Testing : Wind tunnel test intro | tunnel equipment | tunnel results, [pg2], [pg3] | tunnel conclude |
Multiple fans test intro | fan equipment | fan results, [pg2], [pg3], [pg4] | fan conclude

3. TUNNEL TEST - RESULTS AND DISCUSSION
-> continued pg3

3.4 Performance under different wind tunnel speeds

The changes of maximum rotor rpm with tunnel speed are shown in Table 4 and Graph 6. These data show that the maximum tsr increased with wind speed. The exact point where the rotor rpm dropped to 60-120rpm might be higher than the last tunnel rpm of 850 as the rotor has slow respond. If the slow rotation is not included, then the rotor cut in speed is between 3.13 to 3.25m/s.

Table 4: Maximum rotor rpm at different tunnel’s fan rpm

Graph 6: Changes of maximum tsr with wind speed

Ct and Cp curves for average wind tunnel wind speeds of 4.11, 3.81, and 3.52m/s are shown in Graphs 7 and 8. In each wind speed, a total of 6 data sets were selected and plotted together.

Graph 7: Ct vs. tsr for 3 different wind speeds

Graph 8: Cp vs. tsr for 3 different wind speeds

Huge amount of data scatter present but the trend was definitely quite clear as the curves were convincingly ‘spread out’ uniformly in all direction with the increase of wind speeds.

The Cp and Ct increased with wind speeds. The highest maximum Cp during the highest wind speed was only 4% with the optimum tsr of around 2.75. This maximum value drop to about 2.5% under the lowest wind speed. As usual, the torque peaked slightly earlier than the Cp. The optimum tsr was hard to identify due to the scattering but it did seems to increase with wind speeds or at the least, remain constant.

The maximum tsr increased evidently with wind speed. The tsr increased from 2.75 at the lowest wind speed to 3.25 at the highest wind speed.

The rotor was not able to self start under all the wind speeds if self start was defined as ability to reach maximum rpm from stationary without aid. The rotor was able to reach the lower rpm around 60-120rpm but to go further assistance was needed to overcome the negative or zero torque in Figure 2. The cause of this problem might lie in the unfavourable wind condition which blew the counter wind half harder.

The increase in performance was due to increase of airfoil lift to drag ratio as discussed in fan testing. However, the trend of the curve was a bit different from the fan result and Figure 1 in Report I from Jesch and Walsh(1976) as the maximum tsr increased instead of remain constant with Re number. Nonetheless, this trend is more logical in the writer point of view.

3.5 Comparison between tunnel result and fans result

Comparisons were made between wind tunnel result and fan result. The average wind speed from wind tunnel and 3 fans were 4.11, 3.81, 3.52m/s and 4.46, 4.03, 3.66m/s respectively. The fans had higher wind speed and wider wind speed range. The Ct and Cp curves were shown in Graph 9 and 10.

Graph 9: Ct curves from both wind tunnel and fans testing

Graph 10: Cp curves from both wind tunnel and fans testing

The differences under these two testing environment are:
- Narrower operating range in tunnel. The rotor operated from tsr of 1.5 to 3.3 using fans while from 2.0 to 3.25 using tunnel.
- Lower Cp and Ct from tunnel. The fan Cp at 4.03m/s was 6% while the tunnel at 4.11m/s was only 4%.
- Optimum tsr from the tunnel seems to decrease while the fan seems to remain constant. Both these trend were not very clear.
- Maximum tsr decrease with decreasing wind speed in tunnel while the fan tsr remain quite constant. The fans environment might need bigger wind speed differences to show noticeable tsr changes.
- Not able to self start for rotor in tunnel. Might be due to higher wind speeds on counter wind half of rotor in tunnel. Also possible are the restricted wake in tunnel.
Although there are many differences, both results still agreed in the important case of performance increase with wind speed.

The individual and combined curves from fans have better shapes and less scatter than curves from tunnel. This definitely the not the situation expected before the testing. Below are the possible reasons for these scatters:
- Rotor performance might be more consistent at higher Cp.
- Turbulent wind in tunnel disturbed the rotor acceleration and wake
- Confined area in tunnel restricted wake expansion
- Components especially bearings got damaged during waiting period of 2 weeks
- Dying batteries in tachometer

The rotor under fans clearly performed better. The reason might be:
- Unconfined wake expansion in fan environment produced better performance
- Unfavourable wind condition in tunnel as the counter half was blew harder
- Incorrectly adjusted fan direction which favours rotor performance
- My subjective anemometer reading during high fan fluctuation was more biased to lower values.

These two results are definitely not near at all. The tunnel Cp is about 30% lower than fan Cp while the operating range of tsr in tunnel is also about 30% shorter than fan range. It’s hard to decide which result was nearer to the standard wind tunnel result. The main errors in fan testing will probably be the wind speed high fluctuation, low swept area of coverage and too much wake expansion while main errors in the tunnel will be the highly restricted wake expansion and the disturbance of turbulent wind. From the overall observation in these two testing, it should be reasonable to deduce that the tunnel testing was more accurate because of the continuous, constant speed, and constant direction (although curved) wind.

Last updated at November 6, 2002
Comments are welcomed