US1374990A - Steam-ejector - Google Patents

Description

UNHT S'lf ALBERT DELAS, OF PARIS, FRANCE, ASSIGNOR TO DELAS CONDENSER CORPORATION,

OF NEW YORK, N. Y., A CORPORATION OF DELAWARE.

STEAM-EJECTOR.

Specification of Letters Patent.

Patented Apr. 19, 1921.

Application filed October 31, 1918. Serial No. 260,600

To all whom it may concern:

Be it known that I, ALBERT DELAS, a citizen of the Republic of France, residing at 18 Rue de Liege, Paris, France, have invented new and useful Improvements in Steam- Ejectors, of which the following is a full, clear, and exact description.

In the construction of ejectors of a high ratio of compression it has heretofore been the practice to give the diffuser a convergentdivergent form. This form however is only suitable when,- the normal ratio of compression has been attained, as the starting operation is very diflicult. It is difiicult to start because the density of the fluid in the nozzle chamber is much higher than that which must prevail under normal operation. Gonsequently, the jet of steam which is employed as the motive fluid, not only spreads out but loses a part of its speed by virtue of this denser atmosphere, and for a predetermined surface the weight of the fluid that it carries along is considerably increased, due to the high density of this fluid. Therefor, if the distance fromthe nozzle to the neck of the diifuser has been calculated and proportioned for the maximum efliciency' otherwise be required, in order that the steam, which spreads out at starting, should not touch. its inner wall.

I The apparatuses just described have an important bearing on the normal operation because, if the neck of the difiuser is .too large for normal operation, the fluid column forced into the difl'user-will have a contracted section which is smaller than the neck of the diifuser and will therefore not fit the sides of the diffuser; and from the moment that the fluid column leaves said sides eddies will be formed which interfere with the output and with the stability of the ejector. To overcome this builders have been obliged to admit extraneous air or other fluid into the diffuser in order that the latter may always be filled, thereby preventing the column from expanding and thus giving rise to eddies. The loss of efliciency due to the work necessary to compress this air or other fluid admitted is considerable but necessary to stabilize the apparatus. But it is obvious that the maximum of efliciency with such an apparatus is obtained only at the moment of starting and for the poor vacuum. After having used it exclusively certain builders have given up this process of stabilization as too costly and sacrifice both the efliciency and the stability by employing some intermediary. method.

Furthermore, in order to allow the starting in certain types of apparatus, it has been necessary to use severalnozzles, instead of one, and to reduce the total surface according to the laws governing the quantities of movement under normal operation. This again sacrifices the efliciency, and the use of several nozzles necessitates an extreme re--- duction in the dimensions of the neck of the nozzles-few tenths of a millimeter. These nozzles introduce another undesirable characteristic; that is, they are frequently ob structed because of some unfiltered deposits carried along by the motive fluid.

It may therefore be briefly stated that, among others, attempts to stabilize ejectors, the ratio of compression of which are neces sarily subject to fluctuations, have been made either by providing air inlets at'the throat of the diffuser or in the space in which the without varying the cross-sectional area of the throat.

I accomplish the foregoing results by maintaining the convergent part of the vapor ejector at a relatively low temperature, with a result that any eddies produced in the diffuser in the normal operation of the device, are chilled and condensed, and consequently improve the efliciency of the apparatus and the vacuum obtained. Furthermore, if the diffuser of said ejector is cooled, in the manner just stated, it may be made of such dimensions as to secure a maximum of efficiency for normal operation, and that when the weight of the fluid drawn in exceeds said working conditions the steam will condense in contact with the cooled wall and such condensation will prevent congestion I nozzle, the surplus working steam thus sent in condensing upon-the walls of the diffuser with a result that the operation and output of the apparatus will not be affected.

Referring to the drawings:

Figure 1 is a central longitudinal section of an old and well-known form or type of ejector.

Fig. 2 is a similar view of an ejector embodylng my improvements. Y

When the normal working conditions of an ejector of'a high ratio of compression has been attained, the gaseous stream takes the form illustrated in Fig. 1; that is to say, the gaseous stream lines issuing from the nozzles a are parallel, or approximately parallel,

with the portion 6 of the wall which connects the convergent part b with the divergent part 0 of the diffuser. However, when the pressure increases in the combining chamber the fluid stream expands and impinges against the convergent wall b of the diifuser with perceptible incidence. The portion of the stream which strikes the wall b therefore loses, in the ordinary form of ejector, a part or all of its kinetic energy with the result of a stopping or an upsetting of the apparatus. This may be obviated by reducing the quantity of steam, and increasing it only gradually as the ratio of compression increases. 7 But to produce such increase automatically is practically impossible. a v

I have found, therefore, that the' excess steam'must be eliminated at starting and to attain this result such excess steam is caused to con-dense at the point where it impinges fuser at the proper temperature to extract the heat due to the condensation of the steam. However, when the normal ratio of compression of the ejector has been attained the steam jet again becomes parallel with the wall a and the condensation is practically negligible. On the other hand, when the stream of working fluid swells or expands, due for instance, to an increase of pressure in the combining chamber, a portion of the working steam strikes the cold wall of the convergent part of the diffuser and condenses thereon.

By a very simple calculation it can be shown that the amount of work required in removing the water of condensation is insignificant, whereas the work required is considerable in compressing from vacuum to atmospheric pressure the same weight of vapor which has lost its kinetic energy.

From the foregoing description it will.

be clear that I have not only designed anapparatus but also provided a method by which the stability of an ejector is effected for all conditions of normal and abnormal operation and for the conditions which prevail in the starting of the apparatus, and at the same time enabling a maximum efiiciency of the apparatus when operating under normal conditions or working.

What I claim is: v 1

1. The process of ejecting a fluid which consists in passing a motive fluid through a diffuser, and, condensing that portion of the motive fluid which is in excess of that required to fill the neck portion of said diffuser. j

2. The process of ejecting a fluid which consists in passing a motive guid through a diffuser in a quantity sufiici nt to provide for the normal ratio of compression, and condensing that portion of said fluid which strikes the interior wall of said difl'user when the diffuser is not operating at said normal ratio of compression.

3. The process of ejecting a fluid which consists in passing a motive fluid through a diffuser in a quantity suflicient to provide for the normal ratio of compression, and

condensing any eddies produced in the difsaid fluid stream which expands and impinges against the interior wall of said diffuser.

5. The process of ejecting a fluid which 5 consists in passing a motive fluid stream through a diffuser in a quantity suficient to provide for the normal ratio of compression, and in condensing that portion of the fluid stream which is inexcess of what is required when not operating at said normal 10 ratio of compression.

In witness whereof I have hereunto signed my name. 7

ALBERT DELAS.

Images