Expt.

HT 310

Plate Heat Exchanger

Aim
1. To determine the overall heat transfer coefficient in a plate heat exchanger,

2. To study its scaling dependence on the hot fluid flow rate,

3. To determine the laminar turbulent transition, and

4. To suggest a correlation function for this dependence in various regimes (laminar, transition, and
turbulent).

Apparatus
The setup employed for this experiment is as follows.

1. A stainless steel plate heat exchanger with a facility to measure inlet and outlet temperature of hot
and cold fluid with an accuracy of 0.1 ◦ C. The plates are planar (not corrugated), There are a total of
10 plates making 11 chambers for the fluid transport–six for the cold fluid and five for the hot fluid.
The total heat transfer area available is equal to that of the number of plates (10). (TASK: Measure
the dimensions of the plate heat exchanger)

2. The cold fluid used here is water and the hot fluid is ethylene glycol. (TASK: Determine the depen-
dence of both the fluids’ properties–viscosity, thermal conductivity, and specific heat–with tempera-
ture in the range 40–80 ◦ C).

3. A stainless steel insulated tank with a heater to act as a reservoir for the hot fluid.

4. Hot fluid circulation pump with a speed control potentiometer.

5. Cold fluid inlet from the water supply tap.

6. Four temperature sensors at the inlet and outlet points for each of the two fluids. The hot-fluid inlet
thermometer is also a thermostat control, which controls the heater connected to the reservoir by a
simple relay mechanism.

7. Rotameters for fluid flow measurements.

HT 310-1
Procedure
General setup
1. The zero correction of the thermometers are determined by measuring steady the fluid inlet and outlet
temperature under the following conditions (without switching on the heater).

• Stationary (Assuming the equipment is at equilibrium, before the start of the experiment, all the
thermometers should indicate the same temperature. Any deviation indicates the error of the
thermometer/sensor combination)
• Allow minimal flow of the hot fluid (≈ 25 lph) and measure any temperature difference (which
is more than the above error). If the outlet temperature is greater, it indicates viscous dissipation.
• Set the pump to maximum capacity flow rate (≈ 550 lph), and measure the temperature difference
between the outlet and inlet of the hot fluid. (TASK: Calculate the Brinkman number Br =
µ u2 /K∆T , where µ is the fluid viscosity at the mean temperature, u is the velocity of the hot
fluid one chamber, and K is its thermal conductivity, and ∆T is the temperature increase. Br is a
measure of the heating due to viscous dissipation.)

2. Set the temperature of the inlet hot fluid in the dual temperature indicator cum controller. The set
point should be set around 65 to 75 ◦ C.

3. Provide cooling water supply to the plate heat exchanger so that the flowrate is between 13–15 lpm.
This will ensure that the temperature rise is restricted to about 2–3 ◦ C. Keep this flow rate constant
throughout the experiment.

4. Connect the 15 A and 5 A plug pins to a stable 230 V A.C. electric supply. Care should be taken to
connect these two pins in different phases of the power supply. Switch on the heater power supply.

5. Adjust the flow rate of hot fluid through the heat exchanger by adjusting the speed of hot fluid circu-
lation pump. Note down the flow rate of hot fluid as indicated by the rotameter. If during the course
of any experiment, the flow rate changes (due to power fluctuations, or due to temperature changes),
make minor adjustments to the potentiometer (which controls the pump speed) to manually reset the
flow rate to the desired set value. This kind of adjustments should be done for all the experiments to
follow to ensure that the flow rate is maintained at a constant value.

Determination of characteristic settling time

Adjust the set point temperature to a temperature around T = 60◦ C. Set the flow rate to an intermediate
value, V̇ = 300lph. Through out the measurement make sure the flow rate is at this value. Measure the inlet
and outlet temperatures for about 15 minutes at 30 second intervals. Use a graph sheet to plot the variation
in temperature. Use this plot to obtain an estimate of the time it takes for the inlet and outlet temperatures
to settle down to a constant value or to a constant periodic oscillation. Note down if there is any time lag in
the behaviour of the outlet temperature variation with respect to that of the inlet.
Both these readings, settling time and time lag, should be used in the main experiment: The readings
should be taken down only after the settling time (usually one or two time periods of oscillation) and the
outlet temperature after the time lag.

HT 310-2
Determination of the overall heat transfer coefficient
1. Set the inlet temperature to a high value (≈ 70◦ C).

2. Set the flow rate to the highest possible value. Note down the value and maintain it constant (see
above on how to do this).

3. Wait for the predetermined characteristic settling time, and then note down the steady inlet and outlet
temperatures of both the fluids.

4. Repeat steps from 2 (at constant temperature) for at least 10 different flow rates ranging from a maxi-
mum of about 550 lph to a minimum of about 25 lph. It is useful to place the interval in the flow rate
in a geometric progression (GP), which will give equally spaced data in a logarithmic scale. Note that
it is not possible to set the value of the flow rate to arbitrary precision as calculated by the GP. Use a
value that is closest to the resolution provided by the flow meter. For example, if the resolution of the
flow meter is 40, then approximate 28.73 to 40. For higher flow rates, you can also play around with
the value of the spacing in GP to a smaller value (say half of that in the low flow rates) so that there
are values for at least 10 flow rates.

5. In order to ensure reproducibility, for a given set of inlet temperature and flow rate, take at least three
sets of readings of the stream temperatures. The three sets can be obtained by starting (i) at the highest
flow rate and descending to the lowest, (ii) ascending to the highest, and (iii) down again to the lowest.

Theory
The plate heat exchanger normally consists of corrugated plates assembled into a frame. The hot fluid
flows in one direction in alternating chambers while the cold fluid flows in true counter-current flow in
the other alternating chambers. A schematic diagram of the flow is shown in Figure 1. The fluids are
directed into their proper chambers either by a suitable gasket or a weld depending on the type of exchanger
chosen. Traditionally, plate and frame exchangers have been used almost exclusively for liquid to liquid
heat transfer. The best example is in the dairy industry. Today, many variations of the plate technology have
proven useful in applications where a phase change occurs as well. This includes condensing duties as well
as vaporization duties. Plate heat exchangers are best known for having overall heat transfer coefficients
(U-values) in excess of 3–5 times the U-value in a shell and tube designed for the same service.
Plate heat exchanger is an attractive option when more expensive materials of construction can be employed.
The significantly higher U-value results in far less area for a given application. The higher U-values are
obtained by inducing turbulence between the plate surfaces. Owing to this they are also known to minimize
the fouling.

Heat transfer correlation

In general, the heat transfer correlation for a fluid flow past a solid surface is expressed in a dimensionless
form
Nu = Nu(Re, Pr). (1)
where Nu is the non-dimensional heat transfer coefficient Nu = h D/k. For a heat transfer in a laminar fluid
flow past a solid surface, with constant fluid properties, the steady state temperature profile is a function only
of Re, and Pr. The heat transfer coefficient is a function of the temperature profile. Therefore, the above

HT 310-3
 Hot In

Cold Out
Plates

External Wall

Hot Out

Cold In

Figure 1: Schematic diagram of a one pass counter current heat exchanger showing the flow pattern.

relationship. This expression is often used in situations where the properties vary with temperature, and for
turbulent flows.
For fully developed laminar flows (internal flows), we expect the Nusselt number Nu to be constant, however
for a developing flow its is expressed as:

Nu = C1 Reα Prβ (2)

The value of β ≈ 0.4. The value of α is found to be around 0.3 for developing laminar flow and around
0.64 for turbulent flow. The transition from laminar to turbulent region occurs between 10 / Re / 100 for
corrugated plates. It can be expected to be higher for plain plates.
The heat transfer coefficient appearing in the Nusselt number can be calculated from the overall heat transfer
coefficient U, which is given by
1 1 ∆x 1
= + + (3)
U hh Kp hc
where, hh is the hot fluid heat transfer coefficient and hc is the cold fluid heat transfer coefficient, Kp is the
thermal conductivity of the metal plate and ∆x is its thickness. Once the heat exchanger material and its
geometry are fixed, then the metal wall resistance (∆x/Kp ) becomes constant. Similarly, if the flow rate of
cold fluid is fixed and its mean temperature does not differ much for different flow rates of hot fluid, then
the resistance of the cold fluid will remain almost constant. Thus, the overall heat transfer coefficient will
depend upon the value of the hot fluid heat transfer coefficient alone. If the bulk mean temperature does not
differ much for different flow rates, then all the physical properties will remain nearly the same and Eq. (3)
can be re-written in combination with Eq. (2) as
1 1 m
= +C = α +C (4)
U hh u
where m and C are constants. hh can therefore be evaluated from the intercept of the plot of 1/U vs 1/uα .
Since the value of α is not known, it has to be estimated first. A plot of log d(1/U)/du vs log u will eliminate
the constant C and the slope will give (−α−1). The constant m can also be evaluated with this intercept. Then
a plot of 1/U vs 1/uα will provide the intercept value C, which is then used to calculate the heat transfer
coefficient from Eq. (4). The Nusselt number correlation can then be found. For the sake of simplicity, it is
often assumed that α = 1/3. This can be verified if the plot of 1/U vs 1/u1/3 is a straight line for a large
range in the small u limit.

HT 310-4
Plots
The following data need to be plotted. A sample calculation to obtain the values of the variables is shown
below.

1. Plot of the temperature transient for the hot fluid (at the outlet and the inlet) when the heater is turned
on till the temperature attains a steady state value

2. Plot of 1/U vs 1/u1/3 : In this plot, for each value of flow rate, U values corresponding to the three
independent readings should be shown, apart from the U value computed from the average of the
temperature measurement. The average should be used for fitting and computing the intercept. Since
this involves several calculations, it is suggested that programmable spreadsheets be made use of
(OpenOffice.org or MS Exel)

3. Plot of Nu/Pr0.4 vs Re

References
1. G H Hewitt, G L Shires, and T R Bott, “Process Heat Transfer”, CRC Press, NY, 1994

Observations

Height of Plate H = 10cm

Width of Plate W = 5 cm

Gap between two plates b = 1 mm

Number to plates N = 10

Number of hot fluid chambers Nh = 4

Number of cold fluid chambers Nc = 5

Zero error of hot fluid digital thermometers δT z = ◦C

HT 310-5
Observation table

Obs. No. Flow rate V̇ (lph) Hot Fluid Temperature (◦ C) Cold Fluid Temperature (◦ C)

Inlet (T 1 ) Outlet (T 2 ) Inlet (t1 ) Outlet (t2 )

10

Parameters estimation

Total heat transfer area of heat A = NHW =

exchanger

Cup mean temperature (use any T m = 21 (T 1 + T 2 ) =

typical value)

Density of Ethylene glycol at ρ =

Tm

Specific heat of Ethylene glycol Cp =

at T m

Viscosity of Ethylene glycol at µ =

Tm

Thermal conductivity of Ethy- K =

lene glycol at T m
Cp µ
Prandtl number for hot fluid Pr = =
K
2W b
Equivalent diameter De = =
W +b

HT 310-6
Sample calculation

Flow rate V̇ =
V̇
Velocity of hot fluid in a cham- u= =
W b Nh
ber

Total heat transferred Q = ρ C p V̇ (T 1 − T 2 ) =

µ u2
Brinkman number Br = =
K(T 1 − T 2 )
(T 1 − t2 ) − (T 2 − t1 )
Log mean Temperature differ- ∆T LM = =
ln [(T 1 − t2 )/(T 2 − t1 )]
ence (LMTD)
Q
Overall heat transfer coefficient U= =
A ∆T LM
De u ρ
Reynolds number Re = =
µ
Intercept of 1/U vs 1/u1/3 plot C =
1 1
Hot fluid heat transfer coeffi- = −C =
hh U
cient
h h De
Nusselt number Nu = =
K

Student information

Date of Experiment =

Batch Number =

Roll Numbers

HT 310-7