PC-FT 403: Lecture (5) by
PC-FT 403: Lecture (5) by
PC-FT 403: Lecture (5) by
𝑚 100−𝑚
𝐶= 𝐶𝑤 + 𝐶𝑑 kcal/kg °C (5.1)
100 100
❖ The above relationship exists above 8% moisture content of the grain only.
❖ The specific heat of bone dry grain varies from 0.35 – 0.45 kcal/kg. °C.
❖ The specific heat is measured by calorimeter, generally a simple thermos vacuum bottle.
❖ The differential scanning calorimeter (DSC) is suitable for measuring the specific heat.
Thermal Conductivity
❖ The thermal conductivity may be defined as the rate of heat flow through unit thickness of material per unit
area normal to the direction of heat flow and per unit time for unit temperature difference.
❖ It is a measure of ability of the material to conduct heat.
❖ The thermal conductivity can be expressed by the following equation.
𝑄 = 𝐾𝐴∆𝑇 (5.2)
where Q = amount of heat flow (kcal), A = area (m2), ∆T = temperature difference in the direction of heat
flow (°C), K = thermal conductivity (kcal/m.h.°C or W/m. °C).
❖ If m is the moisture content of the grain, then the thermal conductivity of wheat can be expressed as follows:
𝐾 = 0.060 + 0.002𝑚, kcal/m.h.°C (or W/m. °C) (5.3)
The above relationship exists for the wheat bulk with the moisture content range of 10 – 20% (db).
❖ The thermal conductivity of sorghum maybe given as follows:
❖ The thermal conductivity of single grain ranges from 0.3 – 0.6 kcal/m. h.°C.
❖ The thermal conductivity of bulk grain ranges from 0.10 – 0.15 kcal/m. h.°C. The difference is due to the air
spaces present in the bulk grain.
❖ The thermal conductivity of air is 0.02 kcal/m. h .°C.
Determination of thermal conductivity of grains and food materials
❖ The measurement of thermal conductivity of agricultural grains pose much problem because it depends on the structure and
chemical composition of the grain.
❖ Thermal conductivity probe is used for measurement of thermal conductivity of most of the food materials.
❖ A typical line heat source thermal conductivity probe is shown in Figure 5.1.
❖ The probe consists of a hypodermic needle of 0.66 mm diameter and connected to a handle.
❖ A 0.077 mm diameter constantan heater wire insulated with plastic
spaghetti tubing is inside the needle.
❖ The wire is taken from to the top and back.
❖ The purpose of using constantan wire is to avoid the change in
electrical resistance with temperature.
❖ The insulated chromel constantan thermocouples wire of 0.051 mm
Figure 5.1: Diagram of a line heat source for
diameter is wrapped around the upper half of the probe handle and measurement of thermal conductivity. 1. heating wires and
the needle tip. 2. junction of thermocouple wires.
❖ The line-heat source probe is inserted into the sample having uniform initial temperature.
❖ The probe is heated at a constant rate, and the temperature adjacent to the line-heat source is monitored.
❖ After a brief transition period, the plot of natural logarithm of time vs. temperature is plotted, which has a
slope equal to Q/4πK.
❖ The thermal conductivity of material can be estimated by the following equation:
𝑡2 − 𝑡0
ln
𝑡1 − 𝑡0 (5.5)
𝐾=𝑄
4𝜋 𝑇2 − 𝑇1
where Q = power generated by the probe heater (W/m), K = thermal conductivity of the sample (W/m.°C),
t1 and t2 = time since the probe is energized (s), t0 = time conductance factor, T1 and T2 = temperature of
probe thermocouple at time t1 and t2 respectively, °C.
Enthalpy
❖ Enthalpy is the total heat content or energy level of a material.
❖ The enthalpy data are required for frozen foods that freeze over a range of temperatures below 0 °C and not for
those substances that freeze in a narrow temperature limit, as the case of pure substance like water.
❖ The enthalpy of the moist material can be estimated by using following expression:
where h2 – h1 = enthalpy difference (J), m = mass of the product (kg), cp = specific heat of the product (J/kg.
K), Xw = water fraction, T2 – T1 = temperature difference (K), L = latent heat of fusion for water (J/kg).
𝐽 𝐽
𝑘𝑔 × × 𝐾 + 𝑘𝑔 × =𝐽
𝑘𝑔 × 𝐾 𝑘𝑔
Thermal Diffusivity
❖ The thermal diffusivity may be calculated by dividing the thermal conductivity with the product of specific heat and mass
density. 𝑊 𝑚3 𝑘𝑔 × 𝐾 𝐽 𝑚3 𝑘𝑔 × 𝐾 𝑚2
× × = × × =
𝑚 × 𝐾 𝑘𝑔 𝐽 𝑚 × 𝑠 × 𝐾 𝑘𝑔 𝐽 𝑠
❖ It may be expressed as:
𝐾
𝛼= (5.7)
𝜌𝑐𝑝
where α = thermal diffusivity (m2/s), K = thermal conductivity (W/m. K), cp = specific heat of the product (J/kg. K), 𝜌 =
mass density (kg/m3).
❖ Thermal diffusivity is important in determination of heat transfer rates in solid food materials of any shape. Physically it shows
the relationship between the ability of a material to conduct heat to its ability to store heat.
❖ The common method of determination of thermal diffusivity is to calculate from experimentally measured values of thermal
conductivity, specific heat and mass density.
❖ Heat transfer is one of the important unit operations in the agricultural processing activities. e.g. thermal
treatment of food grains, cooling of perishables, pasteurization and sterilization of milk and fruit juices, freezing,
etc.
❖ It is also an essential operation to provide energy for evaporation of moisture during drying, solar heating or
cooling, distillation process, etc.
❖ The heat transfer occurs because of a temperature-difference driving force. Heat flows from the high to the low
temperature region.
❖ Specifically for heat transfer, we get a balanced equation:
Rate of heat in + rate of heat generation = rate of heat out + rate of accumulation of heat (5.8)
❖ Assuming the rate of transfer of heat occurs only by conduction, we can write,
𝑞𝑥 𝑑𝑇
= −𝑘 (5.9)
𝐴 𝑑𝑥
❖ Making an unsteady state heat balance for the 𝑥 direction only on the element of
control volume (Figure 5.2) by using Eqns. (5.8) and (5.9), with the c/s area being A
Figure 5.2: Unsteady-state balance for heat
m 2. transfer in control volume.
𝜕𝑇
𝑞𝑥ȁ𝑥 + 𝑞ሶ ∆𝑥. 𝐴 = 𝑞𝑥ȁ𝑥+∆𝑥 + 𝜌𝑐𝑃 (∆𝑥. 𝐴) (5.10)
𝜕𝑡
where 𝑞ሶ is rate of heat generated per unit volume. Assuming no heat generation and also assuming steady-state heat transfer,
where the rate of accumulation is zero, Eqn. (5.10) becomes
𝑞 𝑥 𝑥 = 𝑞 𝑥 𝑥 + ∆𝑥 (5.11)
This means the rate of heat input by conduction = the rate of heat output by conduction; or 𝒒𝒙 is a constant with time for
steady state heat transfer and the temperatures at various points in the system do not change with time.
Basic Mechanisms of Heat Transfer
Heat transfer may occur by any one or more of the three basic mechanisms of heat transfer:
1. Conduction
o Heat can be conducted through solids, liquids, and gases.
o Heat is conducted by the transfer of the energy between adjacent molecules. This type of transfer is present to some
extent in all solids, gases or liquids in which a temperature gradient exists.
o Energy can also be transferred by ‘free’ electrons (important in metallic solids).
o Examples: Heat transfer through walls of exchangers or refrigerator, drum drying, freezing of the ground during winter
and so on.
2. Convection
o Heat transfer by bulk transport and mixing of macroscopic elements of warmer with cooler portions of a gas or
liquid.
o It also refers to the energy exchange between a solid surface and a fluid.
o A distinction must be made between forced-convection heat transfer, where a fluid is forced to flow past a solid surface by
a pump, fan or other mechanical means, and natural or free convection , where warmer or cooler fluid next to the solid
surface causes a circulation because of a density difference resulting from the temperature differences in the fluid.
o Examples: loss of heat from a car radiator where the air is circulated by a fan, cooking of foods in a vessel being stirred,
cooling of a hot cup of coffee by blowing over the surface, and so on.
3. Radiation
o No physical medium is needed for its propagation.
o Transfer of energy through space occurs by means of electromagnetic waves in almost the same way as electromagnetic
light waves transfer light.
o The same laws that govern the transfer of light, govern the radiant transfer of heat.
o Solids and liquids tend to absorb the radiation being transferred through them, so that radiation is important primarily in
transfer through space or gases.
o Examples: transport of heat to the earth from sun, cooking of food when passed below red-hot electric heaters, heating of
fluids in coils of tubing inside a combustion furnace, and so on.
Fourier’s Law of Heat Conduction
❖ The basic equation is as follows:
𝑑𝑟𝑖𝑣𝑖𝑛𝑔 𝑓𝑜𝑟𝑐𝑒 (5.12)
𝑅𝑎𝑡𝑒 𝑜𝑓𝑎 𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝑝𝑟𝑜𝑐𝑒𝑠𝑠 =
𝑟𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒
The above equation implies that to transfer a property such as heat or mass, we need a driving force to overcome a resistance.
❖ The transfer of heat by conduction through fluids or solids is given by Fourier’s law:
𝑞𝑥 𝑑𝑇
= −𝑘
𝐴 𝑑𝑥
where 𝑞𝑥 = heat transfer rate in the 𝑥 direction (W); A = c/s area normal to the direction of heat flow (m2); T =
temperature (K); 𝑥 = distance (m); k = thermal conductivity (W/m. K).
❖ If the thermal conductivity is not constant but varies linearly with temperature Figure 5.3: Heat conduction in a flat wall: (a)
(mainly applicable for liquids and gases), then it can be expressed as geometry of wall, (b) temperature plot.
𝑘 = 𝑎 + 𝑏𝑇 (5.15)
𝑇1 + 𝑇2
𝑞 𝑎+𝑏 2 𝑘𝑚 (5.16)
= 𝑇1 − 𝑇2 = 𝑇 − 𝑇2
𝐴 ∆𝑥 ∆𝑥 1
𝑇1 +𝑇2
where, 𝑘𝑚 = 𝑎 + 𝑏 . This means that the mean value of k (i.e., km) is the value of k evaluated at the linear average of T1
2
and T2.
❖ As stated earlier, the rate of a transfer process equals the driving force over the resistance. Eqn. (5.14) can be rewritten in that
form:
𝑇1 − 𝑇2 𝑇1 − 𝑇2 𝑑𝑟𝑖𝑣𝑖𝑛𝑔 𝑓𝑜𝑟𝑐𝑒
𝑞= = = (5.17)
∆𝑥Τ𝑘𝐴 𝑅 𝑟𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒
Problem 5.2: The maximum tolerable heat loss through a furnace wall is 1.3 kW/m2. A brick wall is constructed next to the
furnace wall to insulate the heat loss. The temperature on either side of the wall is 200 °C and 40 °C. What should be the
thickness of the brick wall if the thermal conductivity of the brick is 0.72 W/m °C?
Solution: Here q/A = 1.3 kW/m2, T1 = 200 °C, T2 = 40 °C and k = 0.72 W/m. °C
According to Fourier’s Law,
𝑞 𝑘 𝑘𝐴 0.72
= 𝑇1 − 𝑇2 ⟹ ∆𝑥 = 𝑇1 − 𝑇2 = 200 − 40 = 0.088 𝑚
𝐴 ∆𝑥 𝑞 1.3
Conductive Heat Transfer – Through a Composite Wall
❖ In the case where there is a multilayer wall of more than one material is present (Figure 5.4), we proceed as follows.
❖ Since the walls are in series, the heat flow q must be same in each layer. We can write Fourier’s equation for each layer as:
𝑘𝐴 𝐴 𝑘𝐵 𝐴 𝑘𝐶 𝐴
𝑞= 𝑇1 − 𝑇2 = 𝑇2 − 𝑇3 = 𝑇 − 𝑇4 (5.18)
Δ𝑥𝐴 Δ𝑥𝐵 Δ𝑥𝐶 3
𝑇1 −𝑇4 255.4−297.1
𝑞= = = −16.48 𝑊 (Answer). The –ve value indicates that heat flows in from outside.
𝑅𝐴 +𝑅𝐵 +𝑅𝐶 0.0841+2.346+0.1
𝐴 = 2𝜋𝑟𝐿 (5.22)
❖ Substituting Eqn. (5.22) into Eqn. (5.21), rearranging and integrating we get,
𝑟2 𝑇2
𝑞 𝑑𝑟
න = −𝑘 න 𝑑𝑇 (5.23) Figure 5.5: Heat conduction through a hollow
2𝜋𝐿 𝑟
𝑟1 𝑇1 cylinder.
2𝜋𝐿
𝑞=𝑘 𝑇1 − 𝑇2 (5.24)
ln 𝑟2 Τ𝑟1
❖ Multiplying numerator and denominator by (r2 – r1),
𝑟2 − 𝑟1 𝑇1 − 𝑇2 𝑇1 − 𝑇2 𝑇1 − 𝑇2 𝑇1 − 𝑇2
𝑞 = 𝑘2𝜋𝐿 × = 𝑘𝐴𝑙𝑚 = = (5.25)
ln 𝑟2 Τ𝑟1 𝑟2 − 𝑟1 𝑟2 − 𝑟1 𝑟2 − 𝑟1 Τ𝑘𝐴𝑙𝑚 𝑅
𝑟2 − 𝑟1 ln 𝑟2 Τ𝑟1
𝑅= = (5.27)
𝑘𝐴𝑙𝑚 2𝜋𝑘𝐿
𝐴1 +𝐴2
❖ In engineering practice, if A2/A1 < 1.5/1, the linear mean area of is within 1.5% of the log mean area.
2
Problem 5.4: A thick-walled cylindrical tubing of hard rubber having an inside radius of 5 mm and an outside radius of 20 mm is
being used as a temporary cooling coil in a bath. Ice water is flowing rapidly inside, and the inside wall temperature is 274.9 K. The
outside surface temperature is 297.1 K. A total of 14.65 W must be removed from the bath by the cooling coil. How many m of
tubing are required?
Solution: The thermal conductivity (k) at 0 °C (273 K) is 0.151 W/m. K. Since data at other temperatures are not available, this value
will be used for the range of 274.9 to 297.1 K.
Here, r1 = (5/1000) m = 0.005 m, r2 = (20/1000) m = 0.02 m, T1 = 274.9 K, T2 = 297.1 K and q = 14.65 W.
The calculation will be done first for a length of 1 m of tubing (l). Therefore,
𝐴1 = 2𝜋𝑟1 𝑙 = 2 × 𝜋 × 0.005 × 1 = 0.0314 m2 𝐴2 = 2𝜋𝑟2 𝑙 = 2 × 𝜋 × 0.02 × 1 = 0.1257 m2
𝐴2 −𝐴1 0.1257−0.0314 𝑇1 −𝑇2 274.9−297.1
𝐴𝑙𝑚 = = = 0.0680 m2 𝑞 = 𝑘𝐴𝑙𝑚 = 0.151 × 0.0680 × = −15.2 W
ln 𝐴2 Τ𝐴1 ln 0.1257Τ0.0314 𝑟2 −𝑟1 0.02−0.005
The –ve sign indicates that the heat flow is from r2 to r1. Since 15.2 W is removed for 1 m length, the needed length is
14.65
𝐿= = 0.964 m (Answer).
15.2
Problem 5.5: Find the heat transfer rate per one metre length of circular copper pipe of inside dia 10 cm and outside dia 12.5 cm.
The inside temperature of the pipe is 100 °C and the outside temperature of the pipe is 40 °C. Also calculate the percentage error
involved by using average radius in place of log mean radius. The thermal conductivity of the metal is 850 W/m. °C.
Solution: L = 1 m, r1 = 0.05 m, r2 = 0.0625 m, T1 = 100 °C, T2 = 40 °C, k = 850 W/m. °C.
𝑇1 −𝑇2
𝑞2 = = 1441.668 kW Avg. radius = 0.056 m Avg. area = 0.3534 m2
𝑟2 −𝑟1 ൗ𝑘𝐴𝑎𝑣𝑔
1442−1432.08
% error = × 100 = 0.69%
1432.08
Conductive Heat Transfer – Through a Multilayer Cylinder
❖ In the process industries, heat transfer often occurs through multilayers of cylinders, as for example when heat is being
transferred through the walls of an insulated pipe.
❖ Figure 5.6 shows a pipe with two layers of insulation around it i.e., a total of three concentric hollow cylinders.
❖ The temperature drop is T1 – T2 across material A, T2 – T3 across B, and T3 – T4 across C.
❖ The heat transfer rate q will, of course, be the same for each layer as we are at steady state.
❖ Writing an equation similar to Eqn. (5.25) for each concentric cylinder,
𝑇1 − 𝑇2 𝑇2 − 𝑇3 𝑇3 − 𝑇4
𝑞= = = (5.28)
𝑟2 − 𝑟1 Τ𝑘𝐴 𝐴𝐴 𝑙𝑚 𝑟3 − 𝑟2 Τ𝑘𝐵 𝐴𝐵 𝑙𝑚 𝑟4 − 𝑟3 Τ𝑘𝐶 𝐴𝐶 𝑙𝑚
where
𝐴2 − 𝐴1 𝐴3 − 𝐴2 𝐴4 − 𝐴3
𝐴𝐴 𝑙𝑚 = 𝐴𝐵 𝑙𝑚 = 𝐴𝐶 𝑙𝑚 =
ln 𝐴2 Τ𝐴1 ln 𝐴3 Τ𝐴2 ln 𝐴4 Τ𝐴3
𝑇1 − 𝑇4 𝑇1 − 𝑇4
𝑞= = (5.30)
𝑅𝐴 + 𝑅𝐵 + 𝑅𝐶 σ𝑅
Hence, the overall resistance is again the sum of the individual resistances in series.
Problem 5.6: A thick-walled tube of stainless steel (A) having k = 21.63 W/m. K with dimensions of 0.0254 m ID and 0.0508 m
OD is covered with a 0.0254 m thick layer of an insulation (B) having k = 0.2423 W/m. K. The inside wall temperature of the pipe is
811 K and the outside surface of the insulation is at 310.8 K. For a 0.305 m length of pipe, calculate the heat loss and also the
temperature at the interface between the metal and the insulation.
Solution: Calling T1 = 811 K, T2 the interface, and T3 = 310.8 K, the dimensions are
0.0254 0.0508
𝑟1 = = 0.0127 m 𝑟2 = = 0.0254 m 𝑟3 = 0.0508 m
2 2
Conductive Heat Transfer – Through a Hollow Sphere
❖ Heat conduction through a hollow sphere is another case of one dimensional conduction.
❖ Using Fourier’s law for constant thermal conductivity with distance 𝑑𝑟, where r is the radius of the sphere.
𝑞 𝑑𝑇
= −𝑘 (5.31)
𝐴 𝑑𝑟
❖ The c/s area normal to the heat flow is
𝐴 = 4𝜋𝑟 2 (5.32)
❖ Substituting Eqn. (5.29) into Eqn. (5.28), rearranging and integrating we get,
𝑟2 𝑇2
𝑞 𝑑𝑟
න 2 = −𝑘 න 𝑑𝑇 (5.33)
4𝜋 𝑟
𝑟1 𝑇1
𝑞 1 1 4𝜋𝑟1 𝑟2 𝑘 𝑇1 − 𝑇2
− = 𝑘 𝑇1 − 𝑇2 ⟹ 𝑞 = 𝑇1 − 𝑇2 = (5.34)
4𝜋 𝑟1 𝑟2 𝑟2 − 𝑟1 𝑟2 − 𝑟1 Τ4𝜋𝑟1 𝑟2 𝑘
Conduction through Materials in Parallel
❖ Suppose that two plane solids A and B are placed side by side in parallel, and the direction
of heat flow is perpendicular to the plane of the exposed surface of each solid.
❖ Then the total heat flow is the sum of the heat flow through solid A plus that through solid
B. Writing Fourier’s equation for each solid and summing,
𝑘𝐴 𝐴𝐴 𝑘𝐵 𝐴𝐵
𝑞𝑇 = 𝑞𝐴 + 𝑞𝐵 = 𝑇1 − 𝑇2 + 𝑇 − 𝑇4 (5.35)
∆𝑥𝐴 ∆𝑥𝐵 3
where qT = total heat flow; T1 and T2 are the front and rear surface temperatures for solid A; T3 and T4 are the front and rear
surface temperatures for solid B.
❖ If we assume that T1 = T3 (front temperatures same for A and B) and T2 = T4 (equal rear temperatures),
𝑇1 − 𝑇2 𝑇1 − 𝑇2 1 1
𝑞𝑇 = + = + 𝑇1 − 𝑇2 (5.36)
∆𝑥𝐴 Τ𝑘𝐴 𝐴𝐴 ∆𝑥𝐵 Τ𝑘𝐵 𝐴𝐵 𝑅𝐴 𝑅𝐵
❖ An example would be an insulated wall (A) of a brick oven where steel reinforcing members (B) are in parallel and penetrate
the wall. Even though the area AB of the steel would be small compared to the insulated brick area AA, the higher conductivity
of the metal (which would be several hundred times larger than that of brick) could allow a large portion of the heat lost to be
conducted by the steel.
❖ Another example is a method of increasing heat conduction to accelerate the freeze drying of meat. Spikes of metal in the
frozen meat conduct heat more rapidly into the insides of the meat.
❖ It should be mentioned that in some cases two-dimensional heat flow can occur if the thermal conductivities of the materials
in parallel differ markedly. Then the results using Eqn. (5.36) would be somewhat affected.
❖ Upto now we have seen the steady state heat
conduction process through solids which have a
constant temperature on the heating side and
cooling side.
❖ Since sufficient time has been given, steady state
conditions prevail and make the temperature
gradient in the slab to be a straight line (Figure
5.7).
❖ However, in real situations, many times steady state
conditions may not prevail. This kind of heat
transfer by conduction is known as unsteady state
conduction.
Figure 5.7: Steady state conduction. Figure 5.8: Unsteady state conduction.
❖ Let us consider a situation where a solid slab is being heated from both sides (Figure 5.8). It could be a slice of bread in a
toaster or a slab of frozen meat kept outside or kept in an oven where it is heated by radiation. U can also consider a slab of
wada or papad being deep fried in an oil medium.
❖ In all these cases, heat transfer within the solid slabs occur by conduction.
❖ In the unsteady process, the temperature within the slab varies both with time 𝒕, and position in the slab i.e., distance
𝒙.
❖ Let us visualise a small elemental piece of thickness 𝑑𝑥 at a distance 𝑥 from the heating surface (Figure 5.8).
❖ Let us say the temperature profile at the entrance of the strip is 𝜕𝑇Τ𝜕𝑥. We use partial derivatives here because T varies with
both 𝑡 and 𝑥.
𝜕𝑇
❖ Hence, heat entering the strip at position A in a time interval of 𝑑𝑡 = −𝑘𝐴 𝑑𝑡 (X1)
𝜕𝑥
𝜕 𝜕𝑇
❖ The variation in the temperature profile is given by 𝑑𝑥
𝜕𝑥 𝜕𝑥
𝜕𝑇 𝜕 𝜕𝑇
❖ Hence, the temperature profile at position B is: + 𝑑𝑥 (X2)
𝜕𝑥 𝜕𝑥 𝜕𝑥
𝜕𝑇 𝜕 𝜕𝑇
❖ Heat leaving the strip at position B in time interval 𝑑𝑡 = −𝑘𝐴 + 𝑑𝑥 𝑑𝑡 (X3)
𝜕𝑥 𝜕𝑥 𝜕𝑥
❖ Therefore, heat accumulation in the strip is the difference of Eqn. (X1) and Eqn. (X3) and is equal to
𝜕𝑇 𝜕𝑇 𝜕 𝜕𝑇 𝜕2𝑇
−𝑘𝐴 𝑑𝑡 + 𝑘𝐴 + 𝑑𝑥 𝑑𝑡 = kA 𝑑𝑥𝑑𝑡 (X4)
𝜕𝑥 𝜕𝑥 𝜕𝑥 𝜕𝑥 𝜕𝑥 2
❖ Heat accumulation in the strip = (mass of the strip) × (specific heat of the material) × (temperature variation with time) ×
𝜕𝑇
(incremental time) = 𝜌𝐴𝑑𝑥 × 𝐶𝑝 × 𝑑𝑡 (X5)
𝜕𝑡
𝜕2 𝑇 𝜕𝑇
❖ Therefore, kA 𝑑𝑥𝑑𝑡 = 𝜌𝐶𝑝 𝐴𝑑𝑥 𝑑𝑡 (X6)
𝜕𝑥 2 𝜕𝑡
𝑘
❖ Here = 𝛼, which is nothing but thermal diffusivity having unit of m2/s.
𝜌𝐶𝑝
where Tav is average temperature of slab at a time interval of tT, T0 is the initial temperature of the slab.
𝛼𝑡𝑇
𝑁𝐹𝑜 = 2 (X10)
𝑠
Here NFo is a dimensionless number and is known as Fourier Number, tT is total time period, s is the half thickness of the
slab and is equal to B/2.
❖ Half slab thickness is used since the heating is from both sides. If the heating is only from one side, as in the case of baking
of chapati or dosa on a hot pan, then s is replaced by full thickness 2s (B).
❖ Similar type of expression can also be derived for infinitely long cylinder of radius r (where L is much larger compared to r),
and is given by McAdams (1954).
𝑇𝑆 − 𝑇𝑎𝑣
= 0.692 exp −5.78𝑁𝐹𝑜 + 0.131 exp −30.5𝑁𝐹𝑜 + 0.0534 exp −74.8𝑁𝐹𝑜 + ⋯ (X11)
𝑇𝑆 − 𝑇0
𝑇𝑆 − 𝑇0 150 − 30 8 𝑇𝑆 − 𝑇0
= = 2.4 = 1.945
𝑇𝑆 − 𝑇𝑎𝑣 150 − 100 𝜋 2 𝑇𝑆 − 𝑇𝑎𝑣
2
1 12×10−3
𝑡𝑇 = ln 1.945 = 98.69 s ≈ 99 s
9.835×10−8 𝜋
Convective Heat Transfer
❖ Here we will study the process of heat transfer in fluids (liquids, vapours and gases) from a metallic surface or a solid
surface at higher temperature to the fluids at lower temperature which are in contact with it.
❖ Here we see that the fluid layers in contact with the hot surface virtually move upwards because of density differences
caused by temperature differences.
❖ This kind of heat transfer in fluids due to physical movement of layers or droplets of the fluids is known as convection.
Most common example: heating of water in a vessel or in a beaker.
❖ If we start heating water in a vessel, slowly some bubbles start
forming at the bottom of the vessel if the surface temperature is
high (the temp. difference is ≥ 50 °C).
❖ These bubbles consist of water vapour. They rise to the top due
to density difference. Figure 5.7: Convective heating process.
❖ This is an example of convection, and the type of boiling is
known as nucleate boiling.
❖ If the temperature is not very high, we still find some movement of layers of water slowly rising to the top.
❖ The upper layers which are at a relatively less temperature, and hence, are dense will go to the bottom.
❖ Thus there will be a continuous exchange of layers of less density from the bottom going to the top, and the layers of higher
density going to the bottom (Figure 5.7) until the temperature of all layers becomes the same which is almost equal to the
temperature of the surface. Later there won’t be further heat transfer.
❖ If the temperature of the surface is very high (the temperature difference is of the order of 100 °C), the bubbles at the bottom
coalesce and form into a thin quiescent film of vapour (Figure 5.8). The heat transfer has to take place through the film.
❖ This kind of boiling is known as film boiling.
❖ We observe this kind of phenomenon only when
heat transfer is associated with phase change.
❖ The point at which transition occurs between
nucleate boiling and film boiling is known as
Leidenfrost point. Figure 5.8: Concept of film boiling.
Natural Convection and Forced Convection
❖ The type of heat transfer process (Figure 5.7) where heat transfer takes place mainly due to the movement of hot layers and cold
layers due to temperature induced density differences is known as natural convection. The movement of different molecules or
layers of the fluids is by means of gravitational force and buoyancy force. The natural convection is also known as free
convection.
❖ In this type of convection, the heat transfer rates are low , and are generally ≈ 60 W/m2 °C for air and 60 – 3000 W/m2 °C for
water systems.
❖ The heat transfer rates mentioned above in the units of W/m2 °C are known as heat transfer coefficients. The low heat transfer
rates have necessitated the need to increase them by using some additional mechanical force.
❖ Let us imagine a situation where the heating process is assisted by means of a mechanical stirring given to the liquid. The
movement of different layers is faster either through stirring, agitation or by external pumping system. As a result, heat transfer is
faster. This kind of convection is known as forced convection.
❖ Hence, the convective heat transfer rates are not only dependent upon the fluid properties like viscosity, density, thermal
conductivity, but also dependent upon external aid.
❖ The external aids can be characterized by the rate of stirring, type of the agitator, or mechanical horsepower of
the pump used in case of external recirculation, or the velocity of fluid flow and extent of turbulence.
❖ In addition to the above, the heat transfer rates are also dependent upon the system characteristics viz., diameter
of the conduit, smoothness of the surface of the conduit, thermal conductivity of the metal, presence or
absence of baffles in the vessel which break the formation of the vortex and help in creation of better
turbulence.
❖ Indeed convective heat transfer is the major method of heat transfer in any process operation between the
fluids and solid surfaces, which contain fluids.
Heat Transfer Between a Metallic Wall and the Fluid
❖ Whenever the heat is transferring from a metallic wall to a fluid
adjacent to it (it could be in the reverse direction also), it is presumed
that there exists a fluid film of constant thickness (Figure 5.9), which
is responsible for the resistance to heat transfer.
❖ This concept was originally proposed by Prandtl.
❖ The thickness of the film may vary depending upon the agitation or
movement of the fluid. More is the agitation, thinner is the film.
However, it cannot be totally removed.
❖ Within the film, the temperature varies from the wall temperature Figure 5.9: Heat transfer between
metallic wall and fluid.
(TS) to the average temperature of the fluid.
Thermal Boundary Layer
❖ The film within which the temperature varies is known as thermal boundary layer (TBL)
as contrast to the hydrodynamic boundary layer (HBL).
❖ The relative thickness of both the boundary layers are dependent upon a dimensionless
number known as Prandtl number (NPr) and is defined as:
𝐶𝑝 𝜇
𝑁𝑃𝑟 = (5.37) Figure 5.10: Boundary layer concept (T>TW).
𝑘
❖ HBL is thicker than the TBL if NPr > 1.0 which is mostly the case with most of the liquids. HBL is equal to TBL if NPr ≅ 1.0 which is mostly
true for most of the gases and vapours. HBL is thinner than TBL if NPr < 1.0 which happens rarely and is true for molten metals, etc.
❖ The TBL concept is very important in understanding the process of heat transfer in most of the process operations.
❖ This prohibits us from using very high wall temperatures in heating liquid foods because the temperature of the liquid food in the stagnant
film adhering to the wall is equal to the temperature of the wall, and hence, may deteriorate or char the food.
❖ This is why milk heated in a stainless steel vessel on a gas stove invariably results in charring the milk at edges. This is also precisely the reason
why we always prefer to use steam jacketed SS vessels where we can have better control on temperature rather than using direct flame stoves
like gas stoves, etc.
Heat Transfer Coefficient
❖ As originally proposed by Prandtl, the resistance to heat transfer lies in the fluid film adjacent to the hot plate/surface.
❖ The heat transfer rate maybe represented by Fourier law through stagnant fluid film as follows:
𝑞 𝑘
= 𝑇 −𝑇 (5.38)
𝐴 𝛿 𝑠
where 𝛿 is the film thickness and k is the thermal conductivity. Since it is difficult to measure 𝛿, we combine (k/𝛿) into a term
called as the convective heat transfer coefficient and is abbreviated as h. Equation 5.38 can be written as:
𝑞 = ℎ𝐴 𝑇𝑠 − 𝑇 (5.39)
or in differential form, it can be written as:
𝑑𝑞
= ℎ 𝑇𝑠 − 𝑇 (5.40)
𝑑𝐴
in which h has the units of W/m2.°C and A is the area in m2. T is the average temperature of the fluid and TS is the surface
temperature. The reciprocal of h is known as the resistance (R).
1 (5.41)
𝑅=
ℎ
❖ The heat flux at the wall/surface is given by the conductive heat transfer through the wall. Hence,
𝑑𝑞 𝑑𝑇
= −𝑘 (5.42)
𝑑𝐴 𝑑𝑦 𝑤
𝑑𝑇 𝑑𝑇Τ𝑑𝑦 𝑤
ℎ 𝑇𝑆 − 𝑇 = −𝑘 ⟹ ℎ = −𝑘 (5.43)
𝑑𝑦 𝑤
𝑇𝑆 − 𝑇
*h is always positive. Since y is measured in the direction of heat transfer, dT/dy is negative. If Eqn. 5.43 is multiplied on both
sides by d/k, we get
ℎ𝑑 −𝑑 𝑑𝑇Τ𝑑𝑦 𝑤
= 𝑁𝑢𝑠𝑠𝑒𝑙𝑡 𝑛𝑢𝑚𝑏𝑒𝑟 𝑁𝑁𝑢 = (5.44)
𝑘 𝑇𝑆 − 𝑇
NNu is one of the important dimensionless number, and is very useful in convective heat transfer. Most of the equations for
heat transfer are based on Nusselt number.
❖ The heat transfer coefficient h is dependent upon a number of fluid properties and system parameters.
❖ Fluid properties which affect h are:
We know,
o Density (𝜌)
o Viscosity (𝜇) 𝑘
ℎ=
𝛿
o Thermal conductivity (k)
From Fourier’s law
o Specific heat (cP)
𝑘
ℎ=
❖ System properties which affect h are: 𝑥
o Dia or length of pipe (d or l) 𝑥 is the thickness.
Therefore Nusselt number,
o Velocity of the fluid (u)
ℎ𝑑 𝑘 𝑑 𝑑 𝑑
o Roughness of the surface wall, etc. 𝑁𝑁𝑢 = = × = =
𝑘 𝛿 𝑘 𝛿 𝑥
Thus, Thus, Nusselt number is the ratio of the pipe
ℎ = 𝑓 𝜌, 𝑑, 𝑢, 𝜇, 𝑐𝑃 , 𝑘 (5.45) diameter to the thickness of the laminar layer.
Problem 5.7: A fruit juice is passing through a hot pipe of 7.5 cm diameter and 3.3 m long at 60 lpm and the wall temperature is
120 °C. The juice entering at 25 °C and is leaving at 75 °C. Find the heat transfer coefficient and Nusselt number if the fluid
properties are assumed to be those of water.
Solution: The fluid properties are:
𝜌 = 1000 kg/m3, k = 0.6 W/m. °C = 0.6 × 10-3 kW/ m. °C, CP =4.2 kJ/kg. °C, 𝜇 = 1 × 10-3 Pa.s.
Heat transfer rate of fluid, 𝑞 = 𝑚𝑐
ሶ 𝑃 ∆𝑇
60
𝑚ሶ = Volume flow rate × density = × 10−3 × 1000 = 1 kg/s
60
∆𝑇1 = 120 − 25 = 95 °𝐶
∆𝑇2 = 120 − 75 = 45 °𝐶
∆𝑇1 + ∆𝑇2 95 + 45
∆𝑇 = = = 70 °𝐶
2 2
∴ 210 = ℎ × 0.778 × 70
210
ℎ = 0.778×70 = 3.86 kJ/s.m2 °C
a) Heat transfer between the wall and the hot fluid in the shell side by convection
which is designated as outside heat transfer process and the heat transfer
coefficient is represented by ho.
b) Heat transfer through the metallic wall by conduction.
c) Heat transfer between the hot wall and the cold fluid in the tube side by
convection which is designated as inside heat transfer process, and the heat
transfer coefficient is represented by hi.
Figure 5.11: Convective heat transfer between
two fluids through a metallic conductor.
𝑑𝑞 𝑑𝑞
Inside heat transfer flux = = ℎ𝑖 𝑇𝑆𝑐 − 𝑇𝑐 (5.46) Outside heat transfer rate = = ℎ𝑜 𝑇ℎ − 𝑇𝑆ℎ (5.47)
𝑑𝐴𝑖 𝑑𝐴𝑜
𝑑𝑞 𝑘𝑚
Heat transfer through the wall = = 𝑇𝑆ℎ − 𝑇𝑆𝑐 (5.48)
𝑑𝐴𝐿 𝑥𝑤
Since it is not always possible to measure the wall temperatures (𝑇𝑆ℎ and 𝑇𝑆𝑐 ), we consider only one heat transfer coefficient
based on the temperatures of the hot side and cold side viz. 𝑇ℎ and 𝑇𝑐 . We can now write,
𝑞 = 𝑈𝐴∆𝑇 (5.52)
where U is the overall heat transfer coefficient.
In differential form, we can write
𝑑𝑞
= 𝑈∆𝑇 (5.53)
𝑑𝐴
The overall heat transfer coefficient can now be defined based on either inside conditions or outside conditions and accordingly
called as:
o Inside overall heat transfer coefficient 𝑈𝑖
o Outside overall heat transfer coefficient 𝑈𝑜
𝑑𝑞 𝑇ℎ − 𝑇𝑐
= (5.56)
𝑑𝐴𝑜 𝑑𝑜 1 𝑥𝑤 𝑑𝑜 1
+ +
𝑑𝑖 ℎ𝑖 𝑘𝑚 𝑑𝐿 ℎ𝑜
Combining Eqns. 5.55 and 5.56 we get: Since (1/U) is the overall resistance (R) for heat transfer,
or 1 𝑑𝑜 1 𝑥𝑤 𝑑𝑜 1 𝑅 = 𝑅𝑖 + 𝑅𝑚 + 𝑅𝑜 (5.61)
= + + (5.58)
𝑈𝑜 𝑑𝑖 ℎ𝑖 𝑘𝑚 𝑑𝐿 ℎ𝑜 where R is the resistance for heat transfer and Rm is the
resistance for heat transfer by the metallic wall.
1 1 𝑥𝑤 𝑑𝑖 𝑑𝑖 1 (5.59)
And similarly = + +
𝑈𝑖 ℎ𝑖 𝑘𝑚 𝑑𝐿 𝑑𝑜 ℎ𝑜
1 1 𝑥𝑤 1 (5.60)
For a slab we can simply write, = + +
𝑈 ℎ𝑖 𝑘𝑚 ℎ𝑜
Problem 5.8: A 100 L of dilute sugar syrup is heated in a hemispherical jacketed SS vessel of 80 cm diameter with little agitation
under 2 atm pressure. The wall thickness of the vessel is 3 mm. The sugar syrup, whose physico-chemical properties maybe
assumed to be those of water, is heated from 28 °C to 75 °C. What is the heat transfer rate (q)? What is the quantity of steam
required, if the inside heat transfer coefficient is 2 kW/m2 °C, and the outside heat transfer coefficient is 1.5 kW/m2 °C?
Solution: Approach:
i. Initially we calculate what is the quantity of heat required to raise the temperature of 100 L dilute sugar syrup from 28 °C
to 75 °C. From this, we can find out the quantity of steam required.
ii. We calculate U.
iii. We calculate the heat transfer rate. Thermal conductivity of steel
= 43 × 10-3 kW/m. °C.
Properties of sugar syrup:
𝜌 = 1000 kg/m3 k = 0.628 W/m. °C Thickness of wall = 3 mm =
(average at 40 °C) = 3 × 10-3 m.
𝜇 = 1 × 10-3 Pa.s
0.628 × 10-3 kW/m.
Cp = 4.2 kJ/kg. °C °C
∆𝑇 = Temperature rise of sugar syrup = 75 – 28 °C = 47 °C 𝑄
The quantity of steam required =
𝜆𝑠
Quantity of heat required to raise the temperature of 100 L sugar syrup from 28
where 𝜆𝑠 is latent heat of vaporization of
to 75 °C, Q = 𝑚𝐶
ሶ 𝑝 ∆𝑇
steam and is noted from steam table.
𝑄 = 100 × 10−3 × 1000 × 4.2 × 75 − 28 = 19740 kJ.
𝜆𝑠 = 2706.3 − 503.72 = 2202.6 kJ/kg.
Now, Therefore, quantity of steam required =
1 1 𝑥𝑤 1 1 3×10−3 1
= + + = + + = 1.236 m2. °C/kW. 19740
𝑈 ℎ𝑖 𝑘𝑚 ℎ𝑜 2 43×10−3 1.5 = 8.96 kg ≈ 9 kg.
2202.6
Again, 𝑞 = 𝑈𝐴∆𝑇. To find the value of q, we need to know what is ∆𝑇. The steam temperature outside the vessel corresponding
to 2 atm. pressure is 120 °C (from steam table).
The inside temperature is averaged out to be = (28 + 75)/2 = 51.5 °C. 0.8 2
𝐴 = 2𝜋𝑟 2 = 2 × 𝜋 × = 1.01 m2
2
Therefore, ∆𝑇 = 120 – 51.5 = 68.5 °C.
Hence, 𝑞 = 𝑈𝐴∆𝑇 = 0.8088 × 1.01 × 68.5 = 55.96 kW. (Answer)
❖ Till now we have only considered steady state heat transfer.
❖ However, the last problem is more applicable for unsteady heat transfer as there heat transfer process was going on in an
unsteady manner, and ∆𝑇 changed with time.
❖ In the case of unsteady heat transfer, we will follow the approach of heat transfer for a spherical body, for which the heat
transfer coefficient for the sphere is approximated to be 5k/r m. That is, the whole fluid is presumed to be contained in a
metallic spherical body of radius r m, and is being heated. Hence, the volume of the fluid contained in the spherical vessel is
4/3𝜋𝑟𝑚 3 and the heating surface area is 4𝜋𝑟𝑚 2 .
❖ We can write heat balance for unsteady state conditions in which the rate of temperature change is dT/dt and the temperature
change is ∆𝑇.
∆𝑇 = 𝑇𝑠 − 𝑇
where TS is the constant surface temperature and T is the temperature of the fluid at time t.
4 𝑑𝑇
𝜌𝐶𝑝 𝜋𝑟𝑚 3 = ℎ𝑖 𝐴∆𝑇 (5.62)
3 𝑑𝑡
Here,
5𝑘
ℎ𝑖 = (5.63)
𝑟𝑚
and 𝐴 = 4𝜋𝑟𝑚 2
Therefore,
4 3
𝑑𝑇 5𝑘
𝜌𝐶𝑝 𝜋𝑟𝑚 = 4𝜋𝑟𝑚 2 (𝑇𝑠 −𝑇)
3 𝑑𝑡 𝑟𝑚
𝑑𝑇 15 𝑘
or = 2 𝑑𝑡 (5.64)
𝑇𝑆 − 𝑇 𝑟𝑚 𝜌𝐶𝑝
On integration between the limits of initial temperature T0 to the final temperature T at time t,
𝑇𝑆 − 𝑇0 15 (5.65)
ln = 2 𝛼𝑡
𝑇𝑆 − 𝑇 𝑟𝑚
❖ Eqn. 5.65 seems to be over simplification, and does not account for any forced convection conditions prevailing in the heating
system.
❖ Irrespective of the forced convection conditions prevailing in the vessel, we get the same value for t for a given TS, T0 and T,
since the diffusivity (α) for the fluid does not change. This needs a relook into the situation and the validity of Eqn. 5.65.
❖ Alternatively, we can write Eqn. 5.62 as follows:
𝑑𝑇
𝜌𝐶𝑝 𝑉 = ℎ𝑖 𝐴 𝑇𝑆 − 𝑇 = 𝑞 (5.66)
𝑑𝑡
where V is the volume of the fluid, A is the heat transfer area and hi is the inside heat transfer coefficient.
❖ The preceding equation is applicable only when there is no internal resistance to heat transfer or negligible internal
resistance. This kind of situation occurs with most of the foods with high thermal conductivity or if the liquid foods are well
stirred so that there will not be any temperature gradients within the liquid foods.
❖ Eqn. 5.66 can be integrated with the boundary conditions that T = T0 when t = 0 and T = T when t = t.
𝑇𝑆 − 𝑇0 ℎ𝑖 𝐴𝑡
ln = (5.67)… for heating or cooling of liquid foods with negligible internal resistance to heat transfer.
𝑇𝑆 − 𝑇 𝜌𝑉𝐶𝑝
Problem 5.9: For the problem 5.8, find the time taken for the fluid to reach the final temperature, and also compare the time
taken by problem 5.8.
Solution: It is a direct case of Eqn. 5.67.
TS = 120 °C, T0 = 28 °C, T = 75 °C, k = 0.628 × 10-3 kW/m. °C, 𝜌 = 1000 kg/m3 and Cp = 4.2 kJ/kg. °C.
Therefore, 𝑘 0.628×10−3
𝛼= = = 1.5 × 10−7 m2/s
𝜌𝐶𝑝 1000×4.2
𝑇𝑆 − 𝑇0 120 − 28 𝑇𝑆 − 𝑇0
= = 2.04 ln = ln 2.04 = 0.715
𝑇𝑆 − 𝑇 120 − 75 𝑇𝑆 − 𝑇
2×1.01×𝑡
∴ 0.715 = ⟹ 𝑡 = 148.66 ≈ 149 s = 2.48 min (Answer)
1000×0.1×4.2
Now we will see how the results will be by using Eqn. 5.65.
We have to calculate 𝑟𝑚 assuming all 100 L of the fluid is contained in a spherical container.
4
100 × 10−3 = 𝜋𝑟𝑚 3 ⟹ 𝑟𝑚 = 0.2879 m
3
By assuming average temperature for the fluid and heat transfer by steady state:
𝑑𝑞
= 𝑈 ∆𝑇 = 𝑈 𝑇ℎ − 𝑇𝑐 5.68
𝑑𝐴
Where Th is the average temperature of the hot fluid and Tc is the average temperature of the cold fluid.
Integration of Eqn. (5.68) gives the heat transfer rate in the entire area of the heat exchanger. The integration is possible with the
following assumptions:
o The overall heat transfer coefficient is constant in entire range of temperatures.
o The physico-chemical properties of the fluids do not change considerably with the temperature.
o Heat lost to the environment from the shell is negligible.
o Flow through the tube is steady.
The slope of the line in Fig. 5.13 is given by
𝑑 ∆𝑇 ∆𝑇2 − ∆𝑇1
= 5.69
𝑑𝑞 𝑞𝑇
We replace 𝑑𝑞 in Eqn. 5.69 with Eqn. 5.68,
𝑑 ∆𝑇 ∆𝑇2 − ∆𝑇1
= 5.70
𝑈𝑑𝐴 ∆𝑇 𝑞𝑇
∆𝑇2 𝐴𝑇
𝑑 ∆𝑇 𝑈 ∆𝑇2 − ∆𝑇1
න = න 𝑑𝐴 Application of preceding equation has the
∆𝑇 𝑞𝑇
∆𝑇1 0 following constraints:
a) It may give erratic results if q is not
∆𝑇2 𝑈𝐴 𝑇 ∆𝑇2 − ∆𝑇1 changing linearly with ∆𝑇 as drawn in
ln = 5.71 Fig. 5.13.
∆𝑇1 𝑞𝑇
b) U should not vary drastically with ∆𝑇.
On rearrangement
∆𝑇2 − ∆𝑇1
𝑞𝑇 = 𝑈𝐴 𝑇 = 𝑈𝐴 𝑇 ∆𝑇𝐿 5.72
ln ∆𝑇2 Τ∆𝑇1
where ∆𝑇𝐿 is known as the log mean temperature difference (LMTD).
Problem 5.10: A dilute pineapple juice is heated in a double pipe heat exchanger from 28 °C to 75 °C by heat exchanging with
hot water flowing in shell in counter current direction, which is cooled from 95 °C to 85 °C. Calculate the LMTD and compare
with average temperature difference.
∆𝑇2 −∆𝑇1
Solution: LMTD = = 35.32
ln ∆𝑇2 Τ∆𝑇1
∆𝑇2 +∆𝑇1
Average Temperature = = 38.5
2
Correlations for Heat Transfer Coefficient
❖ Heat transfer coefficients are to be generally measured experimentally. However, some correlations are available which are all
based on the physico-chemical and thermal properties of the fluid, the flow conditions and system geometry.
❖ Most of these parameters are characterized by means of various dimensionless numbers.
❖ Reynolds number, 𝑑𝑢𝜌
𝑁𝑅𝑒 =
𝜇
ℎ𝑑
❖ Nusselt number, 𝑁𝑁𝑢 =
𝑘
𝐶𝑝 𝜇
𝑁𝑃𝑟 =
❖ Prandtl number, 𝑘
❖ Grashof number (NGr) is a dimensionless number which relates a large number of parameters viz., a characteristic
dimension of the system like diameter (d), accl. due to gravity (g), coeff. of thermal expansion (β), viscosity and density of
the fluid and temperature difference (∆T).
𝑑3 𝑔𝛽𝜌2 ∆𝑇
𝑁𝐺𝑟 =
𝜇2
The number is a ratio of the force of gravity to the buoyant forces resulting due to ∆T of the fluid, and is associated
with free convection when the fluid movement is mainly due to gravitational pull and buoyancy forces, resulting due to
temperature difference. L/d ratio also plays a significant role if the flow is in laminar conditions.
❖ Pecklet number (NPe) is a product of the Reynolds number and Prandtl number.
𝑑𝑢𝜌 𝐶𝑝 𝜇 𝑑𝑢𝜌𝐶𝑝
𝑁𝑃𝑒 = 𝑁𝑅𝑒 𝑁𝑃𝑟 = × =
𝜇 𝑘 𝑘
❖ Stanton number (NSt) is associated with NNu, NPr and NRe as follows:
𝑁𝑁𝑢 𝑁𝑁𝑢 ℎ𝑑 𝑘 ℎ
𝑁𝑆𝑡 = = = × = where G (= 𝑢𝜌) is known as mass flux.
𝑁𝑃𝑟 × 𝑁𝑅𝑒 𝑁𝑃𝑒 𝑘 𝑑𝑢𝜌𝐶𝑝 𝐶𝑝 𝐺
𝑚𝐶
ሶ 𝑝 𝜋 𝑑
𝑁𝐺𝑧 = = 𝑁𝑅𝑒 × 𝑁𝑃𝑟
𝑘𝐿 4 𝐿
where 𝑚ሶ is mass flow rate in kg/s and L is the length of the pipe or plate. This is also similar to NPe. We may also write
𝜋𝑑
𝑁𝐺𝑧 = 𝑁
4 𝐿 𝑃𝑒
The Graetz number is also related to Fourier number as follows;
𝜋
𝑁𝐺𝑧 =
𝑁𝐹𝑜
𝛼𝑡𝑇
where 𝑁𝐹𝑜 =
𝑟𝑚 2
and 𝐿
𝑡𝑇 =
𝑢
❖ After having defined a number of dimensionless numbers, we shall see how these are useful for calculating the heat transfer
coefficients.
❖ By and large most of these correlations are written in the form of expressions relating Nusselt number to Reynolds number
and Prandtl number. If the viscosity of the fluid changes significantly with temperature as compared to the bulk
temperature of the fluid, we also add an additional term, called 𝜇 Τ𝜇𝑤 where 𝜇𝑤 is the viscosity of the fluid at the wall
temperature. In the light of the above discussion, we may write,
𝐿 𝜇
𝑁𝑁𝑢 = 𝑓 𝑁𝑅𝑒 , 𝑁𝑃𝑟 , 𝑁𝐺𝑟 , ,
𝑑 𝜇𝑤
❖ In case of free convection, NGr is important and L/d is also associated with the correlations particularly in case of laminar
flow. Generally we may write, where 𝜓 is a constant, a is a coefficient which varies between 0.6 to 0.8
𝑐 0.14
𝑎 𝑏
𝑑 𝜇 and is generally of the order of 2/3, b is a coefficient which is generally
𝑁𝑁𝑢 = 𝜓 𝑁𝑅𝑒 𝑁𝑃𝑟
𝐿 𝜇𝑤 0.14
1/3. 𝜇 Τ𝜇𝑤 is represented by 𝜙𝑣 ,
0.14
The value of 𝜇 Τ𝜇𝑤 is generally taken as 1.0, since in food processing we generally do not tend to use high wall
temperatures, and hence, variation of 𝜇𝑤 as compared to 𝜇 is not very significant.
Equations for Free (Natural) Convection
❖ There are some correlations available for calculating heat transfer coefficients for water and air in free convection mode.
❖ For horizontal cylinder,
0.25
∆𝑇
ℎ = 1.3196 × for air 5.73
𝑑
0.25
∆𝑇
ℎ = 291.1 × for water 5.74
𝑑
The above equations are applicable both for heating and cooling.
❖ There are some correlations available for calculating heat transfer coefficients for water and air in free convection mode.
❖ For a vertical cylinder or vertical plate (diameter is used for cylinders, L is used for plates)
0.25
∆𝑇
ℎ = 1.3683 × for air 5.75
𝑑
0.25
∆𝑇 for water 5.76
ℎ = 127.1 ×
𝑑
Equations for Forced Convection
❖ The forced convection process maybe looked upon into two regimes; viz. heat transfer in laminar flow conditions, heat
transfer in turbulent conditions.
❖ Laminar flow conditions: At low Graetz numbers, the Nusselt number approaches a limiting value of 3.66, and it is very
difficult to calculate accurate values for heat transfer coefficients. If the Graetz numbers are more than 20, we can write with
a reasonable amount of accuracy.
1/3
𝑁𝑁𝑢 ≅ 2.0 𝑁𝐺𝑧 5.77
The above equation is applicable for both heating and cooling, and is sometimes associated with viscosity change term as
follows:
Sometimes, the coefficient 2.0 is replaced by more accurate 1.75 and the equation is written as:
1/3 5.79
𝑁𝑁𝑢 = 1.75 𝑁𝐺𝑧 𝜑𝑣
The Eqn. 5.79 is a modified form for non-Newtonian fluids as follows by replacing the viscosities with apparent
viscosities:
1/3 1/3
3𝑛 + 1 𝜇𝑎
𝑁𝑁𝑢 = 2.0 𝑁𝐺𝑧 1/3 (5.80)
4𝑛 𝜇𝑎,𝑤
where n is the flow behaviour index of the power law fluid model and
𝜇𝑎 = 𝐾𝛾 𝑛−1
in which K is the consistency index of the power law fluid model for non-Newtonian fluids.
❖ Turbulent flow conditions: Dittus-Boelter equation adequately represents the turbulent heat transfer during flow through
pipes,
𝑁𝑁𝑢 = 0.023 𝑁𝑅𝑒 0.8 𝑁𝑃𝑟 1/3 (5.81)
Eqn. 5.81 is modified by adding the terms for viscosity correction, and is known as Sieder-Tate equation.
𝐸1 𝐸2
= (5.91)
𝛼1 𝛼2
where E1 and E2 stand for emissive power (or radiating power) of substances 1 and 2. If one of the bodies is a black body
(say body 2), Eqn. (5.91) becomes
𝐸1
= 𝐸𝑏 (5.92)
𝛼1
𝐸1
or 𝛼1 = = 𝜀1 (5.93)
𝐸𝑏
Thus, when a body is in thermal equilibrium with its surrounding, the absorbtivity of the body is equal to its emissivity.
This is an alternate statement of Kirchhoff ’s law and finds wider application.
2. Stefan-Boltzmann Law
The radiating heat from the black body is given by a law known as Stefan-Boltzmann law which states that the total emissive
power of a black body (Eb) is proportional to the forth power of its absolute temperature (T). Thus,
𝑞 𝑞
𝐸𝑏 = ∝ 𝑇 4 ⟹ 𝐸𝑏 = = 𝜎𝑇 4 (5.94)
𝐴 𝐴
The term 𝜎 is called the Stefan-Boltzmann constant (5.67 × 10-8 W/m2. K4). A is the surface area of the body in m2, and T is
the absolute temperature in K. For grey body, the emissive power (E) is expressed as
𝐸 𝐸
× 𝐸𝑏 = 𝜎 × × 𝑇 4 ⟹ 𝐸 = 𝜎𝜀𝑇 4 (5.95)
𝐸𝑏 𝐸𝑏
(5.96)
or 𝑞 = 𝜎𝐴𝜀𝑇 4
Problem 5.14: In a process plant, steam line is drawn from the boiler to the process vessels. The process vessel is covered with glass
wool and is covered with polished aluminium sheets. The outside temperature of the pipeline is 38 °C. Calculate the heat energy
emitted per unit area per unit time. [Emissivity for polished aluminium at 38 °C is 0.04.]
Solution: T = 38+273=311 K; 𝜀 = 0.04; A = 1 m2, t = 1 s
1 1
𝐶= + −1 (5.98)
𝜀1 𝜀2
If the surfaces are not parallel, the angle of vision becomes important. Distances between the two surfaces also would matter.
Considering above factors
𝑞 = 𝜎𝐴𝐹 𝑇1 4 − 𝑇2 4 (5.99)
in which A is arbitrarily chosen area of any one surface. F is a dimensionless geometric factor to account for the
• geometrics of two surfaces
• their spatial relationship, and
• surface area (A) chosen for consideration, etc.
Since consideration of all the above parameters is complex and are not normally required by food engineers, we prefer to restrict
ourselves to contemplate on Eqn. 5.97, and the importance of heat transfer to a small body from its surroundings.
𝑞 = 𝐴𝜎𝜀 𝑇1 4 − 𝑇2 4 (5.100)
in which T1 is temperature in Kelvin of the surroundings radiating heat and T2 is the temperature in Kelvin of the body receiving
heat.
❖ Let the radiative heat transfer process be
𝑞 = ℎ𝑟 𝐴 ∆𝑇 = ℎ𝑟 𝐴 𝑇1 − 𝑇2 (5.101)
where hr is the radiative heat transfer coefficient. In Eqn. 5.101, it does not matter whether T1 and T2 are in °C or K as the ∆T
remains same.
❖ Equating Eqns. 5.100 and 5.101 we get
𝑞 = ℎ𝑟 𝐴 𝑇1 − 𝑇2 = 𝐴𝜎𝜀 𝑇1 4 − 𝑇2 4 (5.102)
Now 𝑇1 4 − 𝑇2 4 = 𝑇1 2 + 𝑇2 2 𝑇1 2 − 𝑇2 2 = 𝑇1 2 + 𝑇2 2 𝑇1 + 𝑇2 𝑇1 − 𝑇2
Therefore
ℎ𝑟 = 𝜎𝜀 𝑇1 2 + 𝑇2 2 𝑇1 + 𝑇2 (5.103)
𝑇1 + 𝑇2
If 𝑇𝐴𝑣 = (5.104)
2
and if ∆T << T1 or T2, we may write with a reasonable amount of accuracy (Earle, 1969),
3
𝑇𝐴𝑣
ℎ𝑟 = 4𝜀𝜎 𝑇𝐴𝑣 3
= 0.227𝜀 (5.105)
100
Problem 5.15: A piece of meat is kept in a deep freezer maintained at – 18 °C. Calculate the radiative heat transfer if the meat is at
25 °C and has an average area of 0.045 m2. The emissivity of the meat may be taken as 0.82.
Solution: T1 = 25 + 273 = 298 K (𝑇1 4 = 7.88 × 109) ; T2 = - 18 + 273 = 255 K (𝑇1 4 = 4.22 × 109); 𝜎 = 5.67 × 10-8 W/m2 K4; 𝜀 =
0.82; A = 0.045 m2
Using Eqn. 5.100, q = 7.64 W
Using Eqn. 5.105 and 5.101, TAv = 276.5 K ;hr = 3.93 W/m2 K ;q = 7.61 W
The values of q from both the approach are reasonably comparable.
Problem 5.16: A loaf of bread is passing through a baking oven, the walls of which are maintained at a constant temperature of 220
°C. The bread has an area of 0.09 m2 and is at 100 °C. The emissivity of bread may be taken as 0.52. In addition to radiation heat,
there is convective heat also by air at 220 °C. Calculate the heat transfer rate.
Solution: Approach – Here we consider both radiative and convective heat transfer. The radiative heat is calculated by Eqn. 5.100 or
5.101, and convective heat transfer is calculated by Eqns. 5.73 and 5.52.
Given T1 = 220 + 273 = 493 K (𝑇1 4 = 5.9 × 1010); T2 = 100 + 273 = 373 K (𝑇2 4 = 1.94 × 1010) ; ∆T = 120 K; 𝜎 = 5.67 × 10-8
W/m2 K4; 𝜀 = 0.52; A = 0.09 m2.
Heat Transfer by Radiation: Using Eqn. 5.100, qradiation = 105 W; Using Eqn. 5.105 and 5.101, TAv = 433 K; hr = 9.58 W/m2 K;
q = 103.4 W
Heat Transfer by natural convection: Consider the bread to be a cube. The surface area will be 6x2 = 0.09 m2 ; x = 0.1225 m
∆𝑇 0.25
ℎ = 1.3196 × = 7.38 W/m2 K 𝑞𝑐𝑜𝑛𝑣𝑒𝑐𝑡𝑖𝑜𝑛 = ℎ𝐴 ∆𝑇 = 𝟕𝟗. 𝟕 W
𝑥
qTotal =184.7 W
Problem 5.17: In a canning process, 100 cans per minute are to be heated from 60 °C to 100 °C and demoistured of the adhering
moisture to the cans. Each can contains on average 0.3 mL of water adhering to it. The cans have an emissivity of 0.7, average area
0.047 m2 and 100 g weight. The specific heat of can material is 0.25 kJ/kg. °C. Calculate the temperature of the oven considering
radiative heat transfer only.
Solution: Approach –
a) The approach for this problem is almost similar to earlier problems using Eqn. 5.100 except that we first find heat requirement
load.
b) Heat load is determined by calculating the heat required for raising the temperature of 100 cans per minute with adhering water
from 60 °C to 100 °C and later latent heat of vaporization of water (0.3 mL) at 100 °C.
Assumption: For all practical purposes we assume the oven to be a radiating oven, and all the area of the can is totally surrounded
by radiating heat.
Given: 𝜌𝑤𝑎𝑡𝑒𝑟 = 1000 kg/m3; Cp of water = 4.2 kJ/kg °C, 𝜆𝑠𝑡𝑒𝑎𝑚 at 100 °C = 2676 – 419.06 = 2257 kJ/kg; A = 0.047 × 100 = 4.7
m2; 𝜎 = 5.67 × 10-8 W/m2 K4, T2 = 100 + 273 = 373 K.
𝑚2 𝑊
𝑊= × 2 4 × 𝐾4
𝑠 𝑚 𝐾
Heat load for cans in one minute:
𝑄𝑐𝑎𝑛𝑠 = 𝑚𝐶𝑝 ∆𝑇 = 100 × 100 × 10−3 × 0.25 × 100 − 60 = 100 kJ/min
𝑄𝑤𝑎𝑡𝑒𝑟 = 𝑚𝐶𝑝 ∆𝑇 = 100 × 0.3 × 10−6 × 1000 × 4.2 × 100 − 60 = 5.04 kJ/min
𝑄𝑠𝑡𝑒𝑎𝑚 = 𝑚𝜆𝑠 = 100 × 0.3 × 10−6 × 1000 × 2257 = 67.71 kJ/min
Total heat requirement 𝑄 per minute = 100 + 5.04 + 67.71 = 172.75 kJ
172.75
Therefore, 𝑞= = 2.87 kJ/s = 2.87 × 103 W
60
0.047×100
Now, 𝑞 = 𝐴𝜎𝜀 𝑇1 4 − 𝑇2 4 = × 5.67 × 10−8 × 0.7 × 𝑇1 4 − 3734 = 3.109 × 10−9 × 𝑇1 4 − 3734
60