Duan C.G., Karelin V.Y. Abrasive Erosion and Corrosion of Hydraulic machinery

Подождите немного. Документ загружается.

Abrasive Erosion and Corrosion of Hydraulic Machinery

k'd

-l)ev

t (1.17)

i.e with the specified sediment matter densities, and all other things being

equal, the intensity of hydraulic abrasion is proportional to the 3rd power of

the diameter of solid particles.

However, the dependence stated in equation (1.17) is in poor agreement

with the data obtained in number of experiments. The erosion curve, for

instance, plotted in relative values shows, according to, that with the particle

grain size exceeding 0.2 mm, the sand grain dimensions have no effect on

erosion rate (see Figure 1.5 a).

AG/AG^/o) Ih(%)

l*"^

0 0.2 0~4 ~~0~6 iUmm) 0 0.2 0.4 0.6 4(mm)

(,) (b)

(a) alteration of relative erosion

(b) erosion rate alteration

(1.

aluminum; 2. steel CT3)

Figure 1.5 Influence of the particle grain size on the hydraulic abrasion

On the other hand, when making use of sharp-sided particles, we shall

have different dependence (see Figure 1.5 a).

The experiment, carried out with utilization with utilization of sand

having grain size values 0.15, 0.3, 0.6 and 1.2 mm, has indicated (Figure 1.5

b) that with sediment concentrations and suspension-conveying mixture

velocities being permanent, the erosion rate is directly proportional to the

grain size and can be expressed as

h=I'

(1.18)

ijb

-—x

J5b

Fundamentals ofHydroabrasive Erosion Theory

where I'

is erosion rate per hour, as related to the sand grain size d

1 mm

and depending on concentration and mineral composition of sediments.

These data fit well with the results obtained for specimens fabricated form

different metals and tested in a rotary-type stand. The test results indicate that

when these specimens are affected by abrasive particles with average

diameters varying 0.1 to 0.5 mm, while they move with a constant speed in a

hydro-mixture having unchangeable concentration, the erosion increases in

direct proportion.

Such significant quantitative and qualitative differences in experimental

data and their discrepancy from the theoretical estimation can be explained by

decrease in actual motion velocity of solid particles c (see equation 1.1) at rise

in their size, in addition to the divergence in the methods used for carrying out

the experiments and difficulties encountered in defining of said divergences.

Assuming that the square of a particle velocity, as related to the liquid, is

proportional to the particle mass and also evaluating the deteriorating

capability of such a particle by its kinetic energy affecting the surface to be

disrupted, it is possible to fine that this energy will be utmost at c = 0.5 v, i.e.

as the particle mass increases, its energy rises at the beginning and then it

starts to decline.

In the actual motion conditions, existing in the inter-blade channels of

pump impellers and created by a suspension-conveying flow, such factors as

separation of solid particles and divergence between their paths and the flow

direction line are of great significance as well.

Within the limits of concentration change and that of grain sizes of solid

matter present and typical of natural waterways, see, for instance, the

granulometric composition described for the Vakhsh irrigation canal:

♦ Sediment particle

sizes (mm): < 0.01 0.01-0.05 0.05-0.1 0.1-0.25 0.25-0.5 >0.5

♦ Fracture

contents (%): 39.6 27.6 14.1 18.4 0.2 0.1

The relationship between the abrasion intensity, taking place in the

components of general application pumps, and grain size of sediment matter

particles should be assumed as linear

[1.14].

With reference to the density of solid particles it should be emphasized

that the experience, gained in the operation of these pumps, confirms the

Abrasive Erosion and Corrosion of Hydraulic Machinery

validity of equation (1.17). Therefore, with the heavier particles contained in

the water, it is assumed that the expected erosion should increase.

In addition to mass, governed by such parameters as size and density, and

hardness, another factor of great significance is the shape of moving abrasive

particles. It is known that solid particles with sharp sides are especially

hazardous in the sense of their ability to deteriorate the surface circumvented

by the flow. But to evaluate this factor quantitatively, with respect to its effect

on erosion intensity, is extremely difficult, since the particle shapes. As a rule,

vary continuously as a result of mutual collisions and their friction against the

surfaces restraining the flow.

As seen from equation (1.4), (1.9) and (1.13) the main factor governing

hydraulic abrasion intensity, at the given parameters of hydroabrasive

mixture, is the local velocity. However, the views, which many researchers

have on the velocity index of power n, are inconsistent with each other.

Although the theoretical analysis suggests that it is equal to the 3rd power, the

results obtained on experimental rotary-type stands correspond to n = 2.5 ~

3.0, those obtained on disk stands measure n = 1.8 ~ 2.7, and those obtained

jet-water impact stands are equals to n

2.0 ~ 2.2.

All the stands of the indicated types are unfit, in our view, for experiments

to be carried out with the aim of defining the effect of a hydro-mixture

velocity on the erosion of

pumps,

since to determine this velocity actual value

is an extremely difficult task. Besides, the influence pattern

itself,

evolved in

the effect performed by hydro-mixture on the surface to be worn,

predetermines emergence of vortexes and various secondary streams, which

distort the whole pattern of interaction taking place between the flow and the

surface. Therefore, in the experiments carried out in the V.V. Kuibyshev

Moscow Civil Engineering Institute (MIEI) we used a Venturi nozzle with

divergence angle 6 = 5°, Figure 1.6 illustrates loss in mass of aluminum

specimens at different velocities of a suspension-conveying flow containing

solid fracture p = 5 g/1 at d

= 0.1~ 0.25 mm. It is found that the index of the

power n is not constant but varies from 2.5 to 3.0 with increase of flow rate

absolute values. It should be noted that in carrying out the quantitative

evaluation of index n the results of our study are in good agreement with

experimental data obtained on hydraulic abrasion of cylindrical specimens

placed in straight pipelines.

Discrepancy of the results obtained can be explained by complexity in

determining the actual flow velocity nearby the surface being disrupted and

Fundamentals of Hydroabrasive Erosion Theory

the velocity of

the

solid particles contained in the flow, as well as due to the

presence of turbulence and boundary layer of variable thickness. The flow

velocity alteration affects the particle concentration nearby the surface (see

equation

1.11),

and the velocity rise, in this case, results in decrease of this

concentration and vice versa.

AG(mg)

150

100

0 24 26 28 30 v(m/s)

Figure 1.6 Dependence of the hydraulic abrasion on the flow rate

During the erosion of planar surfaces by a suspension-conveying flow

with small content of solid particles the erosion intensity, in our view, is

proportional to flow velocity v in power n, where 3 > n

The derivation of equation (1.4) is based on the assumption that an

abrasive particle collides with the surface of a streamlined component with a

normal impact. Alteration in angle corresponding to inclination of the velocity

vector with respect to the component surface, which is often named as an

angle of attack, is accompanied by respective alteration of the external effect

pattern produced and acting on the surface layer and by corresponding

quantitative and qualitative alterations in the surface deterioration process.

At the angle of incidence a = 90° the particles exert normal impacts

against the component surface. Due to differences existing in velocity, form,

mass and mechanical properties of particles, there appear, at the moment of

an impact, stresses of different orders in the surface layer of the material. As

this takes place, a certain part of the kinetic energy possessed by a solid

particle is consumed to produce elastic strain of the material, and the

remaining energy portion is expended for plastic deformation and material

deterioration, as well as for crushing of the solid particle under review.

Abrasive Erosion and Corrosion of Hydraulic Machinery

Normal impacts of abrasive particles are capable of producing brittle and

endurance failures material micro-volumes in the surface layer; failure of soft

materials can also take place and produced by shearing.

As the angle of attack reduces, the strength of particle impact decreases as

well, and its normal component becomes proportional to sina. A number of

authors assume that the tangential component in the impact impulse causes

the material failure only at those values of ctga, which exceed the coefficient

of friction produced by a particle against the surface of a specimen being

worn.

When these angles of incidence are small, the material failure is effected

in such cases by shearing or peeling.

As an example demonstration the process alteration, taking place during

deterioration of

the

surface layer material caused by alteration in the direction

of a stream flowing around a component. Figure 1.7 illustrates gas abrasive

erosion intensity versus angle of incidence relationship, where the specimens

tested were fabricated from hardened and non-hardened steel and from rubber.

These materials were tested on a laboratory stand. The airflow velocity was

set at 100 m/s and the quartz sand particles used had sizes 0.2 to 1.5 mm. It

should be noted here that the material erosion intensity dependence of the

angle of attack are qualitatively similar both in hydro-abrasive and gas-

abrasive processes.

At the incidence angles approaching normal ones, the hardened steel

(curve 1, Figure 1.7) erosions faster than it does at smaller angle, and with a

greater rate than the non-hardened steel and rubber do. This result from

brittleness of hardened steel; the surface layer in such steel cannot at certain

areas withstand, at set conditions of external influence pattern, the impacts

produced by a certain number of abrasive particles.

The soft steel specimens (curve 2) are likely to undergo, at the prescribed

conditions, a sort of poly-deformation failure process. The surface layer in

rubber specimens (curve 3) absorbs, at the present flow velocity of particles

moving with a = 90°, the greater part of the kinetic energy possessed by

abrasive particles, and the rubber in this case erosions out slower, as

compared with the other material,

While the erosion rate in hardened steel specimens reduces with decrease

of the attack angles, which is caused by gradual reduction of the normal

impact strength component, the deterioration process, which develops with

Fundamentals of Hydroabrasive Erosion Theory 19

reference to the steel and rubber, is accompanied at certain values of an angle

of incidence, with qualitative changes.

l^mm/h)

2.0

1.5

1.0

0.5

0 30 60 a

hardened steel, with hardness 840 kgf/mm

non-hardened steel, with hardness 128 kgf/mm

rubber, with Shore hardness 70 ~ 74

Figure 1.7 Dependence of the erosion rate in specimens of steel

and rubber on the solid particle angles of incidence

At a < 73° the erosion rate of soft steel rises due to increase of the

tangential component in the impact impulse, which results in the material

failure effected by shearing. At a < 30° the soft steel erosion rate falls down

again, which results from a continuous decrease, taking place in the normal

component of the impact strength. The rubber erosion rate rises drastically at

a < 64°, and at a = 28°, it already exceeds the erosion rate in the other

materials tested (curve 3 is placed above curves 1 and 2). At small angles of

incidence the rubber ability to absorb kinetic energy of solid particles reduces,

which results in the material failure.

The experimental results obtained have illustrated that during alterations

of incidence angles in the range 15° to 90°, the phenomena observed may

allegedly be classified as an inversion of erosion resistance resulted from

alterations in the external influence pattern affecting the surface layer. It can

be assumed that such erosion resistance inversion will also occur during

alterations in the abrasive action activity (specifically, at the appropriate

Abrasive Erosion and Corrosion of Hydraulic Machinery

alterations in mass, particle shapes and velocity of their motion). At a normal

impact, for instance, a certain combination of particle motion speed and

particle mass may arise, at which the rubber will practically remain unworn,

since the whole particle kinetic energy will be balanced with the elastic strain

of the rubber.

With reference to the effect on the erosion rate caused by duration of the

influence exerted by the hydro-abrasive mixture, there is linear dependence

between the deterioration volume and influence time observed practically in

all the experimental studies. The experiments carried out at the V.V.

Kuibyshev MIEI have proved that the linear nature of function I - f(t) is

preserved with various mixture concentrations (Figure 1.8 a) and solid

particle sizes (Figure 1.8 a) and solid particle sizes (Figure 1.8 b).

AG/G(%)

0 4 8 12 16 20 h 0 4 8 12 16 20 h

(») (b)

1p= 9g/I; 2./?=36g/l; 3.p=72g/\- 4.p= 108

g/1;

5.p= 126 g/1;

6.d

0.15-0.3 mm; l.d

= 0.3-0.6 mm; 8.d

0.6-1.2 mm;

9.d

=1.2 mm; (dash lines-aluminum, solid lines-steel)

Figure 1.8 Relative erosion versus hydro-abrasive mixture action duration

relationship at different concentration contents (a) and particle grain sizes (b)

Fundamentals ofHydroabrasive Erosion Theory

It should be noticed in conclusion that hydraulic abrasion intensity also

depends, undoubtedly, on the material type used to fabricate components

being worn out by a suspended matter conveying flow.

1.3 Abrasive Erosion of Hydraulic Turbine

1.3.1 Illustrative Examples of Hydraulic Abrasion in Hydraulic

Turbines

Let us consider, as examples typifying the hydraulic abrasion occurring in

flow passages of hydro-turbines, the results of regular full-scale study carried

out on hydroelectric units used in derivation water-power station No.l and

No.2 situated on one of the waterways in the Uzbek SSR (Central Asian

region of the USSR).

The following classification was used to evaluate the hydraulic abrasion

patterns:

Metallic luster - A shining surface with no

traces of

paint,

scale or rust

Fine-scaly erosion - A surface with rare, separately

located and skin-deep minute

scales

Scaly erosion - A surface entirely covered with

skin-deep fine scales

Large-sized scaly erosion - A surface entirely covered with

deep and enlarged scales

In-depth erosion - A surface covered with deep

and long channels

6. Through holes or entire erosion-out of the metal.

An average annual sediment concentration for No.l and No.2 accounts

for 1.27 kg/m

and this parameter range varies from 0.23 kg/m

(in

December) to 3.69 kg/m

(in June). The mineralogical composition analyses

have shown that exceeding 5, in accordance with Moh's scratch hardness

number scale, amounts to

41%

of the average annual hard matter content.

Abrasive Erosion and Corrosion of Hydraulic Machinery

The hydroelectric station No.l includes two sets accommodated with

radial-axial turbines having 1.8-m diameter, with the rated head being 34 to

37 m and maximum capacity of 22 kg/m

. The specified power values of

these two units are 6.5 and 5 MW, respectively. The hydro-turbine runners

are fabricated from stainless steel, the other elements in the flow-passage

portion of

the

device are made of carbon steel. The hydroelectric station No.2

has two similar units accommodated with radial-axial turbines with 2-m

diameter, the rated head being 21.5 m and maximum capacity equaling 23

kg/m

; the set power in there units is equivalent to 4.2 MW each. The runner

and other elements of the turbine are made of cast carbon steel.

The periodic full-scale surveys carried out during the scheduled

maintenance overhauls of the components of the hydraulic sets have revealed

that these elements have acquired, in the course of about 10 months of

operation, abrasion of the kind described below.

The spiral case in one of the hydroelectric units, made by welding from

sheet steel, has fine-scaly erosion throughout the entire surface, which is

increased at the approach to the turbine stator. It was found that this fine-

scaly erosion tends to increase over the entire surface in the spiral case. At the

joints of

the

spiral members, i.e on the weld seams (closer to the stator) large-

sized scaly erosion was revealed. Maximum sizes of scales reached up

to 150

x 15 mm with 10 mm in depth.

The guide case is designed to include 20 case steel blades having 560 mm

in height. The blades had the greatest erosion on pressure sides neat the

leading edge. Arc-weld building restored the most damaged places. On the

blades of the runner 4 holes were found in the lower parts of the leading

edges,

the largest of which was up to 300x70 mm. On the inner surface of the

impeller the erosion revealed was not so large and was located on its lower

by-pass. During the survey carried out 3 months later it was found that eleven

blades out of thirteen had through holes nearby the lower rim, the largest

being up to 290x90 mm.

The upper parts in the blades were covered on both sides with metallic

luster, in the middle parts the erosion was scaly and in the lower areas there

were all signs of deep erosion.

The inner face in the runner lower rim was worn out extensively as well.

In the lower portion of the runner upper rim surface, being by-passed by the

flow, there took place large-sized scaly erosion occurred. During the further

visual inspection of a new runner, which had operated in the turbine for about

Fundamentals ofHydroabrasive Erosion Theory 23

12 months, it was discovered that on three blades out of thirteen, through

holes which were placed nearly the runner lower rim; the largest hole was up

to 100 x 70 mm.

The upper portions in these blades had metallic luster, whereas the face

sides in lower parts of all the blades had scaly erosion, which was located

closer to the rim.

A pronounced erosion-out was noticed as well on the inner surface of the

runner lower rim which had scaly erosion turning, in some places, into a deep

one.

In the time interval between two overhauls the upper and lower seal rings

in the runner were subjected to the greatest erosion-out.

The inner surface in the spiral case of

the

other hydroelectric unit had fine

scaly erosion. The stator columns in their upper and lower parts had worn-out

journals, whereas the middle portions in these columns were smooth and

shiny. Ring-shaped erosion grooves of different depths were revealed at the

points where the column border on the bosses accommodated in the spiral

case.

The erosion pattern in the blades of the guide case was scaly and

characterized by different intensity levels. The palaces nearby trailing edges

blades,

ad well as those located at the journal parts of lower necks, had the

greatest erosion.

The inner face in the lower rim of the turbine impeller was subjected to an

enormously pronounced erosion-out. In the external surface of the lower rim,

in addition to the mechanical erosion traces, there were remarkable signs of

cavitational erosion.

The streamlined surface in the runner upper rim has fine-scaly erosion in

its top potion and large-sized scaly erosion in the lower part of the rim. All the

blades in the runner and the inner face in the lower rim have in-depth erosion.

The trailing edges in the blades have been worn out to take a sharpened

"knife"- like shape incorporation jags.

The results obtained on the erosion of other units in the above-mentioned

water-power stations are of the same nature.

Another example deals with description of the erosion taking place in

turbines installed in one of the hydroelectric stations in the Tadjik SSR

(USSR), which was based on the results of surveys carried out during the

full-scale tests in the first stage (the spiral case, the guide case) and in the

second stage (the turbine runner). The average annual sediment matter