The current perfusion pump that we used did not have any indication of the flow rate when the knob is turned. We did not want to assume that this was calibrated on a linear regression (the maximum flow rate given will linearly regress as the knob is turned at a constant speed down to the minimum flow rate) so we decided to test it and adjust for any standard deviation to attain our desired flow rate.
The knob is circled in red on the left.
This test will accurately determine the flow rate of the pump at various settings, essentially calibrating the pump. To do so, we will measure the output volume over a time period and calculate the flow rate at respective pump settings. Once completed, we can set up the rest of the bioreactor system and determine the output flow rate for each trial. We can then observe accurate metabolite concentrations and cell growth information using known variables of the input media and initial cell conditions.
Flow Rate Test Protocol (For Each Setting):
Fill a bottle with water and submerge one end of a tube that leads to the pump input
Connect another tube to the pump output leading to an empty 15mL conical tube
Prime the system and allow the water level to reach 1mL, then turn off pump
Turn on the pump and start the timer
Record the time that it takes for the water to reach each marker on the tube (200uL increments)
Continue until a sufficient amount of data points have been collected (2mL or 10 data points)
Input the data into Excel and plot volume as a function of time
Use the slope to determine flow rate
Repeat at least two more times and average the values for statistical statistical confidence
Mark the pump knob with calculated flow rate value at each respective setting
Given minimum setting:
Recorded minimum flow rate: 64 ± 3 μL/min
Given maximum setting:
Recorded maximum flow rate: 422 ± 8 μL/min
The tests revealed the flow rates of the pump at various settings. The maximum flow rate was determined to be 422 ± 8 μL/min, with a minimum of 64 ± 3 μL/min. Three intermediate settings were also tested. The results from these intermediate settings were used to obtain the knob calibration regression line. We observed a linear dependence between the angle of the knob and flow rate, and constructed a flow rate meter around the knob using this data.
The results were compiled here to show that the flow rates at different degrees of turning resulted in a linear relation between μL/min and degrees.