Let us assume that it is desired to design a hydraulic dredge capable of raising average hydraulic material 15 feet above the pump through 4000 feet of pipe line and at the same time be suitable for general all round work involving also short lines and low heads.
It is generally accepted that a 20 or 22 inch pump is the most advantageous for these conditions. We shall select the 20 inch. The next step is to choose the desired velocity of flow through the 4000 feet of line. Ordinarily the velocity should not be less than 12 feet per second, but as this length of pipe is the ultimate condition and as the assumption of a high velocity here would result in an excessively high velocity for short lines even with a smaller impeller in the pump, we will fix 10 ft. per second as the proper figure for 4000 feet of line and an elevation head of 15 feet. Entering the table, Page 76, we find that for 10 feet velocity in a 20 inch pipe, the friction head per 100 feet of pipe is 1.87 feet. But this value is for water and must be multiplied by 1.35 to obtain the head for dredgings, which is 2.52 feet, and the friction head for 4000 feet of pipe is 40 x 2.52 or 100.8 feet. The pump should be able to hold a vacuum of about 12 inches of mercury.
The peripheral velocity of impeller necessary to develop this head is found by formula (1) Page 77.
P. V. = 435√130 = 4970. feet per minute
bearing in mind that the constant 435 involves the use of the usual type of impeller as previously described. The engine speed for dredges of this size on long lines appears to be generally from about 200 to 225 revolutions per minute. Let us assume for our problem a value of 220 R.P.M. Then the required diameter of impeller is 4970/220 x 3.142 or 7.19 feet or 86 inches.
The water horse power for this condition is from formula (2) Page 78, 9800 x 130/3960 = 322, and, for a pump efficiency of 50 per cent the engine brake horse power would be twice 322 or about 650.
Now let us investigate the same pump under a low head. Assume a short line, say 1200 feet and a low lift, 10 feet. Using the same impeller and engine speed as above, the developed head is again 130 feet, of which 13.6 feet is suction, 10 ft. elevation, and the balance, 106 ft. friction head. For a 1200 ft. line, the friction loss per 100 ft. of pipe is 8.83 ft. corresponding to a velocity of about 19.5 ft. per second, which is excessive. The remedy is either reduced engine speed or a smaller impeller or both. Suppose an impeller 78 in. in diameter be substituted for the 86 in. and driven at the same speed and under the same low head conditions as above.
P.V. = 6.5 x 3.142 x 220 = 4500
and the developed head = (4500/435)² = 107 ft. The friction head = 107 – 24 = 83 ft. or 6.91 ft. per 100. For water, the equivalent head = 6.91/1.35 = 5.12, and the velocity is again excessive, approximating 17 ft. per second requiring a shaft horse power of about 16,500 x 107 x 2/3960 = 900.
The high velocity may be reduced by increasing the size of the discharge pipe or by reducing the R.P.M. Still using the 78 inch impeller, but cutting the engine speed to 200 R.P.M. the total head developed is (6.5 x 3.142 x 220/435)² or 88 ft.; the friction head is 88 — 24 = 64 ft. or 5.33 ft. per 100 ft. of the 1200 ft. line. Dividing by 1.35 we enter the table with a value 3.95 and find that the corresponding velocity is almost 15 ft. per second and the discharge about 14,500 gallons per minute. The engine horse power required is 14,600 x 88 x 2/3960 or about 650.
The foregoing leads us to the following conclusions:— that, for the problem stated, the solution appears to be a 20 in. pump with 20 in. suction and discharge, having at least two sizes of impeller, about 78 in. for short and 86 in. for long lines, driven by a triple expansion steam engine of about 800 horse power, turning over from 200 to 225 R.P.M. Although this is a greater horse power than theoretically required according to the above figures, it is recommended because of the necessarily uncertain nature of the factors involved and the desire for reserve power to take care of severe pumping conditions.
The Machinery.—The pump drive must be capable of variable speed and of running at different speeds for long periods of time because the load on the pump is variable due to the fluctuating suction head, the nature of the material and the varying length of pipe line. A steam engine meets this condition fully and is admirably adapted to the purpose. If the pump is driven electrically the motor must be designed for this varying load. A synchronous motor is, therefore unsatisfactory. Although electrically driven machines have been successfully operated, the usual drive is a triple expansion vertical condensing engine, directly connected to the pump and called the main engine. The 20 in. dredge New Jersey, shown on Page 69, has cylinders 18 and 24 and 40 inches in diameter with a common stroke of 20 inches, 200 R.P.M., and develops 750 h.p. The 22 inch machine “TAMPA” pictured on Page 60, has cylinders 14 and 21½ and 36 inches, 18 inch stroke, 225 R.P.M. and 800 horse power. The boilers are variously Scotch, Heine or Almy water tube, and B. & W. with a working pressure of from 175 to 225 pounds.
Between the pump and the main engine, a thrust bearing on the shaft is required because of the axial pull of the impeller. The shaft bearing at the impeller is kept clean by a water service.
The cutter engine will be about 12 x 12 double cylinder for a 20 in. machine and the winding engine about 8¼ x 12 double cylinder, driving at least 5 drums for swinging wires, spuds and ladder hoist, or more if the dredge be of the swinging ladder type with walking spuds and stem wire.
In addition to the above, there are the condenser and centrifugal circulating pump, the air pump, the generator set for electric lighting and the other auxiliaries.
Operation.—The leverman has full control of the manipulation of the dredge. The lever-room is located well forward and at such an elevation that the leverman can keep a watchful eye on the swinging of the machine or ladder, the ranges to which he is working, a tide gage, the behavior of the cutter engine and the condition of the floating pipe line. In the lever-room, are the vacuum, discharge-pressure and steam-pressure gauges, which guide the operator in the normal operation of the pump and feed. The vacuum in the suction pipe is greater when pumping solids than when water only is passing through, becoming maximum when the suction is choked. The discharge pressure falls off for chokes in the suction and rises for obstructions in the discharge pipe. The leverman learns to keep the gauge readings at that point at which the pump will carry the maximum amount of material without choking. Both vacuum and pressure readings acquaint the operator of restricted suction, the former by rising and the latter by falling, but the vacuum gage is more sensitive than the pressure, responding more quickly to the abnormal condition. Through gratings in the floor of the lever-room, the operator keeps himself informed as to the amount of swinging wire left on the drums in the room below him. The depth of the cutterhead below the water surface is indicated by a sliding weight on the boom or by a dial in the lever- room, operated through reduction tackle.
Under the most favorable conditions, in mud and silt, hydraulic dredges may reach a maximum solid output of 25 to 30 per cent of the pumpings, but in average digging, the percentage will be more nearly 10 to possibly 15 and in heavy material with long lines it will fall as low as 5.
Booster or Relay Pumps.—When a discharge pipe line becomes so long as to reduce the output of the dredge below the economic minimum, material assistance may be rendered the dredge pump by placing in the shore line a second pump called a booster or relay pump. The dredge discharges directly into the suction of the booster, which then forces the pumpings through the remaining length of line between itself and the fill. Thus the dredge pump is relieved of the length of line beyond the booster and is enabled to accelerate the pipe velocity and to increase the quantity of discharge. It is apparent that, for maximum efficiency of the combination, the booster must discharge an amount equal to that delivered to it by the dredge. If the booster receives more than it can handle, the output of the dredge is decreased by retarded velocity, although it may be greater than in the long line without the booster. If the booster is capable of a greater discharge than that delivered to it, it will draw air into its suction and pump the mixture. It is difficult to make the ordinary dredge shore pipe air tight. If the dredge discharge is sufficiently less than that of the booster, the load on the latter will
be intermittent, i.e., the booster receives a column of pumpings and disposes of it so quickly that the continuity of the supply is broken, leaving a void between supply columns during which the impeller rumbles around in a mixture of air and water before receiving the next column, which is as quickly discharged. Consequently the discharge at the end of the pipe on the fill reveals a very appreciable pulsation. It is apparent, therefore, that there is one best place in the line at which to install a given booster, and the usual problem will be the determination of that point rather than to design or select a booster for a particular point, since neither are dredging operations so permanent nor boosters so plentiful as to afford frequent opportunities for the latter.
The solution of the problem again involves the question of peripheral speed of impellers. Knowing the diameter and R.P.M. of the dredge and booster impellers, select a point in the line for trial. Compute the pipe velocity created by the dredge for the length of line from dredge to booster. Compute the pipe velocity created by the booster for the length of line beyond it not forgetting that this length will vary, usually by the periodical addition of extra pipe sections. Reasonable variations in pipe length can be taken care of by varying the speed of the drives on either dredge or relay, or both. For large differences, a second impeller for the booster may be necessary. By a little manipulation the point in the line at which the relay must be located to insure equal pipe velocities of dredge and booster can be determined.
From the above, it is obvious that the booster drive must be capable of variable speed. If electrically driven, as most of them are, to facilitate transportation, installation and operation, the motor should be designed to run for long periods at various speeds without overheating. Synchronous motors are, therefore, wholly unsatisfactory for this purpose.
The pump is directly connected to the motor with a thrust bearing between the two. The latter need not be so heavy as that on the dredge since the water enters the pump normally without vacuum. It is desirable to have a flexible shaft coupling near the motor to provide for changes in alignment. The behavior of the pump should be carefully watched by means of vacuum and pressure gages on the suction and a pressure gage on the discharge. The vacuum gage should read zero for normal behavior. The load on the motor must also be noted by gauge. It is almost essential that there be a by-pass line around the booster so that it may be cut out of the line for repairs or other, causes without interruption to the dredge pump, and that it may be given the load gradually, in starting, to prevent an overload on the motor. The by-pass is effected by two “ Y” branches in the line, one beyond each end of the booster. The “Y” branch between the dredge and booster must have a gate or valve in each leg but that beyond need have a gate in the booster leg only. A typical installation is shown in Figure 35, page 84.
To throw the booster into the line, the procedure is as follows: Start the motor with valves B and C closed; have the dredge pump water only; leaving valve A open, open C partly then B about the same distance; open B wide; then C all the way; close A; have the dredge pump mud. The reasons are obvious. The minimum load on the motor occurs with the valve C closed, giving maximum head, zero discharge and therefore a minimum power requirement which is the desired starting condition. There is considerable danger, however, of blowing the line apart at C if pumping against that valve closed. Hence it is partially opened.
The installation should be housed, and it is essential that telephone communication or some system of signalling be established between dredge and booster so that the orders to stop, start, pump water and pump mud may be quickly conveyed.