EXPERIMENT 4: -
To identify the convective heat transfer of air in natural convection mode
Table 1: Temperature of tube at different heights
Time Electric Heater Surface Temperature Tamb T surf DT Condition
(min) V A W T1 T2 T3 T4 T5 T6 T7 T8
0.00 0 0.02 0 303.5 303.8 303.4 303.3 303.6 303.2 303.5 303.7 303.47
5.00 39 0.25 9.75 305 305.5 305.5 306 306.2 305.3 304.9 303.7 305.49 2.02 Unsteady
10.00 39 0.25 9.75 306.6 307.6 307.9 308.8 308.9 307.2 306.6 303.7 307.66 2.17 Unsteady
15.00 39 0.25 9.75 308.7 309.9 310.2 311.2 311.4 309.5 308.8 303.7 309.96 2.3 Unsteady
20.00 39 0.25 9.75 310.7 311.8 312.2 313.1 313.4 311.5 310.7 303.7 311.91 1.95 Unsteady
25.00 39 0.25 9.75 312 313.2 313.5 314.8 315.1 313.2 312.3 303.7 313.44 1.53 Unsteady
30.00 39 0.25 9.75 313.5 314.8 315.1 316.5 316.8 314.8 314 303.7 315.07 1.63 Unsteady
35.00 39 0.25 9.75 314.6 316 316.5 317.8 318.1 316.2 315.3 303.7 316.36 1.29 Unsteady
40.00 39 0.25 9.75 315.6 317 317.4 318.8 319.1 317.4 316.6 303.7 317.41 1.05 Unsteady
45.00 39 0.25 9.75 316.7 318 318.4 319.8 320.2 318.3 317.6 303.7 318.43 1.02 Unsteady
50.00 39 0.25 9.75 317.5 318.7 319.1 320.6 321.1 319.2 318.5 303.7 319.24 0.81 Unsteady
55.00 39 0.25 9.75 318.3 319.5 320 321.5 322.1 320.2 319.3 303.7 320.13 0.89 Unsteady
60.00 39 0.25 9.75 318.8 320.2 320.6 322.2 322.8 320.9 320.1 303.7 320.8 0.67 Unsteady
65.00 39 0.25 9.75 319.3 320.8 321.3 322.8 323.4 321.6 320.8 303.7 321.43 0.63 Unsteady
70.00 39 0.25 9.75 319.6 321.3 321.6 323.2 323.8 322.1 321.3 303.7 321.84 0.41 Steady
75.00 39 0.25 9.75 320.2 321.9 322.2 323.9 322.9 322.7 321.8 303.7 322.23 0.39 Steady
Table 1 shows the measured surface temperatures at seven different locations (T1–T7) along the tube,
and the ambient temperature (T8). The average surface temperature was found to be 314.621K, while
the average ambient temperature was 303.7K.
Figure 1: Steady-state representation
The Figure 1 above shows the steady state condition where 0.5 is the reference line, below which we
get our steady state readings according to Equation 1
Table 2: Experimental results of heat transfer coefficient
Time(min) Qrad(W) Qconv(W) Hconv(W/m2. K)
0 -0.07 0.07 -5.08
5 0.55 9.20 86.13
10 1.23 8.52 36.07
15 1.96 7.79 20.85
20 2.59 7.16 14.60
25 3.10 6.65 11.44
30 3.65 6.10 8.99
35 4.09 5.66 7.49
40 4.45 5.30 6.48
45 4.80 4.95 5.62
50 5.09 4.66 5.03
55 5.40 4.35 4.43
60 5.64 4.11 4.02
65 5.87 3.88 3.67
70 6.02 3.73 3.45
75 6.16 3.59 3.25
From Table 2, we can see that the radiative heat transfer increased from 6.02W at 70 min to 6.16W
at 75 min while the convective heat transfer decreased from 3.73W to 3.59W correspondingly. This
reflects the temperature dependence of radiation (Equation 4). The coefficient of heat transfer at
steady state (70 min) is 3.45W/m2K. Sample Hexp value has been attached in Figure 2 below.
Figure 2: Experimental Surface Heat Transfer Coefficient
In this experiment, the theoretical heat transfer coefficient was calculated using an empirical
correlation for the Nusselt number:
where C and n are constants determined by the geometry and orientation of the heated surface.
• Grashof number (Gr): it gives us the ratio of inertial forces multiplied by buoyant forces to
that of the viscous forces. It quantifies the buoyancy forces driving air motion along the
heated tube.
• Prandtl number (Pr): It gives us the ratio of momentum diffusivity to that of thermal
diffusivity. It accounts for air properties linking thermal and velocity boundary layers.
• Nusselt number (Nu): It gives us the ratio of heat transfer due to convection to that of heat
transfer due to conduction. The main use of it is to calculate the theoretical heat transfer
coefficient.
This correlation accounts for the combined effects of buoyancy and fluid properties on natural
convection which tells us the exact role of Gr and Pr for the calculation of Nu.
Figure 3: Comparison of experimental and theoretical heat transfer coefficients
From Figure 3 on comparing the Hth and Hexp values we found out that there was an error of about
1.47% in the experimental readings. The Hth value comes out to be around 3.4 W/m2. K while the
Hexp value comes out to be around 3.45 W/m2. K.
Figure 4: Variation of heat transfer mode with time
The heat distribution between Convection and Radiation has been shown in Figure 4. Initially the
convection heat transfer showed dominance contributing to 93% to total heat transfer while the
radiative heat losses were minimal. With increasing temperature, the radiation heat transfer increased
steadily according to Stefan-Boltzmann Law, while the convective heat transfer decreased. Between
30-55 minutes, the convective heat transfer and radiative heat transfer became equal, after which the
radiative heat transfer dominated with 63% and convective heat transfer reduced to 37% of total heat
contribution. The total heat input given by the electric heater was constant at 9.75W throughout the
experiment. This tells us that convection governs at lower temperatures, but radiation becomes
significant and eventually dominant at higher surface temperatures.
Table 3: Theoretical heat transfer coefficient calculations
Time(min) gr pr*gr (pr*gr) m nu Hth
0 0 0 0 0 0
5 20556506.8 14858243.1 62.08574 36.63059 1.904791
10 45476964.7 32870750.1 75.71857 44.67396 2.323046
15 71890353.3 51962347.4 84.90283 50.09267 2.604819
20 94284313.2 68148701.6 90.85826 53.60637 2.787531
25 111854959 80848764.1 94.82402 55.94617 2.909201
30 130574012 94378896.1 98.56409 58.15281 3.023946
35 145388478 105086792 101.2481 59.7364 3.106293
40 157446764 113802521 103.2852 60.93825 3.168789
45 169160528 122269230 105.1548 62.04136 3.226151
50 178462634 128992792 106.5716 62.87723 3.269616
55 188683467 136380410 108.0657 63.75878 3.315457
60 196377802 141941875 109.151 64.39908 3.348752
65 203612774 147171313 110.1427 64.9842 3.379179
70 208321247 150574598 110.774 65.35667 3.398547
75 212800039 153811869 111.3647 65.70516 3.416668
The above Table 3 gives us Hth values which are calculated by the procedure given in Figure 3. The
average Hth comes out to be 3.011 W/m2. K while the average Hexp after steady state comes out to
be 3.35 W/m2. K.
Figure 5: Comparison of experimental and theoretical heat transfer coefficients with time
As seen in Figure 5 at the beginning of the experiment, the experimental heat transfer coefficient
(hexp) appeared very high (up to ~86 W/m²·K). This occurred because the system was in a transient
heating phase, where the temperature difference between the tube surface and the surrounding air
was changing rapidly. During this period, even small heat inputs produced large apparent values of
hexp since the thermal gradients were steep, and the system had not yet reached equilibrium.
As time progressed, the tube temperature increased more gradually, and the rate of change of surface
temperature decreased. The natural convection currents also developed and stabilized. Once the
system approached steady-state conditions, the values of hexp decayed and settled in the range of 3–4
W/m²·K, which matched closely with the theoretical prediction.
By contrast, the theoretical heat transfer coefficient (hth) remains nearly constant throughout the
experiment. This is because hth is obtained from the correlation, where Pr for air is nearly constant
throughout, and Gr depends on geometry and average properties rather than instantaneous transients.
As a result, the theoretical prediction does not account the transient variations seen experimentally but
instead represents the expected steady-state behaviour.
Sources of Error:
The following factors may have contributed to deviations from theoretical values:
• Disturbance of natural convection flow due to air currents caused by observer movement.
• Inaccuracies in temperature measurements from thermocouples or sensors, arising from
calibration issues or delayed response.
• Additional heat losses through conduction in supports, insulation gaps, or surrounding
structures that were not considered in the theoretical model.
• Uneven heating of the rod’s surface leading to non-uniform temperature distribution.
• Variations in ambient conditions, such as fluctuations in room temperature or unintended
airflow/ventilation.
Improvements:
• Ensure the ambient temperature thermocouple is securely positioned, as it was observed
during the experiment that accidental contact with the setup surface led to abnormal ambient
temperature readings.
• Perform proper calibration of temperature sensors prior to the experiment to minimize
measurement errors.
Conclusion
The experiment successfully determined the heat transfer coefficient, 3.45W/m2. K, in a vertical tube
of brass. Steady state was achieved at 70 minutes, with an average surface temperature of 314.621K
and an ambient temperature of 303.7K. Analysis of modes of heat transfer (Figure 4) showed that at
lower temperature, convection heat transfer contributed to 94% of the total heat transfer while the
radiation heat transfer was only 0.55 W. But at 75 minutes when the surface temperature was
322.23K, the radiation heat transfer dominated with 63% contribution to the total heat transfer. The
total heat input remained constant at 9.75W throughout the experiment.
These results support the theoretical expectation that convection dominates at lower temperatures. The
comparison between experimental and theoretical heat transfer coefficients (Figure 5) shows a clear
deviation during the initial heating stage, where the experimental values were much higher due to
transient thermal effects and measurement uncertainties. As time progressed, the experimental
coefficient gradually decreased and approached the theoretical values, indicating the stabilization of
the thermal boundary layer.