Impeller Power Number Test Stand
Team: Josh Jacobson, Dallin Miner, and Jaden Cutler
Sponsor: ThermoFisher
Project Description
ThermoFisher Scientific uses various impellers, some of which are shown below,in their single-use chambers where cell cultures are grown for various medicines. Knowing the power number of these impellers is crucial to understand what rpm to run these chambers at for the optimum mixing flow. Our test stand is capable of testing all impellers and will give the power number at a chosen rpm.
Performance Review
The table below represents some ofthe requirements given to us by ThermoFisher. The target values are the goals that we challenged ourselves to achieve and the threshold values are the bare minimum requirements. Our predicted performance values came from our CAD model and the simulations that were performed as well as theoretical power number calculations under perfect conditions. An analysis of the hydrostatic pressure was done on the polycarbonate window of the tank to ensure it would not deflect or break during turbulent flow. Simulations were also done to ensure the stainless tank would hold when filled to the top with water. Further testing was done by taking the Rushton impeller and running it at various rpm and comparing to CFD models and existing datato ensure accurate and repeatable results. Our findings are further discussed in the results section but overall our design met or exceeded all expectations.
| Requirement | Target | Threshold | Predicted Performance | Actual Performance |
|---|---|---|---|---|
| Internal size (L x W x H) | 2' x 2' x 3' | >14” x 14” x 10” | 2' x 2' x 3' | 20” x 20” x 30” |
| External size (L x W x H) | 2.5’ x 2.5’ x 4’ | < 6’ x 6’ x 6’ | 3’ x 3’ x 4.5’ | 24” x 26.5” x 57” |
| Weight (lbs.) | < 200 | < 1,000 | 143.9 | 148 lbs |
| Motor Capability (rpm) | 30–400 | 30–200 | 1050 | 1050 |
| Torque | 1500 oz-in | 500 oz-in | 1985 oz-in | 1985 oz-in |
| Budget | $6000 | < $8000 | ~$7,000 | $7,608 |
| Time to report power number | 5 min | < 20 min | 30 sec | 30 sec |
| Repeatability | +/- 2% | +/- 5% | +/- 0.2% | +/- 0.05% |
| Accuracy | +/- 15% | +/- 20% | 4.1% | 13–16% |
| Time to switch impellers | < 3 min | < 30 min | 5 min | 1:20 min |
Our Design
Our design features a rigid T-slot frame to support a Teknic motor and electronics box. The Teknic motor comes with a built in VFD to control the speed of the impeller. Teknic also makes the clearcore controller used to program the motor as well as the HMI. The HMI is shownbelow and has options to put in the diameter of the impeller and the speed you want to run it at as shown by the blue arrows on the left. The Program then will record the torque on the shaft and use the inputted values to calculate the average power number over a 10 second interval. This data can be exported to an SD card or laptop as pointed out by the blue arrow on the right. Our design also features a removable safety cover that is detected by a inductive proximity sensor so that the guard has to be in place for the machine to run.
Prox Sensor
HMI
Shaft Assembly
The shaft assembly was made of interchangeable components to aid in the switching of impellers. With an easy release lock on the torque sensor, the shaft can slip down out of the bearing and be pulled out of the tank. The threaded shaft end can then be unscrewed and the impeller will slide off the end. The Impeller adapter is 3D printed so it can be changed to incorporate any impeller that needs to be tested.
Results
The results from testing the Rushton impeller is shown in the graphs below. A few other impellers were tested as well and graphed at various rpm. The trend of each impeller stabilizes as rpm increases and flow becomes more turbulent. As shown by the graph, the power number curve flattens out at different rpm. These curves are close with existing data for these types of impellers.The graph on the left shows how repeatable the test stand is. As rpm increases, the accuracy and repeatability of the test stand increase dramatically. Overall the project was a success in determining the power number of any impeller at any rpm with accuracy and repeatability
Recommendations for future design
In the future itwould be a good idea to come up with a faster way to drain the tank as well as better wheels to move the tank once it is full. Another idea for improvements would be to minimize the components on the shaft to make it easier to align and reduce the uncertainty in any measurements.