-extractors working
twelve hours or more per day or dealing with corrosive substances.

-87

The visits shall be more frequent for machines working under
specially unfavourable conditions.
Each inspection shall be followed by a report by the inspector
appointed, and this report must at any time be available for the factory
inspector.
Temporary Provision
8. The above provisions shall apply to all machines installed before
the publication of the Decree but after 1 April 1928.
C. In t h e Netherlands t h e only legal regulations (which apply
equally t o all machines driven with a rapid rotatory movement)
are t h e following sections of t h e Royal Ordinance of 21 August
1916 (Staatsblad No. 418):
151. (1) Machines whose parts are in danger of cutting, bruising
or crushing the workers, or which are dangerous on account of their
high speed, such as . . . centrifugal machines . . . must be
installed and constructed with such protective devices as will avoid
the dangers as far as possible.
(2) As far as is necessary and possible, workers employed on these
machines must be protected by means of effective safety devices.
153. A mill, or any other machine or tool driven by a motor and
liable to risk of bursting, must always be carefully maintained, suitably
installed, and effectively protected. Except in the case where such
machines are driven by windmills, steps shall be taken to avoid a sudden
change of speed or any excess of the maximum speed allowed for the
safe employment of the machine.
D. In England, according to t h e Factory Inspectorate,
there are no special regulations on this subject.

§ 3 . — Relations between the Constructors and the Manufacturers
who Employ these Machines. — Guarantees Required from
the Constructors. — Periodical Inspections.
As has been seen, there is no touch between the constructors
and t h e manufacturers who employ these machines.
The German m u t u a l insurance associations do not demand
any details of t h e calculations involved from t h e constructors,
b u t merely information regarding the m a x i m u m load and speed.
Certain associations have their centrifugal machines inspected
from time to time by competent experts, who report on any
changes which seem desirable as a result of t h e regular inspection of t h e drum and of the loads.

88 -

The Clothing Association simply leaves to the construe!ing
firm all responsibility regarding the calculations and the construction of the casing, rings, etc.
To avoid all risk of explosion, it recommends the strict
observance of the regulations and principles of the International
Boiler Construction Society (Hamburg Rules, 1905).
According to these regulations, the maximum load and
number of revolutions must be taken into account in the
calculations. The calculations should also take account of a
coefficient of reduction for the natural corrosion of the casing.
Other provisions recommend constructors and manufacturers to
take account of the work for which centrifugal machines are
to be employed — for example, if the substances dealt with are
to be acid and corrosive, the inside and outside of the drum
should be coated with ebonite, lead or enamel.
Other German mutual insurance associations have also
adopted the provisions of the " Bekleidungsindustrie-Berufsgenossenschaft ".
The Factory Inspectorate in Chemnitz, in a letter to the
Departmental Directorate, also states:
In our district there are no definite regulations for the calculation
of centrifugal machines; neither is there any agreement between the
manufacturers of these machines and the industrialists who use them.
There are no provisions or agreements between the different associations regarding preliminary tests and regular inspection of centrifugal
machines, and we can give the assurance t h a t u p to the present the
necessity for this has not made itself felt in our Department.
Although the machines are largely employed in our area, the number
of explosions and accidents resulting from them is almost insignificant,
despite the fact t h a t most of the hydro-extractors are not provided
with a safety cover.

On the contrary, the Saxon Association for the Inspection
of Boilers undertakes the inspection of centrifugal machines
in about eighty factories. With regard to the principles of
accident prevention on hydro-extractors, this association makes
the folluwing statement:
1. The moving parts of the hydro-extractor, and in particular
the casing, the bottom and the cover, must be constructed in such a
way as to offer sufficient resistance for the m a x i m u m speed of revolution
with the maximum load in the drum. For this reason certain factories
talk of a safety coefficient of 5, or even 6.
In Saxony there are no regulations of this sort known, and we arc
not certain whether any agreements have been entered into between
the associations of constructors of these machines and the industrialists
who use them.

— 89 —
2.

We have no experience regarding automatic locking systems.

3. We do n o t know whether any definite agreements have been
arrived at between the industrialists who use hydro-extractors with
reference to tests on delivery and to regular inspection of these machines.
For some years, however, several of our members have entrusted us
with tests on delivery and the periodical inspection of centrifugal
machines with a view to the prevention of accidents.
During the tests on delivery of the machine — t h a t is to say before
it is regularly used — we examine the calculations regarding the
principal parts of the machine, basing our examination on the belief
t h a t the maximum speed of rotation, with the m a x i m u m load, should
correspond to a safety coefficient of a t least 5.
The periodical inspections are carried out partly when the machine
is working and partly when it is at rest, and in m a n y cases the machine
is taken entirely to pieces.
At present we carry out a periodical inspection of seventy-four
centrifugal machines.
4. We know of no accident caused by centrifugal machines which
can be of a n y special interest from the point of view of accident
prevention.

In Wurtemberg, as the Wurtemberg factory inspectorate
reports, there are no local laws concerning the calculation,
construction and inspection of the machines in question. The
centrifugal machines employed in factories are examined by
the factory inspectors, but no calculations are made to test
their power of resistance.
If the owner of the centrifugal machine wishes this test
to be made, the Wurtemberg Association for the Inspection
of Boilers undertakes this operation and bases its calculations
on formulae concerning the stress on the casing given in the
review Mitteilung aus der Praxis des Dampfkessel- und Dampfmaschinenbetriebes (Berlin and Breslau), 1893, pages 105 and
197. Figures for the maximum stresses correspond to those
laid down in the Police Regulations for the construction of steam
boilers, dated 17 December 1908.
Reports of the British factory inspectorate contain a number
of interesting and sensible remarks of a general nature regarding
the construction of centrifugal machines and the materials
employed, but, with reference to the subject with which we are
dealing at present, only the following remarks are made:
. . . All makers (of centrifugal machines) agree as to the desirability of periodic examination by experienced persons. One firm of
makers carry out periodical inspections, at the users' request, of several
machines installed by them, and I understand that at least one insurance
company does business of this kind. The inspection comprises examination as to corrosion and also test as to balance.

— 90 —
Further on the following passage occurs :
Present practice is to test the machines at the maker's works under
the working conditions called for by the users, and the data as to speed
and loading are supplied to the user. This information is, unfortunately,
not always passed on to the proper person, so that frequently the
operators and persons in charge have no proper idea of the safe speeds
and loads.

In the Netherlands there is no State inspection nor are there
any periodical inspections by private associations. But, in terms
of the Royal Ordinance of 21 August 1916 (Staatsblad, section 270
of the Ordinance mentioned) the competent factory inspector
may examine a centrifugal machine, take its measurements,
test its speed and demand from the responsible persons all
necessary data for discovering whether sections 151 and 153
of this Ordinance (quoted already) have been carried out.
In Belgium the supervision of hydro-extractors is regulated
by the Royal Decree of 20 November 1927. Periodical supervision is carried out by agents appointed by the Central Factory
Inspection Service (cf. section 6 of the Royal Decree).
§ 4. — Accidents
It remains now to mention accidents. But it seems superfluous to spend more time on this subject seeing that we have
already discussed it at length in the chapter devoted to the
reasons for these accidents.
It may simply be noted that the accidents which occur are
not very numerous in proportion to the large number of hydroextractors employed in all industries; this relatively small
number is, however, made up for by the extreme seriousness
which characterises these accidents.
Indeed, nine times out of ten the death of the worker or
workers attending the machine is the direct result of the bursting
of a centriiugal machine, and very often other victims, who have
nothing to do with the centrifugal machine, are injured as a
result of the carelessness, clumsiness or ignorance of other people.
This fact need cause no astonishment if one considers that
these machines, when working at full speed, possess kinetic
energy which varies from a minimum of 8,000 to 10,000 kg.
to a maximum of 400,000 to 500,000 kg.
*

— 91 —
Before concluding this monograph, we should like to thank
all those who have been good enough to supply us with the
necessary material for this compilation, and we. hope that the
distribution of this work by the International Labour Office
will speedily result in the filling up of all the blanks which still
exist in this connection in the legislations of various States,
so that wherever hydro-extractors are employed they may be
compelled to undergo periodical inspections, which will not only
safeguard the life of the workers, but, in time, lead to a selection
of those types of machines which will be most complete and
most perfect from the point of view of industrial practice.

BIBLIOGRAPHY

1. Gli Idroestrattori a Forza Centrifuga e i Pericoli
d'Infortunio.
A practical and technical study by Mr. F . MASSARELLI, engineer and
inspector to the Association of Italian Industrialists for the Prevention
of Industrial Accidents. Milan, 1905.
2. "Construction et conditions aptes à garantir la sûreté de fonctionnement des hydroextracteurs à force centrifuge." By Mr. F . MASSARELLI. Comptes rendus du 10T Congrès technique international de Prévention des Accidents et d'Hygiène industrielle, tenu à Milan, 1912.
3. Association des Industriels de Belgique pour l'étude et la propagation des engins et mesures propres ci préserver les ouvriers des accidents
du travail. Bulletin, containing the reports for the year 1912, page 1 7 :
" R e p o r t of Mr. D E H I Î U , Inspector to the Association of Belgian
Industrialists, 1913."
4. Unfallverfiütungs- Vorschriften
der
Bekleidungsinduslrie-Berufsgenossenschaft. Gültig vom 1 November 1922. Berlin, 1921.
5.

Various letters and reports.

6. Various catalogues and explanatory diagrams from various
constructors and institutions according to the following list:
(a) Germany: Haubold. — Rudolph und Co. — Röhring und
König. — Weisbach. — Gebr. Heine, Viersen. — Gebr.
Poensgen A.-G. — Braunschweigische Maschinenbau-Anstalt
(A.-G.). — Rieh. H a r t m a n n . — etc.
(b) England:
Thomas Broadbent and Sons, Huddcrsiicld. —
Johnson and Russell. — Watson, Laidlaw and Co., Glasgow.
— I. I.-Leno, Ltd. — Mann, Lowe, Allioth and Co., Nottingham. — Aublet, H a r r y and Co., Ltd. — Troy Laundry
Machinery Comp., Ltd.
(c) the Netherlands: Arbeidsinspectie. — X.V.X.A. Spronk and
Zonen's Maschinenfabriek, Rotterdam. — Soc. a n . " Reinweld",
Delft.
(d) Belgium:
Ateliers " R a x h o n - T h e u x ".

APPENDIX

SOME ACCIDENTS

CAUSED

CENTRIFUGAL

HYDRO-EXTRACTORS
F I R S T CASE

Figs. 99 and 100 give an idea of the consequences of the bursting
of a centrifugal machine of the Schrceder type employed in a large
refinery for the manufacture of cube sugar. This accident caused
the death of two workers and injured a third. It is sufficiently characteristic to deserve mention here.
It is well known that the cooked mass and the syrup (covering
layer) are separated from each other in centrifugal machines in which
a kind of very heavy cylindrical counter-drum is placed inside the
drum; the base of the counter-drum is formed by a circular ring,
the. outer surface of which fits into the inner surface of the drum
(fig. 101). This ring is formed of a great number of vertical radial
segments placed at a distance of about 1 cm. from each other, so that
once the operation is concluded all these spaces are filled with refined
sugar which forms so many slabs. In the present instance the drum
had an internal diameter of 1 m. 13 mm., the weight of the counterdrum was approximately 750 kg., while that of the molasses was
about 200 kg. and the maximum speed was about 1,000 revolutions.
The drum was provided with an external ring which was not
connected with the drum, but merely rested on the smaller side of
the section. This ring, which could not move in a verlical direction,
because of the buff-joints of the sheet iron drum, had a special purpose : it served not only to strengthen the drum but also to catch in
special projections fixed to the inside of the outer shell, so that once
the operation was concluded and it was necessary, with a great expenditure of effort, to raise the counter-drum containing the sugar, the
drum itself was prevented from being raised at the same time.
The disaster with this machine was caused by the breaking of the
above ring, which was hurled out against the wall of the cast-iron
outer shell, which was 25 mm. thick. The effect is clearly shown by
the photographs in figs. 99 and 100.
It is well to give in full the observations of the competent office
of the Association immediately after the accident:
1. The ring surrounding the drum broke under the action of
centrifugal force at the normal speed of 1,000 revolutions when I he
drum was loaded with the counter-drum.
2. The ring fixed closeiy to the drum is of soft steel, and its section
is 7 0 x 2 0 ; it consists of two halves joined end to end by means of two
double butt-joints with screw bolts. The shorter side of the section
of the ring rests against the casing (fig. 102).

— 94 —

Fig. 99

Fig. 100

Fig. 101

Fig. 102

Fig. 104

— 9b —
3. The break took place at a point on the half ring 20 cm. from
the extremity of the butt-joint (fig. 103), and an examination of the
drum clearly showed that the ring was held together only to the extent
of a few square millimetres, so that the centre part was completely
free for 80 to 90 per cent, of the whole resisting surface (fig. 104).
4. Judging by the shape, colour and appearance of the surface
of the two sides of the ring at the point at which it broke, it seems
clear that the weakness of the material is due to faulty welding of
the two parts of the bar which was employed in the manufacture of
the ring. The two parts carried out their function merely at the
outside and, in particular, on one of the larger sides of the section.
When the ring was turned later the best part of the material which
had been firmly welded was removed, but not to such an extent as
to reveal the part which had not been welded. There was, therefore,
a defect in the material which the worker who fitted the ring and put
the machine in motion could not possibly have observed.
The same characteristic features of the section which led to the
conclusion that a faulty welding had been made may also suggest
that the fault could lie in the rolling of the iron bar.
5. Assuming the normal speed to be 1,000 revolutions per minute
and the maximum radius of the ring to be 640 mm., the peripheral
velocity for the extreme outside edge of the ring would be approximately 67 m., a speed which may be considered as normal for soft
steel, which, provided it is of homogeneous section, can work at
approximately 4 kg. per sq. mm.
Obviously the responsibility falls on the factory which constructed
the ring (the ring was a replacement and did not come from the original
firm), and the break took place a few minutes after the machine had
been set going in order to test it after it had been repaired and the
ring replaced, at a moment when there were very few workers in the
room.
Two of the workers received the full force of the blow from the ring
as it left the machine; the third, who was some distance away, was
struck by a fragment of the outer shell.
SECOND

CASE

An accident of exceptional gravity (three workers killed and eight
severely injured) occurred in a sugar factory in Venetia as the result
of the bursting of a drum in an Adant centrifugal machine for the
manufacture of cube sugar, driven by a direct current motor with
compound excitation of 220 volts.
The necessities of the work demanded that the centrifugal machine,
and consequently also the co-axial motor, should be able to change
their speed very rapidly (cf. accompanying diagram, fig. 105).
These changes were obtained by modifying the exciting current
both in the series winding and in the shunt winding by means of a
rheostat in the corresponding circuits controlled by a moving disc
which revolved above a fixed contact disc (cf. diagram, fig. 106).
After the corresponding resistances in the exciting circuits have
been switched in, the machine is set in motion simply by touching
a button.

— 97 —
While Lhe machine is in motion the various speeds are made to
follow each olher automatically by rotating the moving disc, and
once Lhe centrifugal process has been concluded another pressing of
the button brings into action the electric brake (800 to 300 revolutions)
by the motor which acts as a dynamo and later also acts as a powerful
magnetic brake (350 to 0).
The advantages of this system arc as follows:
(a) By suitable variations of the resistances between two contact
points, it is easy to obtain quickly, precisely and at the required
moment a corresponding variation in the speed of the motor, and
consequently in that of the drum, merely by revolving the moving
disc.
(b) The system is absolutely auLomatic, and thus the worker
has merely to put the machine in motion and stop it.
(c) The brake is applied to the hydro-extractor by the motor,
which is acting as a dynamo, by absorbing more than six-sevenths
of the kinetic energy with a corresponding recovery of electric energy.
(d) The employment of relays which, on account of the faci
that they are actuated by a direct current, work easily and dependably.
The disadvantages may be summed up as follows. The speed
at which the motor rotates depends on two causes (terminal voltage
and intensity of the electric field), and may therefore vary between
very wide limits. It cannot, therefore, be kept with absolute certainty
within, the maximum limits required for the machine unless special
safety devices are introduced in the circuit. This machine, therefore,
as a result of accidental, external causes, may be subjected to stresses
much greater than those for which it was intended and built.
Causes of the Accident
One cause of the accident was the break which was observed as
a result of a short circuit in the shunt winding, which resulted in a
great weakening of the field with a corresponding increase in speed
(the terminal voltage remaining constant).
Another cause observed after the accident was that the joint of
the cylinder had been badly welded (fig. 107) and the cylinder was
only held together at the lower end of the generatrix for a quarter
of its length. Thus, allowing for a bursting load of 50 kg., reduced
to 37.5 (by applying a co-eificient of 0.75 for the welding) and recalling
that the stress resulting from centrifugal force increases in proportion
to the square of the velocity, and that the resisting section, allowing
for the presence of strengthening rings, could be calculated at a third
of the normal, then a speed of 1,150 revolutions per minute was
sufficient to make the drum burst and result in the breaking of the
outer shell, causing the deal h of three men and injuries to other eight.
New Safely Devices which have been Adopted and their Effectiveness
As a result of this accident, the sugar factory, in agreement with
the firm which had constructed the centrifugal machine, adopted

2 4 2%mintiH

Fig. 105

Fig. 106

r^^^_^

^~T
r

o
o

0
^
*O ' 0
s o! 0
So! 0
7 o; o

Fig. 107

Fig. 108

—--

— 99 —
the following provisions for preventing the danger of these accidents
in future:
1. A centrifugal regulator which opens the circuit at 850
revolutions.
2. Relays with the minimum switch fitted on the shunt circuit
and taking effect as soon as the current in the latter falls below
0.6 A., which is the normal strength for 850 revolutions.
This latter device naturally does not allow the interrupter to be
closed if the current in the same circuit is lessened or ceases entirely.
3. The adoption of cylinders formed of a single-piece of steel
without welding and forged. Bursting load : 50 kg. per sq. mm. Stress
observed: 5.76 kg. per sq. mm. Corresponding co-efficient of safety:
50:5.7=8.7.
The sugar factory also fitted a tachymeler with a dial showing
the angular speed of the machine and always in view of the worker
controlling the machine.
We believe that these devices are very effective and give the best
guarantee that accidents of this type will not occur in the future in
machines worked on this system.
Comments

As far as we know, there seems to be no firm for the construction
of electric machines which has so far studied a system for regulating
the speed of an induction motor which can be compared, from the
point of view of relative simplicity and accuracy, with that found
on the Adant motor for centrifugal machines.
We now ask whether it would not be possible to drive these machines by means of alternating current motors whose safety characteristic is that their maximum number of revolutions corresponds to the
synchronous speed and makes any other safety device unnecessary.
On the other hand, an alternating current motor cannot be used
as a brake except at a speed greater than that of synchronism. Ils
adoption would therefore make it necessary to have recourse to some
other system of brake which would supplement or replace the electromagnetic brake at present employed for low speeds (under 300 revolutions).
It is very desirable that a complete scientific study should be carried
out of the methods of regulating the speed of a three-phase current
motor, the adoption of which for driving machines of this class would
not only make them absolutely safe for the reasons mentioned above,
but would avoid the necessity of supplementing the installation by
a very expensive converter group.
THIRD CASE

F"* This accident, which seems worthy of mention, was brought
about by the breaking of the drum of an old hydro-extractor in a
dye works at Busto Arsizio. There were fortunately no fatal consequences, but we think it well to mention the accident as an example
of how the inexperience and rashness of persons working the machine

— 100 —
without realising the dangers involved may result in very serious
accidents.
The machine was of a light type with an oscillating drum of sheet
iron without strengthening rings. It had been bought and put in use
not only without testing whether the number of revolutions (800)
advised by the seller was compatible with the state of preservation
and the strength of the machine, but the number of revolutions was
even increased to 900.
As a matter of fact, the outer shell was of very thin sheet iron
as is the case in most of these machines, and consequently it was
quite impossible to lessen even to the slightest extent the effects of
any explosion of the drum.
The drum was very worn and the riveting of the double butt-joint
gave no guarantee of strength, because the heads of the rivets were
almost completely worn away and the edges of the joint were so worn
that they no longer offered any resistance to the rivets (fig. 108).
Without entering into too many technical details, we may mention
that:
(a) The riveting consisted merely of seven rivets of 6 mm. diameter on each side.
(b) The distance between the edges of the metal of the casing
and the centre of the rivets was too small and had been decreased
by the wear and tear of the metal.
(c) The iron casing had been perforated with a punch and not
by boring. This is a serious mistake, because punching involves a
great strain on sheet iron in the neighbourhood of the holes and
greatly diminishes the resistance.
Reconstructing the accident in view of these observations, it may
be concluded that it occurred as follows:
In the first place, rivets 1 and 3 broke because they were too close
to the edge of the metal casing. As a result, the resistance was reduced
to five rivets instead of seven for a load of 35 kg. (burstingload), which
immediately caused the remaining rivets to break and the drum to
burst, and subsequently to break through the outer shell, which was
too thin.
It should be noted that this force (35 kg.) should, according to
our calculations, have been produced even under 800 revolutions, so
that the machine would have burst even if the speed had not been
increased to 900.
It would therefore be a good rule for all firms purchasing hydroextractors, especially second-hand, to have the machines carefully
examined by experts before starting to use them.
In the preceding case, the only advice we could give was to sell
the machine as old iron, because' it did not seem possible to use it
again in view of the serious damage which had been done to other
' parts of the machine as well.
FOURTH CASE

This was an accident due to the explosion of the drum of a hydroextractor employed in a dye works, which unfortunately resulted in the
death of one man.

101 —

This firm had had Lhe machine for leu years and only the motive
power had been changed. The sLeam engine had been replaced by an
electric motor with a slight increase of speed.
This latter fact, however, is only a partial explanation of the
accident, because the examination has shown that there was a very
slight difference between the two speeds.
The real reason was the stress to which the machine and the sheet
iron of the drum were subjected, and to the faulty welding, as was shown
by an examination of the various parts after the accident.
This machine was of the ordinary type driven from below (with
elastic suspension of the case, a lower bearing and a shoe brake)
and may be described as follows:
(a) The drum, which had no strengthening rings, was of sheet
copper and measured 1 m. in diameter, 370 mm. in height, and 2.75 mm.
in thickness, while the perforations, 4 mm. in diameter, were arranged
in quincunxes. The casing was welded with partial overlapping. The
casing was attached to the bottom by fifty-eight copper rivets of
8 mm. in thickness and the bottom was coated on the upper side with
copper and had a conical nave of cast iron. The upper edge of the
truncated cone was of copper, with a copper strengthening ring at the
narrowest part (750 mm.) joined to the casing by forty-two copper
rivets also of 8 mm. thickness.
(ft) The outer shell was of sheet iron (5 mm.) of 1,280 mm.
diameter.
How lhe Accident Occurred
The hydro-extractor was working at. full speed with a load which
at the beginning had been abouL 140 kg., but which had been reduced
to 97 kg. (weight of the cotton which had been dried) because the operalion was at an end, and one of the workers had already moved the belt
to the loose pulley. This same worker was about to apply the brake
when the drum, still surrounded by its shell, exploded with a loud
report and fell about, six metres away, broken and twisted, but still
covered by the shell, and containing most of the yarn which had
been dried. In its flight it struck and killed one worker and then
struck some baskets and dye-containers which stopped U.
Probable Causes
According to an examination of the various parts made on the
spot, the causes of the disaster may be stated as follows:
1. The speed, load, and consequently the stress, were much
greater than those generally allowed for copper. Our calculations
gave traction stress on the sheet metal of about 9 kg., whereas generally
with welded copper one should not exceed 3 kg. per sq. mm.
2. The welding of the drum was insufficient, as is proved by the
fact that the latter had torn along two-thirds of its height, following
the line of the welding, while only at the lower end did the break move
to an oblique line following a row of perforations (line of least
resistance).
Keeping these points in mind, it is easy to reconstruct the phases
of the accident.

— 102 —
Under the action of centrifugal force the drum gave way along the
welding (which had been insufficient right from the beginning as is
shown by the carbonous edges, but which had become weaker with time
under the action of acid liquids used in dyeing) and the part which
remained intact had immediately burst also because it was too small
lo bear the whole stress ( 9 x 3 : 1 = 2 7 kg. per sq. mm.).
3. The drum having thus
the direction opposite to that
thus tearing the rivets which
breaking the three radial rods

lost its equilibrium moved quickly in
in which the break had taken place,
attach the casing to the bottom and
of the upper elastic support.

4. Finally, the whole mass — drum, yarn, and the outer shell
(which simply rested on the base and was carried away by the drum) —•
broke off from the machine and was hurled like a meteor for a certain
distance. The comparison with a meteor is by no means exaggerated
if one considers that the peripheral speed of the drum revolving at
940 was 50 m. per second, and that the centrifugal force was as
much as 45,000 kg.
Observations and Conclusions
Was this accident inevitable or not and had the firm done all in
its power to avoid it ? The reply must be in the negative for the
following reasons:
1. The firm bought the machine second-hand with a guarantee
from the constructor.
2. Prudence ought to have made the firm have the machine
inspected by a competent person who would not have been satisfied
with a general examination, but who would have made the necessary
calculations to test the machine and see whether with the material
and the thickness of the casing which he observed, the load (140 kg.)
and the speed (940 revolutions) were not too great.
It would at once have been noticed that it was dangerous to make
a copper drum work at 9 to 10 kg. per sq. mm., and that it was necessary to reduce the speed to 600 revolutions, thus bringing the traction
stress down to 3.6 kg. per sq. mm. It would also have been found
desirable to strengthen the drum by means of two or three iron rings
shrunk on. To these facts must be added the negligence in the supervision and maintenance of the machine. It must not be forgotten that
this hydro-extractor had for long years been subjected to the corrosive
action of slightly acid substances. This action particularly affects the
welding, where a solution of continuity may take place, but probably
no precautions had been taken to prevent this and no one had troubled
to examine carefully the interior uf the drum to discover whether the
edge of the welded sheet metal had begun to wear.

P R I N T E D B Y ATAR, GENEVA