Sie sind auf Seite 1von 22

Nama : Veni Tri Agustin

NPM : 1655041002
Matakuliah : Teknik Scale-Up

1. It is desired to make a powder addition to a batch of molten aluminum. The batch is


being mixed with an impeller. This was done on the smaller scale and the rotational
speed for submergence of the powder into the molten aluminum was determined.
Solution :
1) The scaleup objective is to incorporate powder into a batch of molten aluminum.
2) What is doing the work in the process?
The batch is being stirred with an impeller.
What is the controlling mechanism?
Solid submergence is sought. The flow at the surface of the batch pulls the
material below the surface.
3) What are the process physics and chemistry?
Froude number is the ratio of inertia forces to gravitational forces. For drawing
down the powder into the batch, the surface velocity must overcome gravity
Buoyancy or gravity is causing the material to float. At the small scale, material
has been drawn into the batch. Hence, for equal solid submergence, equal
modified Froude numbers are sought on scaleup to submerge the powder. The
chemistry is not an issue. What are the process flow regimes? The flow regime is
turbulent.
4) Geometric similarity is assumed for the scaleup.
5) From the various data and correlations, the scaleup method, i.e. equal Froude
numbers or Dg/V2, is obtained.
Fr1 = Fr2
6) At this point, a judgment is made as to whether the scaleup is appropriate.
Are the results reasonable? Yes
Can the scaleup be accomplished? Yes
Will it be a working design? Yes
7) Unfortunately, the scaled process did not make good quality casted aluminum
parts. The scaleup was done correctly but did not provide the expected results.
Plan Bs are needed.
Plan B: What was the real problem? What was the real process objective? As it turns
out, the real objective was to disperse the powder into the batch of molten aluminum.
Submerging the powder did not disperse the powder. To disperse the powder, equal
power per volume on scaleup is the more likely scaleup objective. Power/volume is
another example of the worker/volume ratio. Power is the worker.
2. A metallic electrical wire of diameter d = 5 mm is to be coated with insulation of
thermal conductivity k = 0.35 W/mK. It is expected that, for the typical installation,
the coated wire will be exposed to conditions for which the total coefficient
associated with convection and radiation is h = 15 W/m2K. To minimize the
temperature rise of the wire due to ohmic heating, the insulation thickness is
specified so that the critical insulation radius is achieved. During the wire coating
process, however, the insulation thickness sometimes varies around the periphery of
the wire, resulting in eccentricity of the wire relative to the coating. Determine the
change in the thermal resistance of the insulation due to an eccentricity that is 50% of
the critical insulation thickness.
Solution :
Knows : Wire diameter, convective conditions, and insulation thermal
conductivity.
Find : Thermal resistance of the wire coating associated with peripheral
variations in the coating thickness.

Assumptions:
1. Steady-state conditions.
2. Two-dimensional conduction.
3. Constant properties.
4. Both the exterior and interior surfaces of the coating are at uniform temperatures.
Analysys :
the critical insulation radius is,
𝑊
𝑘 0,35 .𝐾
𝑚
rcr = ℎ = = 0,023 m = 23 mm
15 𝑊/𝑚2 𝐾

Therefore, the critical insulation thickness is


0,005 𝑚
tcr = rcr – d/2 = 0,023m – = 0,021 m = 21mm
2

The thermal resistance of the coating associated with the concentric wire may be
evaluated using Equation below :
𝑟𝑐𝑟 0,023𝑚
ln{ 𝑑 } ln{ 𝑚}
0,005
R’t,cond = 2 2
= 2𝜋(0,35 𝑊/𝑚𝐾0 = 1,0 m.K/W
2𝜋𝑘

For the eccentric wire, the thermal resistance of the insulation may be evaluated
using case 7 of Table 4.1, where the eccentricity is z=0.5 x tcr = 0.5 x 0.021 m =
0.010 m.
𝐷2 + 𝑑2 −4𝑧2
1 cosh−1 ( )
R’t,cond(2D) = 𝑆𝑘 = 2𝐷𝑑
2𝜋𝑘
(2 .0,023 𝑚)2 + (0,005 𝑚)2 −4(0,010 𝑚)2
cosh−1 ( )
2(2.0,023 𝑚).0,005𝑚
= 2𝜋𝑘

= 0,91 m.K/W

Therefore, the reduction in the thermal resistance of the insulation is 0.10 m. K/W,
or 10%.

3. There has been considerable burning of material on heat transfer surfaces in the
small-scale batch stirred pot, 1 liter in volume, which has been used for the
laboratory studies. Studies have been done where using higher levels of agitation in
the pot prevented the burning. What can be done to prevent the burning of the
material as we move into the pilot plant facility, 100 liter? The scaleup factor is 100
in volume.
Solution :
- State the scaleup objectives. What are the important aspects of the process?
What is the process similarity you seek? What are you trying to do?
The process has had significant problems because of the burning issue. Process
similarity is that since the material did not burn in the lab scale in some cases,
the material should not burn on the large scale if the scaleup of these conditions
is done properly. On scaleup, the prevention of material from burning is the
scaleup objective.
- What is doing the work in the process? What is the controlling mechanism?
What is the size of the job?
Agitation has prevented burning by increasing the heat transfer. The actual work
is being done by the heat transfer coefficient. Having a high enough heat transfer
coefficient at the surface to take away the heat prevents burning. The controlling
mechanism is the heat transfer. The size of the job, 100 liter, determines the
scaleup factor, a factor of 100 in volume and 4.64 in length.
V2 = 100 V1
and
D2 = 4.64 D1
- What are the process physics and chemistry? What are the process flow
regimes?
The process physics is that we want the same heat transfer coefficient on the
large scale that we had on the small scale. The burning on the large scale will
not likely occur. The process chemistry is the burning product but chemistry is
not an important issue in this case. The process flow regime is turbulent.
- At this point, geometric similarity is assumed for scaleup.
Geometric similarity serves as a starting point to fix the geometry.
- From the various data and correlations, the scaleup method is obtained.
One correlation for turbulent heat transfer in agitated tanks is the Nusselt
Reynolds-Prandtl correlation with a slight correction for wall viscosities.
Nu = 0.332 Re0.66 Pr0.33 (μ/μw)0.14
If all the fluid properties, the leading constant and constant geometric ratios are
removed from the equation, one obtains:
hD α (ND2)0.66
or
h α N0.66D+0.33
The scaleup objective is to have:
h1 = h2
with D2 =4.64 D1. This becomes
(N0.66D+0.33)1 = (N0.66D+0.33)2
Or
(N0.66D+0.33)1 = (N0.66) ((4,64D)+0.33)1
(N0.66)1 = (N0.66)2 (4,64)+0.33
The relationship between the agitator rotational speeds is now obtained. Since
N1 is known from lab studies and the two diameters are known as well, N2 can
be calculated.
N2 = 0.464 N1
From the rotational speed, N2, and impeller diameter, D2, the motor size of the
agitator for the pilot plant facility can be determined. Material properties remain
the same on scaleup and cancelled in the heat transfer correlation. Geometric
ratios can be removed from the correlation, since geometric similarity is
assumed. The ratios are the same between the small and large scale and they
cancel.
4. A dusting problem exists. One way to handle this problem is to use
agglomeration. A new product or by-product may be obtained. Testing has been
done on the small scale using a pan agglomerator. A pan agglomerator is a flat
pan that rotates at an angle. Powder is added above the pan and rolls as the pan
rotates. Powder forms balls as it rolls. These grow to large uniform size spheres
that flow over the side of the pan.
Solution:
1) State the scaleup objectives.
The objective of the process is to make the same type of agglomerate that
was made on the small scale.
What are the important aspects of the process?
The roll mechanics of the agglomerate in the pan become important. As the
agglomerate seeds roll over fine powder, the seed grows into larger
agglomerates. The process is like making a snowman in the winter.
However, this occurs only in a rolling flow regime. Powder can easily slide,
be lifted or centrifuged to the wall. These regimes have to be avoided. The
powder is assumed to be free flowing and there are no effects of solid
loading.
Another equipment choice is a drum agglomerator, which typically recycles
4 times the quantity of fresh feed. The recycle level is unacceptable for this
problem. In drum agglomerators, there may also be asubstantial effect of
solids loading. The lifting/carryover regime occurs in drums and is
undesirable for making agglomerates. Lifting and dropping agglomerates
will cause them to break. This regime is minimized in a pan agglomerator.
The pan agglomerator may be the best equipment for the task. Drums are
less desirable.
The transitions between 1) the sliding and rolling regime and 2) rolling and
the centrifugation regimes may be failure boundaries for the process in the
pan agglomerator. It may be desirable to operate in the rolling regime as
close as possible to the centrifugation regime to maximize production. This
operation will maximize roll area.
There is an angle effect in pan agglomeration, which should be studied as
part of the piloting studies.
Process similarity exists when the seed agglomerate has the same roll
distance and is exposed to the same powder content in the two sizes of
equipment. Roll distance, S, should be a function of rotational velocity, ND,
and residence time, θR. θR is holdup volume divided by feed rate.
S α θR ND
On scaleup, the seeds will grow to the same size when
S1 = S2
This has to occur above sliding regime and below centrifugation.
(θR ND)1 = (θR ND)2
At this point, several options exist. With residence time kept constant, (θR)1
= (θR)2 then
(ND)1 = (ND)2
Since the information, N and D, on the small scale and the size of the large
scale are known, then the rotational speed of the larger pan, N2, can be
calculated and the pan agglomerator can be specified.
5. For an atomization process, it is desired to increase the production of a spray
from 2 lb/min to 10 lb/min using a pressure swirl nozzle. At 2 lb/min the
atomization pressure drop is 30 psi. Keeping the drop size the same would keep
the two processes relatively similar to each other.
Solution :
1) State the scaleup objective:
Keep the drop size the same on scaleup.
2) What is doing the work in the process?
Nozzles are going to be used similar to those used on the small scale. Let’s
see if this is possible.
3) What is the controlling mechanism?
The controlling mechanism for nozzle flow is the flow number. The flow
number consists of the mass flow rate, m, divided by the square root of
nozzle pressure drop and liquid density or:
m/(Δpρ)0.5
The flow number can be interpreted as the product of a discharge coefficient
and a flow area. This product of discharge coefficient and flow area is very
important because this determines the drop size. If the flow number is kept
constant on scaleup, the drop size will likely remain the same. So between
scale 1 and scale 2, equal flow numbers are desired which result in the
following relationship:
m/(Δpρ)0.51 = m/(Δpρ)0.52
4) What is the size of the job? 10 lb/min
What are the process physics and chemistry?
It is thought that the process physics is well understood. Chemistry only
appears through affecting the surface tension in this case. When reaching
their smallest sizes, drop size depends upon surface tension. However, the
material and the chemistry are assumed to be the same between the two
scales.
5) Geometric similarity is assumed.
With the mass flow rates and pressure drop given, the above equation
becomes:
2/(30)0.51 = 10/(Δp)0.52
The pressure drop for the large scale becomes 750 psi. Unfortunately, this
pressure drop is not available at the facility.
6) The scaleup is not appropriate for the conditions that exist in the facility.
Are the results reasonable? No
Can the scaleup be accomplished? No
Will it be a working design? No
7) Plan B.
The geometry must be changed if the scaleup is to be accomplished. The
maximum available pressure in the facility is 90 psig. Using this pressure,
the mass flow for scale 2 is 3.46 lb/min as determined from:
2/(30)0.51 = m/(90)0.52
Whereas 10 lb/min is desired and 3.46 lb/min can be delivered with one
nozzle, three nozzles are needed.
6. It is desired to cook at the same rate large scale as small scale. Comparisons will
be made between two processes differing by a factor of 1000, i.e., a scaleup of
1000 in volume.
Solution :
1) State the scaleup objectives. What are the important aspects of the process?
What is the process similarity you seek? What are you trying to do?
The desire is to have the same cooking rate large scale as small scale. This
can be written as an equation:
(Q/V)1 = (Q/V)2
Volume varies with the cube of the impeller diameter for geometrically
similar systems.
V α D3
2) What is doing the work in the process? What is the controlling mechanism?
What is the size of the job?
The heat transfer coefficient, area for heat transfer and the temperature
difference are doing the actual work.
Q = h A ΔT
The controlling mechanism is heat transfer. The large-scale process will be
heated the same way as the small scale, i.e., through a jacket. The
temperature difference, ΔT, will be the same between the two scales.
The size of the job, 1000 times scale one, determines the scaleup factor.
V2= 1000 V1 and
D2= 10 D1
The area, A, for heat transfer varies with the square of the impeller diameter,
D. This is the case when heating through a jacket for geometrically similar
systems.
A α D2
So on scaleup, the following is desired.
(Q/V)1 = (Q/V)2
Or
(h A ΔT/V)1 = (h A ΔT/V)2
Since the ΔTs are the same, A is D2 and V is D3, we have:
(h/D)1 = (h/D)2
3) What are the process physics and chemistry? What are the process flow
regimes?
The process physics is heat transfer. The process chemistry is not important
in this case. The process flow regime is probably turbulent but a Reynolds
number calculation is needed to check this.
4) At this point, geometric similarity is assumed for scaleup.
Geometric similarity serves as a starting point to fix the geometry. This has
been already assumed in the above discussion.
5) From the various data and correlations, the scaleup method is obtained.
One correlation for turbulent heat transfer in agitated tanks is the Nusselt-
Reynolds-Prandtl correlation with a small correction for wall viscosities.
Nu = 0.332 Re0.66 Pr0.33 (μ/μw)0.14
If all the fluid properties, the leading constant and constant geometric ratios
are removed from the equation, one obtains:
hD α (ND2)0.66
or
h α N0.66D+0.33
From (h/D)1 = (h/D)2 above, the following is obtained:
(N0.66D-0.66)1 = (N0.66D-0.66)2
Now the relationship between the agitator rotational speeds can be obtained.
Since N1 is known from lab studies and the two diameters are known, i.e.,
D2 = 10 D1, N2 can be calculated.
N2 = 10 N1
From the rotational speed and impeller diameter, the motor size of the
agitator for the pilot plant facility can be determined.
7. Materials often have to be well mixed in processing. If they are not well mixed,
poor quality products usually result. Mixing is not instantaneous in most cases.
For effective batch mixing geometries, the number of revolutions to mix (i.e.
NRM) is a fairly small value, e.g., 30, 100, 300. Excessive values, e.g., above
10,000, should be considered unacceptable. The number of revolutions to mix is
simportant for continuous flow systems as well. As materials move through a
unit, they have to experience an appropriate number of revolutions to become
mixed. If the materials do not, then they are going to leave the continuous flow
system partially mixed. This can be viewed as a moving NRM. The process
objective is to have the material remain in the mixer long enough to become
mixed.
Solution :
Two ways to improve mixing are: 1) have an effective mixing geometry and 2)
eliminate unneeded mixing or procedural mixing. Effective mixing geometry is
one with a low number of revolutions to mix. Procedural mixing occurs in
procedures where statements are made which are not supported by need. “Add
and stir for thirty minutes” is an example. The mixing may actually be over in
one minute. The other twenty-nine minutes is processing time being wasted for
no reason.
The number of revolutions to mix is a constant determined by geometry. For
geometrically similar systems, this becomes:
Nθ1 = Nθ2
For equal mixing times,
θ1 = θ2
therefore
N1 = N2
The rotational speed for scale 1 equals the rotational speed for scale 2. The
difficulty comes from the power calculations. For turbulent flow, power is a
function of rotation speed and impeller diameter as:
P α N3 D5
For a scaleup of 1000x in volume or 10x in diameter, D, in turbulent flow, the
power used in scale 1 to power used in scale 2 are related as:
P2 = 105 P1
This is an unrealistic result. As a result, equal mixing times on scaleup are
unrealistic as well. Larger scale tanks will take longer to mix than smaller tanks.
However, one possible way to help in the processing is to determine the mixing
time that might be needed on the large scale. From that, the calculation of power
input can be done. For example, the larger scale will mix 10 times more slowly
than the small scale or:
θ2 = 10 θ1
This means that, in terms of rotational speed,
N2 = 0.1 N1
In terms of power then:
P2 = 102 P1
This power level for scale 2 may be possible. The Reynolds numbers between
the two scales are related as
Re2 = 10 Re1
Although this may appear to be satisfactory, it is well known that larger tanks
are far less turbulent than smaller tanks for the same Reynolds number.
8. The pressure drop due to friction for flow in a long smooth pipe is a function of
average flow velocity, density, viscosity, and pipe length and diameter: ∆p =
fcn(V, ρ, μ, L, D). We wish to know how p varies with V. Use the pi theorem to
rewrite this function in dimensionless form.
Solution :
There are six variables with three primary dimensions involved {MLT}. The
refore we expect that j = 6 – 3 = 3 pi groups. We are correct, for we can find
three variables which do not form a pi product, for example, (ρ, V, L). Carefully
select three (j) repeating variables, but not including ∆p or V, which we plan to
plot versus each other. We select (ρ, μ, D), and the pi theorem guarantees that
three independent power-product groups will occur:

We have omitted the algebra of finding (a, b, c, d, e, f, g, h, i) by setting all


exponents to zero M0, L0, T0. Therefore we wish to plot the dimensionless
relation

9. Then plot this function, using the following data for three pipes and three fluids:
Solution :
We plot III1 versus III2 with III3 as a parameter. There will be nine data points.
For example, the first row in the data above yields

The values of L/D are listed for each point, and we see a significant length
effect. In fact, if we connect the only two points which have the same L/D
(=200), we could see (and cross-plot to verify) that p increases linearly with L,
as stated in the last part of the problem. Since L occurs only in II3=L/D, the
function II1=fcn(II2, II3) must reduce to II1= (L/D) fcn(II2), or simply a function
involving only two parameters:
flow in a long pipe
10. Suppose it is further known that _p is proportional to L (which is quite true for
long pipes with well-rounded entrances). Use this information to simplify and
improve the pi-theorem formulation. Plot the dimensionless data in this
improved manner and comment upon the results.
Solution :
We now modify each data point in Fig. above by dividing it by its L/D value.
For example, for the first row of data, ρD3∆p/(Lμ2) = (3.73 E9)/500=7.46 E6.
We replot these new data points as solid circles in Fig. 5.10. They correlate
almost perfectly into a straight-line power-law function:

11. The smooth-sphere represent dimensionless drag versus dimensionless viscosity,


since (ρ, V, d) were selected as scaling or repeating variables. Replot these data
to display the effect of dimensionless velocity on the drag.
Solution :
To display the effect of velocity, we must not use V as a repeating variable.
Instead we choose (ρ, μ, d) as our j variables to nondimensionalize, F=fcn(d, V,
ρ, μ). The pi groups form as follows:
12. Use your new figure to predict the terminal (zero-acceleration) velocity of a 1-
cm-diameter steel ball (SG=7.86) falling through water at 20°C.
Solution :
That is, a=1, b=2, c=0, e=1, f=1, and g=1, by using our power-product
techniques. The refore a plot of ρF/μ2 versus Re will display the direct effect of
velocity on sphere drag. This replot is shown as Fig below. The drag increases
rapidly with velocity up to transition, where there is a slight drop, after which it
increases more quickly than ever. If the force is known, we may predict the
velocity from the figure. For water at 20°C, take ρ = 998 kg/m3 and 𝜇 = 0.001
kg/(m.s). For steel, ρs = 7.86ρwater = 7840 kg/m3. For terminal velocity, the
drag equals the net weight of the sphere in water. Thus

From Fig. below, at ρF/μ2=3.5E7, a magnifying glass reveals that Red 2 E4.
Then a crude estimate of the terminal fall velocity is
13. To illustrate these various steps we consider the problem which concerns with
the steady flow of an incompressible Newtonian fluid through a long, smooth-
walled, horizontal circular pipe. We are interested in the pressure drop per unit
length, Δpℓ along the pipe.
Solution :
Step 1 : list all of the pertinent variables that are involved based on the
experimenter’s knowledge of the problem. In this problem we assume

where D is the pipe diameter, ρ andμ are the fluid density and viscosity,
respectively, and V is the mean velocity.
Step 2 : express all the variables in terms of basic dimensions. Using F,L, and T
as basic dimensions it follows that

Step3: There are five variables(k=5) and three required referenc dimensions(r=3)
then according to the pi theorem there will be(n=k-r=5-3=2) or two pi terms
required.
step4: we would try to select from repeating variables those that are the simplest,
dimensionally. we will use D,V, and ρ as repeating variables.
step5: We are now ready to form the two pi terms

Since this combination is to be dimensionless, it follows that

The exponents, a, b, and c must be determined

It follows that a=1, b=-2, c=-1 and, therefore,


step6: The process is now repeated for the remaining nonrepeating variables. In
this example there is only one additional variable (μ) so that

step7: At this point make sure the pi terms are actually dimensionless. We will
check using both FLT and MLT dimensions

step8: Finally, we can express the result of the dimensional analysis as

In the fig. below , an illustrative plot of pressure drop data using dimensionless
parameters .
14. A long structural component of a bridge has the cross section shown in Fig. It is
known that when a steady wind blows past this type of bluff body, vortices may
develop on the downwind side that are shed in a regular fashion at some definite
frequency. Since these vortices can create harmful periodic forces acting on the
structure, it is important to determine the shedding frequency. For the specific
structure of interest, D=0.1 m , H=0.3 m and a representative wind velocity is 50
km/hr. Standard air can be assumed. The shedding frequency is to be determined
through the use of a small-scale model that is to be tested in a water tunnel. For
the model Dm=20 mm and the water temperature is 20℃ . Determine the model
dimension, Hm and the velocity at which the test should be performed. If the
shedding frequency for the model is found to be 49.9 Hz, what is the
corresponding frequency for the prototype?

Solution :

Since there are six variables and three reference dimensions (MLT), three pi
terms are required. Application of the pi theorem yields

to maintain similarity between model and prototype


From the first similarity requirement

For air standard conditions, μ = 1,79 x 10-5 kg/m.s, ρ = 1,23 kg/m3, and for water
at 20oC, μ = 1,00 x 10-3 kg/m.s, ρ = 998 kg/m3. The fluid velocity for the
prototype is

and the predicted prototype vortex shedding frequency is


15. Consider convective cooling of a two-dimensional streamlined strut of
characteristic length𝐿𝐻2 = 2 atm. The strut is exposed to hydrogen -owing at𝑝𝐻2 =
2 atm,𝑉𝐻2 =8.1 m/s, and 𝑇∞,𝐻2 = -30℃ Of interest is the value of the average heat
transfer coeficientℎ̅𝐻2 , when the surface temperature is𝑇𝑠,𝐻2 = -15℃. Rather than
conducting expensive experiments involving pressurized hydrogen, an engineer
proposes to take advantage of similarity by performing wind tunnel experiments
using air at atmospheric pressure with 𝑇∞,𝐴𝑖𝑟 = 23℃. A geometrically similar
strut of characteristic length𝐿𝐴𝑖𝑟 = 60mm and parameter P = 150 mm is placed in
the wind tunnel. Measurements reveal a surface temperature of𝑇𝑠 ,𝐴𝑖𝑟 = 30℃.
when the heat loss per unit object length (into the page) is 𝑞′𝐴𝑖𝑟 =
𝑊
50 𝑚.Determine the required air velocity in the wind tunnel experiment 𝑉𝐴𝑖𝑟 and

the average convective heat transfer coeficient in the hydrogen ℎ̅𝐻2 .


Known: Flow across a strut. Hydrogen pressure, velocity, and temperature. Air
temperature and pressure, as well as heat loss per unit length. Surface
temperatures of the strut in hydrogen and in air.\
Find: Air velocity and average convective heat transfer coeficient for the
strut that is exposed to hydrogen.
Schematic:

Assumptions:
1. Steady-state, incompressible boundary layer behavior.
2. Ideal gas behavior.
3. Constant properties.
4. Negligible viscous dissipation.
23℃+30℃
Properties: Table A.4, air ( p=1atm, 𝑇𝑓 = ( = 26.5℃ ≈ 300 𝐾) ;
2
𝑊
Pr=0.707, v=15.89× 10−6 𝑚2 /𝑠, k=26.3× 10−3 𝑚 . 𝐾.Table A.4 Hydrogen (

p=1atm, 𝑇𝑓 = (−22.5℃ ≈ 250 𝐾); Pr=0.707, v=81.4× 10−6 𝑚2 /𝑠, k=157×


𝑊
10−3 𝑚 . 𝐾.

The properties k, Pr, 𝑐𝑝 and 𝜇may be assumed to be independent of


pressure to an excellentapproximation. However, for a gas, the kinematic
𝜇
viscosity 𝑝will vary with pressurethrough its dependence on density.

From the ideal gas law, p/RT, it follows that the ratioof kinematic
viscosities for a gas at the same temperature but at different pressures,
𝑝1 and 𝑝2 , is (𝑣1 , 𝑣2 )=(𝑝2 /𝑝1). Hence, the kinematic viscosity of hydrogen
1𝑎𝑡𝑚 10−6 𝑚2
at 250 K and 2 atm is𝑣𝐻2 =81.4× 10−6 𝑚2 /𝑠 × 2𝑎𝑡𝑚 = 40.7 × .Since
𝑠

Pr is independent of pressure, 𝑃𝑟𝐻2 (𝑝 = 2 𝑎𝑡𝑚, 𝑇𝑓 = −22.5℃) =


𝑃𝑟𝑎𝑖𝑟 (𝑝 = 1 𝑎𝑡𝑚, 𝑇𝑓 = 26.5℃) = 0.707.

Analysis: From Equation 6.50, we know that the average Nusselt


numbers are related to
the Reynolds and Prandtl numbers by the functional dependence
ℎ̅𝐻2 𝐿𝐻2 ℎ̅𝐴𝑖𝑟 𝐿𝐴𝑖𝑟
̅̅̅̅̅̅̅
𝑁𝑢𝐻2 = ̅̅̅̅̅̅̅
= 𝑓(𝑅𝑒𝐿𝐻2 , 𝑃𝑟𝐻2 ) 𝑎𝑛𝑑 𝑁𝑢 𝐴𝑖𝑟 = = 𝑓(𝑅𝑒𝐿𝐴𝑖𝑟 , 𝑃𝑟𝐴𝑖𝑟 )
𝑘𝐻2 𝑘𝐴𝑖𝑟

Since 𝑃𝑟𝐻2 = 𝑃𝑟𝐴𝑖𝑟 , similarity exists if𝑅𝑒𝐿𝐴𝑖𝑟 = 𝑅𝑒𝐿𝐻2 , in which case the
average Nusselt
numbers for the air and hydrogen will be identical,𝑁𝑢𝐴𝑖𝑟 = 𝑁𝑢𝐻2 .
Equating the Reynolds
numbers for the hydrogen and air yields the expression
𝑅𝑒𝐿𝐴𝑖𝑟 𝑣𝐴𝑖𝑟 𝑅𝑒𝐿𝐻2 𝑣𝐴𝑖𝑟 𝑉𝐻2 𝐿𝐻2 𝑣𝐴𝑖𝑟
𝑉𝐴𝑖𝑟 = = =
𝐿𝐴𝑖𝑟 𝐿𝐴𝑖𝑟 𝑣𝐻2 𝐿𝐴𝑖𝑟
8.1𝑚/𝑠 × 0.04𝑚 × 15.89 × 10−6 𝑚2 /𝑠
= = 2.10 𝑚/𝑠
40.7 × 10−6 𝑚2 /𝑠 × 0.06𝑚
With 𝑅𝑒𝐿𝐴𝑖𝑟 = 𝑅𝑒𝐿𝐻2 and 𝑃𝑟𝐻2 = 𝑃𝑟𝐴𝑖𝑟 , we may equate the Nusselt
numbers for the hydrogen and air, and incorporate Newton's law of
cooling. Doing so gives
𝐿𝐴𝑖𝑟 𝑘𝐻2 𝑞 ′ 𝐴𝑖𝑟 𝐿𝐴𝑖𝑟 𝑘𝐻2
ℎ̅𝐻2 = ℎ̅𝐴𝑖𝑟 = =
𝐿𝐴𝑖𝑟 𝑘𝐴𝑖𝑟 𝑃(𝑇𝑠,𝐴𝑖𝑟 𝑇𝑠,𝐻2 ) 𝐿𝐻2 𝑘𝐴𝑖𝑟
0.157𝑊
50𝑊/𝑚 0.06𝑚 × 𝑚 . 𝐾 𝑊
= × = 426 .𝐾
150 × 10−3 𝑚 × (30 − 23)℃ 0.04𝑚 × 0.0263 𝑊 . 𝐾 𝑚
𝑚
16. The drag of a sonar transducer is to be predicted, based on wind tunnel test data.
The prototype, a 1-ft diameter sphere, is to be towed at 5 knots (nautical miles
per hour) in seawater at 40F. The model is 6 in. in diameter. Determine the
required test speed in air. If the drag of the model at these test conditions is 0.60
lbf, estimate the drag of the prototype.
Solution :
Given: Sonar transducer to be tested in a wind tunnel.
Find :
a. Vm?
b. Fp?

Since the prototype operates in water and the model test is to be performed in air,
useful results can be expected only if cavitation effects are absent in the prototype
flow and compressibility effects are absent from the model test. Under these
conditions,

and the test should be run at


to ensure dynamic similarity. For seawater at 400F, ρ 5 1.99 slug/ft3 and v ≈ 1,69 x
10-5 ft2/s. At prototype conditions,

Thus,

For air at STP, ρ = 0,00238 slug/ft3 and ν = 1,57 x 10-4 ft2/s, the twind tunnel must
be operated

This speed is low enough to neglect compressibility effects.


At these test conditions, the model and prototype flows are dynamically similar.
Hence

If cavitation were expected—if the sonar probe were operated at high speed near
the free surface of the seawater—then useful results could not be obtained from a
model test in air.

Das könnte Ihnen auch gefallen