JOMO KENYATTA UNIVERSITY

OF
AGRICULTURE AND TECHNOLOGY

HEAT TRANSFER EXPERIMENTS

September 1994

By Mr. B.K .Kariuki

Lecture department of mechanical Engineering

1.Objective

To know heat transfer coefficient of single tube in forced and natural

convection and also heat transfer enhancement due to turbulence in tube
bundle.

2. Introduction

In the forced convection, in general, heat transfer is explained by the

relation between Nusselt and Reynolds number.
In the natural convection it is by Nusselt Grashoff and Prandtl numbers. It
is well known that the heat transfer coefficient is enhanced by turbulent
flow.
3.0.CROSS FLOW EXPERIMENTS

3.1. General experimental details and data.

The cross flow apparatus supplied enables experimental work to be

undertaken, on forced convection heat transfer from bundle of tubes
mounted transversely to the forced air flow. Heat transfer coefficients
corresponding to differing air flow rates, different tube positions within
the bundle and different tube diameters can be determined. A heated
tube consisting of a cartridge type electrical heater with an internally
mounted thermocouple can be used on its own or located in any bank
of the tube bundle. Thus the heat flow under varying conditions can be
determined. Additionally, natural convection studies relating to a single
tube may be undertaken.
The setting up procedure is to mount the black acrylic test section at
the inlet of the bench mounted fan, using the toggle catches provided.
The height of the test section mounting feet should be adjusted as
required. Note that the fan outlet flow straightener is not essential for
these particular experiments and may be omitted if desired.
Thermocouples are mounted at the inlet to the flow tube and on the
cartridge type electrical heater. They should be connected to the
temperature read out facility on the control unit. Which also contains
the regulated power source supplying the heater?
Level and zero the manometer to be used, and with each limb in the
vertical position, connect the smaller manometer to the pressure
tapping points on the cross flow apparatus, i.e. the points located
immediately before and immediately after the test section. The
upstream pressure tapping should be connected to the manometer
reservoir and the downstream tapping to the manometer limb. Position
the pitot static tube in the upstream position on the test section with
the downstream position blanked off and connects the pitot static tube
to the larger manometer with side outlet, static pressure tapping
connected to the manometer and the top vertical tapping to the
reservoir.
In all experiments involving the control unit, care should be taken to
allow the equipment to stabilize by switching on the unit, with the
heater supply turned to minimum, for at least 30 minutes prior to
taking any results.

Data for P3219 Cross flow heat exchanger.

Voltmeter: 0- 30 volts RMS dc – digital panel meter
Ammeter: 0 -4 amps RMS dc – digital panel meter
Working section: 140mm x 140mm
Tube bundle: pitch (12.7mm dia. tubes) = 20mm ( 1.6x d)
Blank tube: length 142mm diameter 12.7 mm
Heated tube: length 142mm diameter 9.5mm
Length 142mm diameter 12.7 mm
(Each heated tube contains an embedded type K
thermocouple)
Temperature readout: digital panel meter with switch selection to
read Ta (inlet air) or Th –Ta (heater/air differential) ⁰C
Heater trip: factory set to heater temperature of 80⁰C
3.2.To determine the relationship between Nusselt’s and
Reynolds’ numbers applied to a single heated tube positioned
transversely to a stream of air.( Forced convection study

Preliminary. Position the larger (12.7mm diameter) heated tube in the

clear acrylic resin housing provided, and mount this housing in the black
acrylic resin working section, using the brass knurled head screws to
clamp it into position . Couple supply and thermocouple leads from the
heated tube to the heater supply turned to a minimum value, and allow
the control unit to stabilize in temperature before commencing the
experiment. This stabilization period should normally be of the order 30
mins so that meaningful results can be obtained.
Method. With the bench supply switched on, start the benched-
mounted fan with the flow control valve fully closed. When the fan has
run up to full speed, fully open the flow control valve to give a
maximum reading on the pitot static tube manometer, with the pitot
head positioned approximately in the centre of the duct.
Adjust the input voltage to the heater tube to arrive at steady
temperature difference between the tube surface (measured by
embedded thermocouple) and inlet ambient air ( Th – Ta), of around
40⁰C.
Record the values of Th – Ta, heater voltage and current, and pitot static
tube reading. Repeat investigation for a wide range of pitot static tube
readings, i.e. from outlet air control valve fully open to minimum
measurable flow. In every case maintain Th – Ta at 40⁰C. Repeat the
above experiment using the 9.5mm diameter heated tube.
NOTE: ‘Teflon brushes’ will be required here for mounting purposes.
Allow heated tube to cool down before attempting to remove it, since
thermal expansion will cause the tube to initially lock in position.
Any permanent offset difference between the temperature of the tube
and the air temperature reading within the working section should be
subtracted from the temperature difference reading.

Results and calculations (See notes on calculation)

1. Using the equation V =√ (2Δp/ρ1) m/s , convert the pitot tube
reading of pressure into values o fair velocity within the working
section .
Note that ΔP = mmH2O x 9.807 pascals.
2. Determine the Reynolds number associated with each of the above
velocities from the equation
Re = d1vρ1/μF
using constant units to give dimensionless number

ρ1 =1.293 Pa/ Po x To/Ta kg/m3

Where, Pa : atmospheric pressure in Nairobi (=870mmbar)
Pa : atmospheric pressure at sea level (=1013mmbar)
Ta : duct inlet air temperature [K]
To : 0⁰C (=273)
NOTE: - Values of viscosity, μF used in the above equation are
determined at the film temperature, TFILM
3. Knowing the dc voltage and current supplied to the heater in the
question, the watts dissipated Q can b calculated and by determining
the tube surface area A, (∏d1l) a value for q/A power dissipation in
watts/m2 can be calculated.
Since q/A = h ( TSURFACE – TFLUID) and TSURFACE – TFLUID
has been determined in the experiment, a value for h (heat transfer
coefficient) can be calculated in watts per m2 - ⁰C.
4. Calculate the Nusselt numbers from Nu =hd1/KF
NOTE: values of thermal conductivity KF are those determined at the
film temperature, TFILM
 5. Plot Nusselt numbers (Nu) against Reynolds numbers (Re) on log-log
graph paper. From the plot determine B and n in the equation Nu =
Ren and compare the experimental values with published data.

i.e.

Re B n
0.4 to 4 0.891 0.33
4 to 40 0.821 0.385
40 to 4000 0.615 0.466
4000 to 40,000 0.174 0.618
40,000 to 400,000 0.0239 0.805
Typical results for section 3.2.

Relationship between Nusselt’s and Reynolds numbers for a single

heated tube positioned transversely to a steam of air.
Results obtained – 12.7mm diameter tube (d)

Inlet ΔT Heater Heater

Temp Ta voltage current
Pitot reading Pv T h - Ta
⁰C volts Amps.
⁰C

mmH2O Pa

24.2 237.4 21.0 42 15.95 1.91

11.6 113.8 21.0 43 14.75 1.77

6.8 66.7 21.0 42 13.90 1.70

4.0 39.2 21.0 41 12.76 1.54

 2.0 19.6 21.0 41 11.98 1.40

0.65 6.4 21.0 43 10.45 1.26

Calculations from the above results

Free Re (note 3) Power (q) HO w per KF w per m- Nu (note 6)

velocity v watts(note4) m2-K K
m/s (note (note5)
2)
19.9 15,670 30.5 128 2.758 x 10-2 58.9
13.8 10,870 26.1 110 (from 50.6
10.5 8,270 23.6 99.2 tables)n 45.7
8.1 6,380 19.7 82.8 38.1
5.7 4,490 16.8 70.6 32.5
3.3 2,600 13.2 55 25.3

Notes on calculations –Single heated tube (calculation below are made under the
conditions of sea level Pa = 1013mmbar, Ta =16 ⁰C

1. ΔP = mmH2O x 9.807 pascals

2. Free velocity V =1.291 (ΔP)1/2 m/s
3. Fluid properties
Inlet Temp = Ta + 273k =21 +273 = 294K
Surface Temp = (Th –Ta) + Ta +273K =42 + 21+ 273K
Film Temp = (TINLET + TSURFACE)/2
= ½ { 294 + [42 + 21 + 273]}
I.e. TFILM = 315 K
d of tube = 12.7mm
ρI at inlet tube = 1.2022 kg/m3
μF at TFILM =19.391 x 10-6 Ns per m2
KF =2.758 x 10-2 w per m-K at film temp. (315k) from tables
4. Re = vdρI /μF =vx {12.7 x 10-3} x 1.2022 x {106/19.391}
= 787.4 v (v in m/s)
5. Power q = ( RMS dc volts x RMS dc amps) watts
6. q/A =hO ( TSURFACE –TFLUID) = hOΔT watts/m2
A = ∏dl of tube = ∏12.7 x 142 = 5666mm@
7. Nusselt number Nu = hOd/KF
d =12.7mm
= hO(12.7 x 102)/ 103 2.758
= hO 0.4605
Results
A graph of Nusselts number versus Reynolds number was drawn on log-log
graph paper and points joined by a straight line whose equation was:
Nu = 0.614 Re0.474
Two points were inserted on the graph from published data and reasonable
correlation was found.
i.e. at Re = 10,000 difference 12.7%
Re = 2,000 difference 2.2%
Nusselt number verses Reynolds number experiment 3.2
Conclusion
The wide range of results that include figures derived from tubes of
different diameters aptly demonstrate the effectiveness of dimensionless
number groups (Re and Nu) to correlate a wide range of experimental data.
 3.3. To determine the heat transfer coefficient for heated tube within
tube bundle.
Preliminary. Position the clear acrylic resin housing containing the bank of
12.7mm diameter tubes in the black acrylic resin working section.
Remove the desired blank tube from the third bank in the bundle and
replace with the heater tube.
Method. Close the fan flow control valve and start the fan. When the fan Is
running, open the flow control valve to obtain the maximum reading on the
pitot static tube when it is positioned in the upstream position. Switch on
the supply at the control box, and adjust the input voltage to the heater
tube to give a temperature difference between the embedded
thermocouple and inlet air (Th – Ta) of approx. 40⁰C. Record Th – Ta, voltage
supplied to the heater, current flowing in the heater, and pitot static tube
reading. Repeat the above procedure for a range of flow velocities,
maintaining Th – Ta at approx. 40⁰C for each velocity investigated.
Results and calculations. The calculations detailed for experiment 3.2 are
also applicable to the test on heated tube within a tube bank. The cross
sectional area used for the derivation of velocity across the tubes is now
given by; cross sectional area = (a –n.d1) x b (duct dimensions a x b) m2
which is the cross-sectional area of the duct less the plan area of the tubes
in each tube bank. The velocity derived from the results given by a pitot
static tube positioned upstream of the tube bundle will obviously have to
be increased by multiplying by the ratio of cross sectional area of the duct
at the pitot static tube plane, divided by the cross-sectional area used for
flow of air through the tube bundle.
i.e. VT = V x p/(p-d) m/s
Where p =tube pitch. For this model p = 1.6d
Therefore VT = V x (1.6/0.6) m/s

A pitot is again made of Nusselt versus Reynolds numbers using log-log paper. The
plot should be compared to that obtained for single heated tubes, to show the
effect of increased turbulence, of the air passing over the tube, on the heat
transfer rate.
3.4. To determine the Effect of change of position of Heated tube within
the Tube Bundle.
Preliminary. As air flows through a tube bundle, its turbulence will increase
progressively, depending on the number of tubes within the bundle. The
effect of this increase of turbulence on heat transfer rate can be
demonstrated by taking a series of results, as in the previous experiment,
for the heated tube in each of the four available positions.
Method. The previous experiment (3.3) is repeated for a fixed value of air
speed, with the heated tube positioned in the first tube bank and then in
each of the other tube banks in turn, the relevant results being taken in
each case.
Results and calculations. The effect of increasing turbulence on heat
transfer coefficient, as the tested tube is moved from left to right within the
tube bundle, is evident from the derived results. There is no increase in
heat transfer coefficient between banks 3 and 4 because bank 4 is the final
bank, and air is free to follow in a parallel path from the middle of bank 4
and is not forced around the back of the tubes as in the previous three
banks.

Table 1.

N 1 2 3 4 5 6 7 8
HN/h͚ 0.63 0.76 0.93 0.98 0.99 1.0 1.0 1.0
Hm/h͚ 0.63 0.70 0.77 0.83 0.86 0.88 0.90 0.91
In table 1, hN/h is the ratio of heat-transfer coefficient for any row N to that
for a row where h no longer change and hm is the row-mean coefficient for
rows 1 through N.
3.5. Natural convection study.
Preliminary. The fan should be switched off for the experiment, since the
movement of air being studied is the around a single cylinder by natural
convection only. The single tube test module, with the 12.7mm diameter
tube in position, is fitted into the duct, and coupled to the control unit.
Method. Switch on the heater, adjusting the output until Th –Ta settles at
around 20⁰C. Record values of Th –Ta heater current, and heater voltage.
Repeat the procedure increasing Th –Ta in 10⁰C increments up to the
maximum permitted by the heater trip ( th = 80⁰C).
Results and calculations. A simplified expression for natural convection heat
transfer from a horizontal cylinder to air is given by the expression :-
H = A(ΔT/L)b watts/m2-K
For values of product of Grasshof and Prandtl numbers lying between 10 3
and 109,
Since A =m 10.97, b= ¼ and L =d
Then h = 10.97(ΔT/d)1/4 watts/m2 –K
The accuracy of this relationship can be determined by plotting a series of
values of h against ΔT using log- log paper. The slope of the best straight
line through the results will give the value of the power to which ΔT is
raised, whilst the value of the abscissae (h) for which the ordinate (ΔT) is
zero will give log A.
Alternative procedure. The experiment can be repeated using the 9.5mm
diameter tube to confirm the effect of diameter. Also the experiment could
be performed with the test module supported outside of the duct in a
position so as to allow a free vertical airflow past the horizontal heated
tube.
OPERATING INSTRUCTIONS.

EXPERIMENT 1.
Demonstrate that heat flow is directly proportional to temperature difference
between faces.

1. The apparatus is assembled with the full length (102mm) aluminum sample
in position. Ensure that the ends of the sample are free from dirt and apply
a thin smear of contacting fluid to enable efficient thermal contact to be
made. Suitable contacting fluids are silicone grease or heat sink compound.
2. Hold the clamping lever positioned on the front of the apparatus in the
downward position and place the specimen between the heater and
calorimeter block, and clamp in position by releasing the lever.
3. Insert the thermocouples into the holes provided in the sample. A small
amount of thermal Grease should be used on each thermocouple.
4. Place the Dewar vessel over the specimens.
5. Fit the thermometers into the special leak proof connections provided on
top of the calorimeter base (left hand water out, right hand water in), and
connect the water pipes from the water supply to the header tank, the
header tank to the inlet on apparatus, the apparatus outlet connection to
drain via the spring valve provided and the header tank overflow to drain.
6. Turn on the water supply and adjust the flow at source to give a small
regular overflow from the constant head tank to drain. Adjust the height
tank and the clip on the outlet hose to obtain a water flow through the
apparatus of 1.0 to 3.0cc? Sec, whilst maintaining the overflow. During the
experiment, if necessary, re-adjust the clip on the outlet hose to prevent
the difference in temperature between the two mercury –in-glass
thermometers from exceeding 12⁰C, whilst maintaining the small overflow
to drain.
 7. Check the supply voltage as indicated on the serial number label positioned
on the back of the apparatus is correct. Connect the apparatus to a single
phase AC supply point using the socket provided in the right hand side of
the apparatus.
8. Switch on the unit.
9. The heat delivered to the sample is controlled by regulating the current
supplied to the heater block using the controlled knob positioned on the
front panel under the ammeter. Turn the knob fully clockwise so that the
maximum current is supplied to the heater until temperature T 4, as
indicated by the thermocouple selection knob on the front panel
approaches 80⁰C. turn the heater control knob until temperature T 4
stabilizes at approximately 80⁰C and maintain this temperature until each
of the three other thermocouples are reading constant temperature.
10.The following readings should be recorded.
W1 = water inlet temperature ⁰C
W2 = water outlet temperature ⁰C
T1-4 = thermocouple temperature ⁰C
M = mass of water collected Kg.
T = time to collect M kg of water, secs.

The flow rate of water is determined by timing the collection of a 100ccs sample
of water using a stopwatch (provided).

11.Progressively increase the heat supplied to the sample so that temperature

T4 increases at increments of about 40⁰C up to a maximum value of about
200⁰C and allows the temperatures to stabilize for each progression.
Record values of W1, W2, T1-4, M and T, each time.
12.Switch of the apparatus.

Interpretation of results. The value of heat supplied to the calorimeter

from the heater is given by: - Q = J x M/t x (W2-W1) watts.
Where J = conversion factor of heat (4816 joules/Kilocal)
A graph of heat transmitted through the sample Q, by a temperature force
(T4 – T1) will be straight line with origin at zero, demonstrating that heat
flow is directly proportional to the temperature difference between the
faces – fig 2.1
It will also be observed that for each value of T4, the temperature
differences T4- T3, T3 –T2 and T2 – T1 are the same, (the distances between
the thermocouples also being equal). In other words the temperature
gradient along the sample can be determined by the extrapolation of the
temperature versus length curve fig 2.2.