Failure data of a microgripper for intracorporeal surgeries (2024)

The experimental setup is composed of two 316L stainless steel wires, two MF4005V2 brushless motors, two pulleys of 20 mm in diameter assembled to each motor, two 316L stainless steel wires of 0.08 mm in diameter, a 316L stainless steel tube of 2 mm in diameter, three supports to hold the 316L stainless steel tube, two tensioner mechanism for each 316L stainless steel wire, a wires' guide for preventing wire friction in the sharp edge of the steel tube, a mobile jaw made of 316L stainless steel, a fixed jaw made of photopolymer resin, and an opening limitation shaft of the mobile jaw of 0.40 mm in diameter.
One side of the 316L stainless steel wire (w1) is attached to the bottom of the mobile jaw, and the other side is attached to the pulley (p1) that is coupled to the shaft of the motor (M1). On the other hand, one side of the 316L stainless steel wire (w2) is attached to the top of the mobile jaw, and the other side is attached to the pulley (p2) that is coupled to the shaft of the motor (M2). The two wires are passed through a tube of 2 mm in diameter. A wire guide for preventing wire friction in the sharp edge of the 316L stainless steel tube is placed on one side of this tube. The jaw tested is the mobile jaw, and its opening range of 2 mm is limited by the shaft of 0.40 mm in diameter.
In addition, to get equal tension in the two 316L stainless steel wires during testing, two tensioner mechanisms are placed close to the motors.
The brushless motors incorporate a control unit and one encoder in the shaft of each one. The encoders are used to collect data on angular position. The angular velocity, acceleration, and jerk are generated by computing its derivatives. For velocity by deriving position with respect to time, for acceleration by deriving velocity with respect to time, and for jerk by deriving acceleration with respect to time. All are computed at the same time that the encoder collects the data. The derivatives are computed by using numerical differentiation methods.
The tests consisted of repetitive opening and closing movements controlled by torque. All tests were conducted using a positive and negative trapezoidal input torque signal for motors M1 and M2 with an amplitude of +0.00854 Nm, and −0.00854 Nm respectively, a frequency of 1 Hz and a period T = 1 second.
The tests were performed considering five scenarios:
- Scenario 1: Only the mobile jaw manufactured in 316L stainless steel was subjected to cyclical opening and closing movements, with a maximum opening range of 2 mm, and a closing range of 0 mm.
- Scenario 2: Only the mobile jaw manufactured in 316L stainless steel was subjected to cyclical opening and closing movements, changing the closing range by holding a small object of 0.39 mm.
- Scenario 3: Only the mobile jaw manufactured in 316L stainless steel was subjected to cyclical opening and closing movements, changing the closing range by holding a small object of 0.65 mm.
- Scenario 4: Only the mobile jaw manufactured in 316L stainless steel was subjected to cyclical opening and closing movements, changing the closing range by holding a small object of 0.91 mm.
- Scenario 5: Only the mobile jaw manufactured in 316L stainless steel was subjected to cyclical opening and closing movements, holding small pieces of raw chicken.
The tests were saved in CSV files and labeled T1001 to T1025, T2001 to T2025, T3001 to T3025, T4001 to 4025, and T5001 to T5025 respectively for scenarios 1, 2, 3, 4 and 5. The first digit represents the scenario, and the last digit represents the number of tests. Each CSV file contains data on variables Time, Torque_output, Torque_input1, Pos_input1, Velocity_input1, Acceleration_input1, Jerk_input1, and Temp_input1 from encoder 1; and Torque_output2, Torque_input2, Pos_input2, Velocity_input2, Acceleration_input2, Jerk_input2, and Temp_input2 from encoder 2.
A script program performed in Python "Data_Processing.ipynb" is provided for data structuration, visualization, and filtering.
Data filtering involves extracting the maximum value from the position, velocity, acceleration, and jerk data during one operational cycle. One operational cycle is composed of one opening and closing movement and encompasses 20 data points. The time was converted to cycles using the equation: number of cycles = time*frequency. After filtering, the data were gathered as T1001_max_min for each test.
The folder "Test_folder" contains raw data, and the folder "Filter_tests" contains the filtered data. The Python script "Data_Processing.ipynb" is for data filtering and visualization of filtered data.