REPORT 
ON 
D. C. GREEN’S SYSTEM 
OF 
VENTILATION. 
WASHINGTON: 
GOVERNMENT PRINTING OFFICE 
1 879. REPORT 
ON 
D. C. GREEN’S SYSTEM 
OF 
VENTILATION. 
WASHINGTON: 
GOVERNMENT PRINTING OFFICE. 
1879. REPORT CN D. C. GREEN’S SYSTEM OF VENTILATION. 
New York Navy-Yard, July 30, 1870. 
Sir : In obedience to the orders of the bureau, we have examined the 
system of ventilation proposed for ships by Mr. D. 0. Green, of No. 88 
Liberty street, New York City, and the mechanical apparatus with 
which he gives it practical effect. Both the system and the apparatus 
are shown in the accompanying drawing, on which are marked the neces- 
sary explanations and references. 
The system consists in either exhausting the stagnant air from the 
compartments of a vessel, in which case fresh air flows in through open 
hatches or pipes to fill the vacuum, or in injecting fresh air into the 
compartments, thereby forcing out the stagnant air through open hatches 
or pipes; or both methods may be employed in combination. 
The mechanical means by which these operations are effected is an 
injector discharging compressed air, the current of which induces a much 
larger volume than itself of either the stagnant or the fresh air that may 
surround it. 
The compressed air is furnished by an ordinary double-acting air- 
pump worked by a common steam-cylinder and fitted with a reservoir 
wherein the air-pressure is maintained practically constant during the 
strokes of the pump-piston. The air-pump, with its steam-cylinder and 
reservoir may be placed in any convenient part of the vessel, the com- 
pressed air being conveyed by pipes to the injectors. The air-pressure 
in the reservoir preferred by Mr. Green is from 1 to 5 pounds per square 
inch above the atmosphere. The injectors may also be placed in any 
convenient part of the vessel, suitable pipes leading from them to the 
compartments to be ventilated. But the nearer the air-pump and the 
injectors are to these compartments, the more economical both in piping 
and in air-supply will be the ventilation. 
What we term the injector consists of two parts, namely: the one, of 
an automatic valve placed at the end of the pipe which brings the com- 
pressed air from the reservoir. This valve is shown in Figure 1 of the 
accompanying drawing, and is so designed that it can be counter-loaded 
by a spring or weight to any air-pressure it is desired to employ, its 
action under this load being such that the air flowing from the valve 
will always have the pressure for which the valve was loaded, whether 
the air-supply be great or small from the pump. By this arrangement, 
the injector working always with a constant pressure of air, the velocity 
of the air-discharge will be constant also, but the quantity of air dis- 
charged will vary with the quantity supplied by the pump, the velve 4 
graduating itself to a greater or less area of discharge proportional to 
this quantity. The valve is, in fact, a peculiar throttle-valve worked 
automatically by the pressure of the compressed air. It is frictionless 
when in the vertical position, and has but a trifling friction when hori- 
zontal. The valve is placed entirely within the hollow frustum of a cone 
forming the end of the pipe bringing the compressed air from the reser- 
voir, and consists of a solid frustum of a cone; the two frusta having 
the inclinations of their conical sides in opposite directions, the sliding 
of the inner solid frustum out or in within the hollow frustum, under 
the action of the spring or weight, will increase or decrease the width 
of the annular space between the frusta which is the area of discharge 
for the confined air. The valve or inner frustum being counter-loaded 
by a spring or weight to an equilibrium against a given air-pressure in 
the reservoir, will automatically assume a position within the outer or 
hollow frustum that will give the proper area for discharging under that 
pressure all the air supplied by the pump. If the pump should change 
the quantity of air supplied in a given time the valve will move auto- 
matically to a new position which will give exactly the area required to 
discharge the new rate of supply under the same pressure. This valve 
is the novel, and also the important, feature in the mechanical details of 
the ventilating system employed by Mr. Green, maintaining without 
supervision an uniform velocity of air-discharge, be the quantity of air 
discharged what it may. The amount of movement of the valve within 
the outer hollow frustum is exceedingly small for very great differences 
in the amount of air-supply during equal times. 
The second part of the injector is the receiver, into which the com- 
pressed air is discharged by the valve just described. This receiver, 
lettered S in Figure 2 of the accompanying drawing, is a short, cylin- 
drical pipe, with its two ends formed into frusta of cones; the outer 
diameter of these frusta is the same as the inner diameter of the pipe 
containing them; while the intermediate cylindrical portion is, of course, 
of much less diameter than this pipe. The receiver, together with the 
valve that discharges compressed air into it, are placed in the ventilat- 
ing pipe, or pipe through which the stagnant air is exhausted from the 
compartments of a vessel, or fresh air is delivered into them. The form 
and dimensions of the receiver, relatively to the valve, exert a marked 
influence on the economy and the efficiency of the ventilation. The 
proper proportions can only be determined by careful experiment on a 
considerable scale. 
If it be desired to exhaust a compartment of its stagnant air, the large 
ventilating pipe must extend from the bottom of the compartment to 
any point convenient for delivering this air into the atmosphere, the 
fresh air replacing the stagnant air thus exhausted being delivered into 
the top of the compartment through hatch openings or a suitable pipe; 
the pressure causing the replacement being in proportion to the vacuum 
formed by the injector. 5 
If, on the contrary, it be desired to force air into the compartment, 
one end of the ventilating pipe must communicate with the atmosphere, 
while the other end terminates at the top of the compartment. An out- 
let to the atmosphere must be provided through another ventilating pipe 
of equal diameter extending from the bottom of the compartment to the 
atmosphere; or the fresh air may be delivered into the compartment 
through a ventilating pipe extending to the bottom of the compartment, 
while the top is open to the atmosphere by means of a hatch or pipe; 
the stagnant air, in both cases, being expelled by the greater pressure 
of the fresh air forced in. 
Finally, both methods may be employed simultaneously for the same 
compartment, without additional ventilating pipes. All three arrange- 
ments are shown in Figure 2 of the accompanying drawing, which repre- 
sents a compartment of a vessel, with the steam-cylinder, air-pump, in- 
jectors, and pipes necessary for each and all the methods of ventilation 
jiroposed. 
We witnessed the working of the system of ventilation employed by 
Mr. Green. The compartment ventilated was 25 feet by 20 feet on the 
floor and 9 feet in height; and was made reasonably air tight for the 
pressure used by means of paper pasted over all its apertures. The 
steam-cylinder was double-acting, 10 inches in diameter, with 18 inches 
stroke of piston, and was connected directly to the air-pump, which was 
also double-acting, 14 inches in diameter, with 18 inches stroke of piston. 
The pistons were worked at 30 single strokes per minute. The pressure 
of the compressed air in the air-pipe, just beyond the pump, was 
pounds per square inch above the atmosphere; and the injector-valve 
was counter-loaded to the same. 
The greatest diameter of the valve of the injector was 2| inches, and 
it was supplied with compressed air through a pipe of 2 inches inner 
diameter; the receiver into which this valve discharged consisted of a 
cylindrical pipe, 9 inches in inner diameter and 24 inches in length, with 
its ends formed into conical frusta of 15 inches extreme diameter and 
3 inches length; the entire length of the receiver being 30 inches. It 
was placed in a horizontal ventilating pipe of 15 inches inner diameter 
and 10 feet length, forming a communication between the atmosphere 
and the compartment at one side’ near the lloor. 
The outlet from the compartment was near its top, and consisted of a 
rectangular aperture 18 indies by 18 inches, in which was placed a del- 
icate helical anemometer that recorded the velocity of the outgoing air. 
Assuming the air-pump to have filled to nine-tenths of its capacity, and 
taking the vena contracts of the rectangular outlet at five-eighths, each 
volume of air delivered by the pump entrained or induced about twenty- 
five volumes of external air. 
When the pump delivered the same quantity in equal time of the same 
compressed air directly into the compartment without passing through the 
injector, for which experiment provision had been made in the apparatus, 6 
the difference of effect was very marked, the ventilation being greatly 
poorer, and the pressure in the compartment falling lower than was 
required to turn the anemometer, which accordingly remained motion- 
less. In fact, with the same expenditure of power, and the same com- 
pression of air in the same air-pump, both as regards density and bulk, 
twenty-six times the volume of fresh air was thrown into the compart- 
ment when the injector was used in connection with the air-pump than 
when the pump alone was used. This is an immense gain in both the 
efficiency and the economy of the ventilation. The whole of the addi- 
tional twenty-five volumes was clear gain, and cost absolutely nothing 
above the cost of producing the one volume by the direct action of the 
pump. 
The air compressed in the pump is heated to the thermal equivalent 
of the work done upon it by the compression. With an extended experi- 
ment, the temperature of the pump might be thus increased to an ob- 
jectionable degree, for its loss of heat by radiation might be less than 
its gain of heat from the compressed air; should such prove to be the 
fact, the cylinder of the pump could be surrounded by a jacket through 
which a circulation of cold sea-water could be maintained by a special 
pump with but little expenditure of power. 
With a given injector, the efficiency of the induced ventilation, or weight 
of fresh air entrained by the injector into a compartment, without regard 
to economy in the power producing this ventilation, will be in the ratio 
of the velocity of the air issuing from the injector, and of some function 
of the mass of air injected in equal times. This mass, with the same 
injector and pressure, will be in proportion to the width of the annular 
opening made by the valve. Air compressed to two atmospheres above 
zero discharges into the atmosphere with a velocity of 1,173.61 feet per 
second, and this velocity remains constant at all higher pressures; there* 
fore, any increase of pressure in the pump beyond 15 pounds per 
square inch above the atmosphere would not cause any increase in the 
efficiency of the induced ventilation. Under the pressure of 41 pounds 
per square inch above the atmosphere, as adopted by Mr. Green, air 
rushes into the atmosphere with the velocity of 649.61 feet per second, 
which is a reduction of 44.65 per centum below the maximum in the 
efficiency of the induced ventilation, but with a great increase in the 
practicability of the system and ease in managing it, together with an 
enormous increase of economy in the power-cost of the ventilation pro- 
duced. At the low pressure of 41 pounds per square inch above the 
atmosphere, but little difficulty would probably be experienced from the 
heating of the air produced by its compression. 
The weight of air, other things equal, received by the compartment 
through the injector from the pump, will, with equal speed of pump 
piston, of course, be the same, let the degree to which the air is com. 
pressed be what it may, the breadth of the annular space made by the 
valve through which the compressed air is discharged being necessarily 
reduced, automatically, proportionally to the pressure. The efficiency 7 
of the direct ventilation, therefore, as this air-supply may be called, with 
the same injector, as far as the weight of air furnished directly from the 
pump is concerned, will not be affected by the degree of compression ; 
but the expenditure of power producing the compression will be in the 
direct ratio of the degree of compression, twice the compression requir- 
ing twice the power. Hence, the less the compression the greater the 
economy with which the ventilation is produced for all pressures over 15 
pounds per square inch above the atmosphere ; below that, the cost is 
no longer pro rata with the compression, the effect of the lessening ve- 
locity of the induced air current entering now as a factor. It has been 
shown that all the increased ventilation which could be obtained by in- 
creasing the air-pressure in the pump from 44 to 15 pounds per square 
inch above the atmosphere would be what was due to the difference iu 
the effluent velocities in the two cases, namely, 619.61 and 1,173.61 feet 
per second, or 44.65 per centum of the induced ventilation; but this 
would be obtained at an increased power-expenditure in the ratio of 
( —2 : —~— \ o4 per centum. 
Both the efficiency and the economy of the ventilation produced by 
injecting air into a compartment with pressures from 4J to 15 pounds 
per square inch above the atmosphere is from 4 to 5 per centum greater 
than that obtained by exhausting the air, because, in the former case, 
the direct ventilation is utilized, while in'the latter it is not. 
The system of ventilation proposed by Mr. Green is not original, ex- 
cept so far as relates to the particular form of his automatic valve. This 
system, as applied to the ventilation of compartments, seems to have 
been first suggested by Mons. Mondesir, and was afterwards proposed 
by Mons. Decante to the French Government for the ventilation of the 
French Horse Transports, as appears from an account in the July, 1872, 
number of the Revue Maritime et Coloniale. It is highly recommended 
by Mons. J. B. Fonssagrives, retired chief of the Medical Department of 
the French Marine, in the second edition, published in 1877, of his clas - 
sical u Traite d’hygiene navale.” 
The same system was applied in 1863 by the presiding officer of this 
board to the furnaces of an experimental boiler in the New York navy- 
yard, for the purpose of forcing the combustion. In this case, however, 
steam was used as the injecting vehicle instead of air, on account of the 
facility with which it could be obtained under considerable pressure from 
the boiler itself, without requiring additional apparatus to produce the 
compression. The mingled steam and air, in very small proportion of 
the former to very large proportion of the latter, were delivered into the 
front of the closed ashpits of the boiler by means of ordinary injectors 
encircled by conical tubes. At the same date an experiment was tried 
with the same object on the same boiler, by forcing or blowing atmos 
pheric air up the chimney through an injector by means of an ordinary 
Dimpfel blower. Although this blower delivered air of very slight pre s s 
ure, the result was, in measure, satisfactory; but the air of this low pressure was so wanting in velocity that the apparatus would be too 
bulky for use on board a vessel. Circumstances prevented the continu- 
ance of these experiments which were to have been extended, varied, 
and enlarged, and which might otherwise have produced results of value 
in connection with the ventilation of naval ships, as they were being 
made in the right direction. 
The only vessels to which a complete system of ventilation has been 
applied in the United States Navy were the monitors. This system was 
also devised by the presiding officer of this board, and was first applied 
by him at the Washington navy-yard to the original historic Monitor 
shortly after her action with the Merrimack, and afterwards to all the 
iron-clads, from a general design furnished by him. It consisted in 
drawing, by means of ordinary Dimpfel blowers, the atmospheric air 
down an armored pipe extending from the deck to a level with the top 
of’the turrets, and then blowing it into the different compartments of 
the vessel at large through horizontal pipes of adequate cross-section 
provided with the necessary registers; the stagnant air thus forced out 
of the compartments escaped up the turrets. 
This was a very simple system, and it acted with great efficiency, 
giving the vessels perfect ventilation. They had no natural ventilation 
whatever, and depended vrholly on the artificial ventilation for habita- 
bility. The blowers were driven through belts by small - cylinders, 
and were geared to make fivd revolutions to each double stroke of the 
steam-pistons. 
It is, of course, possible with this system to perfectly ventilate any 
vessel, so that in absolute efficiency it is equal to the best which can be 
devised; but economically it is greatly inferior to the system by in- 
jectors, requiring a very much greater expenditure of power and very 
much bulkier piping. 
With the injector system there is the same small steam-cylinder as 
with the blower system, but it directly works an air-pump which replaces 
the blower. The quantity of atmospheric air to be compressed in the 
pump being only from 4 to 5 per centum of what has to pass through 
the blower, the expenditure of power is proportionally less. With the 
blower, air-pipes of the full capacity for the air required for the entire 
ventilation have to start from the blower, but proceed with diminishing 
area of cross-section from register to register. With the injector only 
a small pipe for the compressed air is used, even when starting from the 
pump, and it too has a diminishing area of cross-section from injector 
to injector. The money cost of the apparatus in the two cases will not 
differ to any extent of practical importance, but the enormous reduction 
in the bulk of the ventilating pipes is of the first consequence on board 
ship, as is also the equally large reduction in the coal cost of the power 
expended on the ventilation. In either case the cost of the ventilation 
in coal is not material, so that for ship’s use the great advantage of the 
injector system is in the very greatly less bulk of its apparatus. 
If artificial ventilation is to be generally adopted in naval vessels, 9 
then an exhaustive series of experiments should be made with the 
injector system, to ascertain the best proportions for the apparatus, and 
the laws governing its application. Such experiments would be costly 
in time and money, but they would be found true economy in the end. The 
same remarks apply to the blower system, little or nothing concerning 
which is known, as relates to the absolute efficiency obtainable by it, or 
to the cost in power of the ventilation produced with it. The two sys- 
tems appear to be the best, or, rather, the only ones practicable on board 
ship. The blower lias the advantage of extreme simplicity, but the dis- 
advantage of greater bulk of apparatus and greater power-cost for the 
production of a given ventilation. The injector system has the advan- 
tage of a less bulky apparatus aud less power-cost for the production of 
a given ventilation, but the apparatus is more complicated and more 
delicate. The absolute power-cost of producing a given ventilation with 
either system, and, consequently, the absolute money cost, are at pres- 
ent unknown. 
The system employed by Mr. Green has not yet been patented, the 
case still pending before the Patent Office, so that the particular points 
of originality which may be allowed by that office have not been deter- 
mined. It is impossible, however, to sustain a claim to the system of 
ventilating by injectors, and the patentable portion will probably be 
confined to the use of the particular automatic counter-loaded throttle- 
valve hereinbefore described. Any other valve producing an equiva- 
lent effect, or the use of an ordinary throttle-valve depending for adjust- 
ment on manipulation by an attendant, can be employed without either 
legal or moral invasion of Mr. Green’s rights. 
The effect of Mr. Green’s valve is to maintain always the same pres- 
sure of compressed air at the point of discharge into the receiver, let 
the speed at which the compressing air-pump works be as variable as it 
may, owing to variations in the boiler-pressure. With a constant boiler- 
pressure the air-pump would work at a constant speed, and always supply 
the injector with compressed air of the pressure to which an ordinary 
throttle-valve might be regulated, which latter being once adjusted 
would require no further attention. In this case the ordinary throttle- 
valve would be in every way as efficient as Mr. Green’s valve. How, a 
sufficiently near uniform pressure of steam for practical purposes upon 
the piston of the small steam-piston driving the air-pump, can be ob- 
tained by means of a “governor” applied to the throttle-valve of each 
cylinder; and with a uniform pressure upon the steam-piston the air- 
pump piston would have with its invariable load a practically uniform 
speed, so that equally good results can thus be obtained without the use 
of Mr. Green’s valve, from an ordinary throttle-valve throttling the air- 
pressure. Mr. Green’s valve, however, is a most elegant and appropri- 
ate device for accomplishing the purpose, and accomplishes it perfectly 
without the assistance of other apparatus. 
As an essential portion of this subject we have calculated and placed 
in the following table the velocities in feet per second with which air compressed to the tabular pressures flows into the atmosphere. In thus 
flowing, compressed air follows two entirely different laws, according as 
the compression is less or more than that of two atmospheres above zero. 
For the first case, the effluent air at the moment of entering the at- 
mosphere has the pressure of one atmosphere above zero, let its com- 
pression in the compressing vessel be what it may between the limits of 
one and two atmospheres above zero, so that the weights of air dis- 
charged within these limits, other things equal, are in the direct ratio of 
the tabular velocities, which velocities are themselves in the ratio of the 
square roots of the differences of the atmospheric pressure and the 
pressures of the compression, both counted from zero and at constant 
temperature. 
For the second case, the effluent air at the moment of entering the 
atmosphere has exactly the same velocity let its compression be what 
it may above the limit of two atmospheres above zero; the velocity 
being uninfluenced by any greater compression than two atmospheres 
above zero, and consequently constant, the temperature being supposed 
constant. But, in this case, the effluent air at the moment of entering 
the atmosphere has exactly half the pressure to which it was com- 
pressed in the compressing vessel; consequently, for all compressions 
above that of two atmospheres above zero, the weight of compressed 
air flowing into the atmosphere, other things equal, will be in the direct 
ratio of the compressions above zero. 
In making the calculations, the height of a homogeneous column of 
air of average composition, including aqueous vapor and carbonic-acid 
gas, weighing one pound per square inch, has been taken at 1457. 6525 
feet; and 2g. has been taken at 64£. The atmospheric pressure has been 
taken at 14.688 pounds per square inch, or at 29.92 inches of mercury. 
Table of the velocities with which air compressed to the tabular pressures flows into the 
atmosphere. 
Pressure of tlie 
compressed air 
in the com- 
pressing vessel 
in pounds per 
square inch 
above zero, at 
constant tem- 
perature. 
Pressure of the 
compressed air 
in the com- 
pressing vessel 
in pounds per 
square inch 
above the at- 
mosphere, at 
constant tem- 
perature. 
Velocity with 
which the com- 
pressed air 
flows into the 
atmosphere in 
feet per sec- 
ond. 
Relative weights 
of compressed 
air discharged 
in equal times 
into the atmos- 
phere, from the 
same orifice. 
15. 688 
1 
306. 23 
1. 0000 
16. 688 
2 
433. 07 
1. 4142 
17. 688 
3 
530. 40 
1. 7320 
18. 688 
4 
612. 45 
2. 0000 
19. 688 
5 
684. 75 
2. 2361 
20. 688 
6 
750.10 
2. 4495 
21. 688 
7 
810. 20 
2. 6457 
22. 688 
8 
866.14 
2. 8284 
23. 688 
9 
918. 69 
3. 0000 
24. 688 
10 
968. 33 
3.1621 
25. 688 
11 
1015. 64 
3. 3166 
26.688 ' 
12 
1060. 80 
3. 4640 
27. 688 
13 
1104.12 
3. 6055 
28. 688 
14 
1145. 80 
3. 7416 
29. 376 
14. 688 
1173. 61 
3. 8324 
29. 688 
15 
1173. 61 
3. 8731 
30. 688 
16 
1173. 61 
4. 0036 
31. 688 
17 
1173. 61 
4.1340 We have been particular in the announcement of the foregoing true 
laws governing the flow of compressed air into the atmosphere, and in 
the making of the corresponding computations, because these laws are 
not correctly given in treatises on the dynamics of elastic fluids. 
Very respectfully, your obedient servants, 
B. F. I SHERWOOD, 
Chief Engineer, U. S. N. 
THEO. ZELLER, 
Chief Engineer, U. S. N. 
HENRY L. SNYDER, 
Chief Engineer, U. S. N. 
Engineer-in-Chief Wi. H. Shock, U. S. N., 
Chief of the Bureau of Steam Engineering, 
Navy Department. 
Commandant’s Office, Navy- Yard, New York. 
Forwarded July 31, 1870. 
W. W. QUEEN, 
Captain Commanding. if mum f 
SECTION OF VALVES [A B,C,JFigure 2] 
ABC Air Values 
D Air Inlet 
E Air Distributor 
F E xhdust mt>L Pipes to Rooms 
G E xh&ust without Pipes to Boons 
H H DedC 
P Pump 
K Disclarpe of Air Power -from Pmp 
L Suction Pipe to Pump 
Fmum 
(o) 
V Outsiie A ir to Suction L 
S S S A cLj lxrfcdbl e A xr Pipes 
N Steel Spring 
R Rod or Tube for Adjusting Pressure 
T Cap Covering End of Veuve Stem 
U Beariud at U ton Vatire S tem 
X Air Inlet to Valve 
Y 0 wild of Air from Valve 
TFF W£MT!lLATm@ BBMPAIMf 9 HEW YORK a