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 pg2

3.2 Rotor rpm, torque, and power curves

Initially, the rotor was blew from stationary state. It did start but it never got pass the rpm of 60 to 120rpm (around 1 to 2 revolution per second) even at the highest possible tunnel speed of 3.07m/s. The rotor just rotated slowly around these speeds. However, when the rotor rotation was assisted to around 300rpm by shielding off the counter wind half of swept area, the rotor started to accelerate by itself to maximum rpm of above 450 and stayed there. This behaviour shows that the torque value around the centre range of the rotor tsr which were normally very low dipped to below zero this time, as shown in Figure 2. This behaviour was also observed during the fan testing using 3 fans at the lowest fan speed. The first starting range of positive torque was weak and slow so it’s quite useless while the second range of positive torque was strong and fast so it’s the productive area. No attempt was made to measure the first phase of positive torque.

Figure 2: Performance curves with negative values between starting and acceleration

The changes of rotor rpm with time are shown in Table 3 and Graph 1. When the rotor was released from rope tension at slightly below 300rpm in the first few seconds, the rotor hesitated to accelerate. Then, it began to increase very fast before slowing down at the maximum rpm. Small fluctuation occurred continuously after the maximum.

Table 3: One of the collected rpm data

Graph 1: Rpm vs. time for one data set from wind tunnel test under fan rpm of 1200 or average wind speed of 4.11m/s.

Torque and power were calculated from Eq. 2 and 3 in Report I and shown in Graph 2 and 3. The torque at the starting of curve (above 300rpm) was quite high so rpm lower than 300 might be possible. But the slope was quite steep so the starting point might not be a really bad decision after all. These specific curves show that the power peaked after torque. The curve shape is not very nice but considered good when compared to other data.

Graph 2: Torque vs. rpm for wind tunnel test with fan rpm of 1200

Graph 3: Power vs. rpm for wind tunnel test with fan rpm of 1200

3.3 Errors in testing

The total error calculation for wind tunnel testing are similar to fan testing (Calculation 1 in Report I) with changes in wind speed and angular acceleration as shown in Calculation 2 below.

Calculation 2: Percentage of error in Cp and Ct

The error percentage above did not include some not measured and immeasurable errors. Both tunnel and fan result contained similar errors except the quality of wind:
- Confined space might restricted the normal wake expansion and increased the wind speed. No wind tunnel correction factor was made to compensate this.
- Effects of high turbulent and huge obstruction (me) on one side of wall
- Slightly curved wind speed

9 out of 27 collected data sets were discarded due to bad shapes and extreme values which occurred especially at lower wind speeds. Graphs 4 and 5 show some of the good and bad shapes among the included sets (6) at the highest wind speed.

Graph 4: 2 best Cp curves out of 6 from wind tunnel test with fan rpm of 1200

Graph 5: 2 worst Cp curves out of 6 from wind tunnel test with fan rpm of 1200

Last updated at November 6, 2002
Comments are welcomed