load can be maintained upon one machine of several running in parallel.
This may be carried out as follows: Suppose, in a station having a total
capacity of 20,000 kilowatts, there are three machines, two of 6000
kilowatts each, and one of 8000 kilowatts, and it is desired to carry
out a steady full-load test upon one of the 6000 kilowatts units.
Assuming that the test is to be of six hours' duration, and that the
conditions of load fluctuations upon the station are well known, the
first step to take is to select a period for the test during which the
total load upon all machines is not likely to fall below, say, 8000
kilowatts. The tension upon the governor spring of the turbine to be
tested must then be adjusted so that the machine on each peak load is
taxed to its utmost normal capacity; and even when the station load
falls to its minimum, the load from the particular machine shall not be
released sufficiently to allow it to fall below 6000 kilowatts. Under
these conditions, then, it may be assumed that although the load on the
test machine will vary, it cannot fall below 6000 kilowatts. Therefore,
all that remains to be done to insure a perfectly steady load equal to
the normal full load of the machine, or 6000 kilowatts, is to fix the
main throttle or governing valve in such a position that the steam
passing through at constant pressure is just capable of sustaining full
speed under the load required. When this method is adopted, it is
desirable to fix a simple hight-adjusting and locking mechanism to the
governing-valve spindle. The load as read on the indicating wattmeter
can then be very accurately varied until correct, and farther varied, if
necessary, should any change occur in the general conditions which might
either directly or indirectly bring about a change of load.
Preparing the Turbine for Testing
All preliminary labors connected with a test being satisfactorily
disposed of, it only remains to place the turbines under the required
conditions, and to then proceed with the test. For the benefit of those
inexperienced in the operation of large turbines, we will assume that
such a machine is about to be started for the purpose outlined.
It is always advisable to make a strict practice of getting all the
auxiliary plant under way before starting up the turbine. In handling a
turbine plant the several operations might be carried through in the
following order:
(1) Circulating oil through all bearings and oil chambers.[5]
(2) Starting of condenser circulating-water pumps, and continuous
circulation of circulating water through the tubes of condenser.
(3) Starting of pump delivering condensed steam from the condenser
hot-well to weighing tanks.
(4) Starting of air pump, vacuum being raised as high as possible within
condenser.
(5) Sealing of turbine glands, whether of liquid or steam type, no
adjustment of the quantity of sealing fluid being necessary, however, at
this point.
(6) Adjustment of valves on and leading to the water-weighing tanks.
(7) Opening of main exhaust valve or valves between turbine and
condenser.
(8) Starting up of turbine and slowly running to speed.
(9) Application of load, and adjustment of gland-sealing steam.
[5] In a self-contained system, where the oil pump is usually driven
from the turbine spindle, this would of course be impossible. In the
gravity and allied systems, however, it should always be the first
operation performed. The tests for oil consumption, described
previously, having been carried out, it is assumed that suitable means
have been adopted to restrict the total oil flow through the bearings to
a minimum quantity.
The running to speed of large turbo-alternators requires considerable
care, and should always be done slowly; that is to say the rate of
acceleration should be slow. It is well known that the vibration of a
heavy unit is accompanied by a synchronous or non-synchronous vibration
of the foundation upon which it rests. The nearest approach to perfect
synchronism between unit and foundation is obtained by a gradual rise in
speed. A machine run up to speed too quickly might, after passing the
critical speed, settle down with little visible vibration, but at a
later time, even hours after, suddenly begin vibrating violently from no
apparent cause. The chances of this occurring are minimized by slow and
careful running to speed.
Whether the machine being tested is one of a number running in parallel,
or a single unit running on a steady water load, the latter should in
all cases be thrown on gradually until full load is reached. A
preliminary run of two or three hours--whenever possible--should then be
made, during which ample opportunity is afforded for regulating the
conditions in accordance with test requirements. The tester will do well
during the last hour of this trial run to station his recorders at their
several posts and, for a short time at least, to have a complete set of
readings taken at the correct test intervals. This more particularly
applies to the electrical water, superheat and vacuum readings. In the
case of a turbo-alternator the steadiness obtainable in the electrical
load may determine the frequency of readings taken, both electrical and
otherwise. On a perfectly steady water-tank load, for example, it may
be sufficiently adequate to read all wattmeters, voltmeters, and
ammeters from standard instruments at from one- to two-minute intervals.
Readings at half-minute intervals, however, should be taken with a
varying load, even when the variation is only slight.
The water-measurement readings may of course be taken at any suitable
intervals, the time being to an extent determined by the size of the
measuring tanks or the capacity of the weighing machine or machines.
When designing the measuring apparatus, the object should be to
minimize, within economical and practical range, the total number of
weighings or measurements necessary. Consequently, no strict time of
interval between individual weighings or measurements can be given in
this case. It may be said, however, that it is not desirable to take
these at anything less than five-minute intervals. Under ordinary
circumstances a three- to five-minute interval is sufficient in the case
of all steam-pressure, vacuum--including mercurial columns and
barometer--superheat and temperature readings.
Gland and Hot-Well Regulation
There are two highly important features requiring more or less constant
attention throughout a test, namely the gland and hot-well regulation.
For the present purpose we may assume that the glands are supplied with
either steam or water for sealing them. All steam supplied to the
turbine obviously goes to swell the hot-well contents, and to thus
increase the total steam consumption. The ordinary steam gland is in
reality a pressure gland. At both ends of the turbine casing is an
annular chamber, surrounding the turbine spindle at the point where it
projects through the casing. A number of brass rings on either side of
this chamber encircle the spindle, with only a very fine running
clearance between the latter and themselves. Steam enters the gland
chamber at a slight pressure, and, when a vacuum exists inside the
turbine casing, tends to flow inward. The pressure, however, inside the
gland is increased until it exceeds that of the atmosphere outside, and
by maintaining it at this pressure it is obvious that no air can
possibly enter the turbine through the glands, to destroy the vacuum.
The above principle must be borne in mind during a test upon a turbine
having steam-fed glands. Perhaps the best course to follow--in view of
the economy of gland steam consumption necessary--is as follows:
During the preliminary non-test run, full steam is turned into both
glands while the vacuum is being raised, and maintained until full load
has been on the turbine for some little time. The vacuum will by this
time have probably reached its maximum, and perhaps fallen to a point
slightly lower, at which hight it may be expected to remain, other
conditions also remaining constant. The gland steam must now be
gradually turned off until the amount of steam vapor issuing from the
glands is almost imperceptible. This should not lower the vacuum in the
slightest degree. By gradual degrees the gland steam can be still
farther cut down, until no steam vapor at all can be discerned issuing
from the gland boxes. This reduction should be continued until a point
is reached at which the vacuum is affected, when it must be stopped and
the amount of steam flowing to the gland again increased very slightly,
just enough to bring the vacuum again to its original hight. The steam
now passing into the glands is the minimum required under the
conditions, and should be maintained as nearly constant as possible
throughout the test. Practically all steam entering the glands is drawn
into the turbine, and thence to the condenser, and under the
circumstances it may be assumed the increase in steam consumption
arising from this source is also a minimum.
There is one mechanical feature which has an important bearing upon the
foregoing question, and which it is one of the tester's duties to
investigate. This is illustrated in Fig. 71, which shows a turbine
spindle projecting through the casing. The gland box is let into the
casing as shown. Brass rings A calked into the gland box encircle the
shaft on either side of the annular steam space S. As the clearance
between the turbine spindle and the rings A is in a measure instrumental
in determining the amount of steam required to maintain a required
pressure inside the chamber, it is obvious that this clearance should be
minimum. An unnecessarily large clearance means a proportionally large
increase in gland steam consumption and _vice versa_.
[Illustration: FIG. 71]
When the turbine glands are sealed with water, all water leakage which
takes place into the turbine, and ultimately to the condenser hot-well,
must be measured and subtracted from the hot-well contents at the end of
a test.
The foregoing remarks would not apply to those cases in which the gland
supply is drawn from and returned to the hot-well, or a pipe leading
from the hot-well. Then no correction would be necessary, as all water
used for gland purposes might be assumed as being taken from the
measuring tanks and returned again in time for same or next weighing or
measurement.
General Considerations
There are a few principal elementary points which it is necessary always
to keep in mind during the conduct of a test. Among these are the
effects of variation in vacuum, superheat, initial steam pressure, and,
as already indicated, in load. There exist many rules for determining
the corrections necessitated by this variation. For example, it is often
assumed that 9 degrees Fahrenheit, excess or otherwise, above or below
that specified, represents an increase or reduction in efficiency of
about 1 per cent. It is probable that the percentage increase or
decrease in steam consumption, in the case of superheat, can be more
reliably calculated than in other cases, as, for example, vacuum; but
the increase cannot be said to be due solely to the variation in
superheat. In other words, the individuality of the particular turbine
being tested always contributes something, however small this something
may be, to the results obtained.
These remarks are particularly applicable where vacuum is concerned.
Here again rules exist, one of these being that every additional inch of
vacuum increases the economy of the turbine by something slightly under
half a pound of steam per kilowatt-hour. But a moment's consideration
convinces one of the utter unreliability of such rules for general
application. It is, for instance, well known that many machines, when
under test, have demonstrated that the total increase in the water rate
is very far from constant. A machine tested, for example, gave
approximately the following results, the object of the test being to
discover the total increase in the water rate per inch decrease in
vacuum:
From 27 inches to 26 inches, 4.5 per cent.
From 26.2 inches to 24.5 inches, 2.5 per cent.
This illustrates to what an extent the ratio of increase can vary, and
it must be borne in mind that it is very probable that the variation is
different in different types and sizes of machines.
There can exist, therefore, no empirical rules of a reliable nature upon
which the tester can base his deductions. The only way calculated to
give satisfaction is to conduct a series of preliminary tests upon the
turbine undergoing observation, and from these to deduce all information
of the nature required, which can be permanently recorded in a set of
curves for reference during the final official tests.
In conclusion, it must be admitted that many published tests outlining
the performances of certain makes of turbine are unreliable. To
determine honestly the capabilities of any machine in the direction of
steam economy is an operation requiring time, and unbiased and accurate
supervision. By means of such assets as "floating quantities," short
tests during exceptionally favorable conditions, and disregard of the
vital necessity of running a test under the proper specified conditions,
it is comparatively easy to obtain results apparently highly
satisfactory, but which under other conditions might be just the
reverse. These considerations are, however, unworthy of the tester
proper.
VII. AUXILIARIES FOR STEAM TURBINES[6]
[6] Contributed to _Power_ by Thomas Franklin.
The Jet Condenser
The jet condenser illustrated in Fig. 72 is singularly well adapted for
the turbine installation. As the type has not been so widely adopted as
the more common forms of jet condenser and the surface types, it may
prove of interest to describe briefly its general construction and a few
of its special features in relation to tests.
[Illustration: FIG. 72]
Referring to the figure, C is the main condenser body. Exhaust steam
enters at the left-hand side through the pipe E, condensing water
issuing through the pipe D at the opposite side. Passing through the
short conical pipe P, the condensing water enters the cylindrical
chamber W and falls directly upon the spraying cone S. The hight of this
spraying cone is determined by the tension upon the spring T, below the
piston R, the latter being connected to the cone by a spindle L. An
increase of the water pressure inside the chamber W will thus compress
the spring, and the spraying cone being consequently lowered increases
the aperture between it and the sloping lower wall of the chamber W,
allowing a greater volume of water to be sprayed. The piston R
incidentally prevents water entering the top vapor chamber V. From the
foregoing it can be seen that this condenser is of the contra-flow type,
the entering steam coming immediately into contact with the sprayed
water. The perforated diaphragm plate F allows the vapor to rise into
the chamber V, from which it is drawn through the pipe A to the air
pump. A relief valve U prevents an excessive accumulation of pressure
in the vapor chamber, this valve being obviously of delicate
construction, capable of opening upon a very slight increase of the
internal pressure over that of the atmosphere. Condensed steam and
circulating water are together carried down the pipe B to the well Z,
from which a portion may be carried off as feed water, and the remainder
cooled and passed through the condenser again. Under any circumstances,
whether the air pump is working or not, a certain percentage of the
vapor in the condenser is always carried down the pipe B, and this
action alone creates a partial vacuum, thus rendering the work of the
air pump easier. As a matter of fact, a fairly high vacuum can be
maintained with the air pump closed down, and only the indirect pumping
action of the falling water operating to rarify the contents of the
condenser body. It is customary to place the condenser forty or more
feet above the circulating-water pump, the latter usually being a few
feet below the turbine.
Features Demanding Attention
When operating a condenser of this type, the most important features
requiring preliminary inspection and regulation while running are:
(a) Circulating-water regulation.
(b) Freedom of all mechanical parts of spraying mechanism.
(c) Relief-valve regulation.
(d) Water-cooling arrangements.
The tester will, however, devote his attention to a practical survey of
the condenser and its auxiliaries, before running operations commence.
A preliminary vacuum test ought to be conducted upon the condenser body,
and the exhaust piping between the condenser and turbine. To accomplish
this the circulating-water pipe D can be filled with water to the
condenser level. The relief valve should also be water-sealed. Any
existing leakage can thus be located and stopped.
Having made the condenser as tight as possible within practical limits,
vacuum might be again raised and, with the same parts sealed, allowed to
fall slowly for, say, ten minutes. A similar test over an equal period
may then be conducted with the relief valve not water-sealed. A
comparison of the times taken for an equal fall of vacuum in inches,
under the different conditions, during the above two tests, will reveal
the extent of the leakage taking place through the relief valve. It
seems superfluous to add that the fall of vacuum in both the foregoing
tests must not be accelerated in any way, but must be a result simply of
the slight inevitable leakage which is to be found in every system.
On a comparatively steady load, and with consequently only small
fluctuation in the volume of steam to be condensed, the conditions are
most favorable for regulating the amount of circulating water necessary.
Naturally, an excess of water above the required minimum will not affect
the pressure conditions inside the condenser. It does, however, increase
the quantity of water to be handled from the hot-well, and incidentally
lowers the temperature there, which, whether the feed-water pass through
economizers or otherwise, is not advisable from an economical
standpoint. Thus there is an economical minimum of circulating water to
be aimed at, and, as previously stated, it can best be arrived at by
running the turbine under normal load and adjusting the flow of the
circulating water by regulating the main valve and the tension upon the
spring T. Under abnormal conditions, the breakdown of an air pump, or
the sudden springing of a bad leak, for instance, the amount of
circulating water can be increased by a farther opening of the main
valve if necessary, and a relaxation of the spring tension by hand; or,
the spring tension might be automatically changed immediately upon the
vacuum falling.
The absolute freedom of all moving parts of the spraying mechanism
should be one of the tester's first assurances. To facilitate this, it
is customary to construct the parts, with the exception of the springs,
of brass or some other non-corrosive metal. The spraying cone must be
thoroughly clean in every channel, to insure a well-distributed stream
of water. Nor is it less important that careful attention be given to
the setting and operation of the relief valve, as will be seen later.
The obvious object of such a valve is to prevent the internal condenser
pressure ever being maintained much higher than the atmospheric
pressure. A number of carefully designed rubber flap valves, or one
large one, have been found to act successfully for this purpose,
although a balanced valve of more substantial construction would appear
to be more desirable.
Importance of Relief Valves
The question of relief valves in turbine installations is an important
one, and it seems desirable at this point to draw attention to another
necessary relief valve and its function, namely the turbine atmospheric
valve. As generally understood, this is placed between the turbine and
condenser, and, should the pressure in the latter, owing to any cause,
rise above that of the atmosphere, it opens automatically and allows the
exhaust steam to flow through it into the atmosphere, or into another
condenser.
A general diagrammatic arrangement of a steam turbine, condenser, and
exhaust piping is shown in Fig. 73. Connected to the exhaust pipe B,
near to the condenser, is the automatic atmospheric valve D, from which
leads the exhaust piping E to the atmosphere. The turbine relief valve
is shown at F, and the condenser relief valve at G. The main exhaust
valve between turbine and condenser is seen at H. We have here three
separate relief valves: one, F, to prevent excessive pressure in the
turbine: the second, D, an atmospheric valve opening a path to the air,
and, in addition to preventing excessive pressure accumulating, also
helping to keep the temperature of the condenser body and tubes low; the
third, the condenser relief valve G, which in itself ought to be capable
of exhausting all steam from the turbine, should occasion demand it.
[Illustration: FIG. 73]
Assuming a plant of this description to be operating favorably, the
conditions would of necessity be as follows: The valves F, D, and G, all
closed; the valve H open. Suppose that, owing to sudden loss of
circulating water, the vacuum fell to zero. The condenser would at once
fill with steam, a slight pressure would be set up, and whichever of the
three valves happened to be set to blow off at the lowest pressure would
do so. Now it is desirable that the first valve to open under such
circumstances should be the atmospheric valve D. This being so, the
condenser would remain full of steam at atmospheric pressure until the
attendant had had time to close the main hand-or motor-operated exhaust
valve H, which he would naturally do before attempting to regain the
circulation of the condensing water. Again, assume the installation to
be running under the initial conditions, with the atmospheric valve D
and all remaining valves except H closed.
Suppose the vacuum again fell to zero from a similar cause, and,
further, suppose the atmospheric valve D failed to operate
automatically. The only valves now capable of passing the exhaust steam
are the turbine and condenser relief valves F and G. Inasmuch as the
pressures at exhaust in the turbine proper, on varying load, vary over a
considerably greater range than the small fairly constant absolute
pressures inside the condenser, it is obviously necessary to allow for
this factor in the respective setting of these two relief valves. In
other words, the obvious deduction is to set the turbine relief valve to
blow off at a higher pressure than the condenser relief valve, even
when considering the question with respect to condensing conditions
only. In this second hypothetical case, then, with a closed and disabled
atmospheric valve, the exhaust must take place through the condenser,
until the turbine can be shut down, or the circulating water regained
without the former course being found necessary.
There is one other remote case which may be assumed, namely, the
simultaneous refusal of both atmospheric and condenser relief valves to
open, upon the vacuum inside the condenser being entirely lost. The
exhaust would then be blown through the turbine relief valve F, until
the plant could be closed down.
Although the conditions just cited are highly improbable in actual
practice, it can at once be seen that to insure the safety of the
condenser, absolutely, the turbine relief valve must be set to open at a
comparatively low pressure, say 40 pounds by gage, or thereabouts. To
set it much lower than this would create a possibility of its leaking
when the turbine was making a non-condensing run, and when the pressure
at the turbine exhaust end is often above that of the atmosphere. From
every point of view, therefore, it is advisable to make a minute
examination of all relief valves in a system, and before a test to
insure that these valves are all set to open at their correct relative
pressures.
It must be admitted that the practice of placing a large relief valve
upon a condenser in addition to the atmospheric exhausting valve is by
no means common. The latter valve, where surface condensing is adopted,
is often thought sufficient, working in conjunction with a quickly
operated main exhaust valve. Similarly, with a barometric condenser as
that illustrated in Fig. 72, the atmospheric exhaust valve D (seen in
Fig. 73) is sometimes dispensed with. This course is, however,
objectionable, for upon a loss of vacuum in the turbine, all exhaust
steam must pass through the condenser body, or the entire plant be
closed down until the vacuum is regained. The simple construction of the
barometric condenser, however, is in such an event much to its
advantage, and the passage of the hot steam right through it is not
likely to seriously warp or strain any of its parts, as might probably
happen in the case of a surface condenser.