**4. Prototype Test Verification**

#### *4.1. Prototyping and Testing Methods*

Three-dimensional printing technology was used to process complex key components, such as the seeding wheel and pneumatic distribution cover, with a Stratasys F900 3D printer (Stratasys Ltd., Rehovot, Israel) and 8000 photosensitive resin, and the rest of the components were processed using metal. The prototype is shown in Figure 12a. The measuring equipment included a DN2000 intelligent anemometer (Beijing Instrument Equipment Factory, Beijing, China), as shown in Figure 12b. Its measuring range of wind speed was 1.00–365.00 <sup>m</sup> · <sup>s</sup>−<sup>1</sup> with an accuracy of 0.01 <sup>m</sup> · <sup>s</sup><sup>−</sup>1, its wind pressure range was ±80 kPa, with an accuracy is 0.01 kPa, and its air volume range was <999,999 m3 · <sup>h</sup>−<sup>1</sup> .

**Figure 12.** Test prototype and measuring equipment: (**a**) the prototype; (**b**) DN2000 intelligent anemometer; (**c**) the Pitot tube.

The prototype was fixed according to the relative position of the pneumatic distribution cover and the seeding wheel in the simulation experiment, and the airflow velocity of the airway at this position was measured by an anemometer. When the seeding wheel was processed, a measuring slot hole in an air pipe was created to facilitate the insertion of the Pitot tube. During the measurements, the front end of the Pitot tube was inserted into the hollow airway and then hot-melt adhesive was used to seal the measuring hole to ensure air tightness. In addition, it is necessary to ensure that the extended part of the Pitot tube is concentric with the pipe. The negative pressure fan was turned on, and the vacuum degree of negative pressure was 7 kPa during the test. After the working state was confirmed as stable, the value on the DN2000 intelligent anemometer was read, the airflow velocity of this group was recorded, the measurement was repeated 3 times, and the average value was recorded. After one measurement was completed, the seeding wheel was rotated to the next position and the steps above were repeated to derive the actual airway wind velocity at the same location as in the simulation.

However, because of the small diameter of the seed-sucking hole, it was difficult to use an anemometer to measure air velocity accurately at this position. It would also affect flow field distribution if it was too close to the seed-sucking hole. Therefore, the measurement position was adjusted by 20 mm. From the simulation results, it can be concluded that the airflow velocity trend at this position is consistent with the results for the seed-sucking hole.

#### *4.2. Results and Analysis*

Figure 13 shows the simulated and tested values of the airflow velocity in the air pipe.

**Figure 13.** Dot–line graph of test results.

The data in the figure were plotted after taking the average of the three groups of tests. In fact, the results of the anemometer fluctuated in a narrow range due to the uncertainty of the flow field during the measurement, and the recorded value was taken as the fluctuation center. The analysis of the readings obtained from the three groups of tests showed that there was little difference in the values measured, so it was reasonable to characterize the values using averages. The dot–line graph shows that the test results were generally lower than the simulation results. There are three main reasons for the results. First, the prototype processing could not guarantee absolute air tightness as in the numerical analysis, resulting in additional pressure loss. Second, the detection head of the anemometer inserted into the air pipe inserted an artificial obstacle, which would affect the airflow in the airway. Third, the inserted part of the Pitot tube should be concentric with the air pipe during the measurements. However, it was hard to ensure this during the actual measurements, which would have affected the measurement results.

However, both the simulation results and the test results show a trend of narrow fluctuations around the mean value, and the airflow velocity of each airway shows little difference, which can verify the accuracy of the simulation.
