MAXIMUM ANGULAR DISPLACEMENT POSSIBLE: 131.591796875 MAXIMUM ANGULAR DISPLACEMENT STEP COUNT: 26950

Data is written to a csv file identified by the date and time the collection began (in UTC). The data written is written in the following order: raw position, processed position (which adjusts the position value if it is negative), raw index, and processed index (which adjusts the index value if it is negative).

This is the first version-controlled release of the nexys calibrator motor control repository. It was used for all vibration test activities as well as the first round of the lifecycle testing for the calibrator mechanism.

In order to begin running test operations, a user must first clone this repository from gitserver, then create three additional directories at the same level as /inc and /src : /data, /bin, and /obj. The Makefile is included for ease of use. Once the directories are created, run make all from the root directory.

To run test operations, execute ./bin/motor-control-cli. The interface will open unless there is an issue with creating a serial connection to the Nexys FPGA board. This connection is hardcoded as /dev/ttyUSB0, so if there are additional connections needed, make sure the Nexys FPGA board is connected first and powered on before making any other connections. It does not appear as a serialized connection, and so will appear as the next available USB device. Adjust these connections as needed to establish connection and launch the interface.

Upon launching the interface, the uptime will automatically begin incrementing. Data is not recorded until commanded to do so, so this counter is meant to be strictly informative. To begin recording data, press 1. Before performing any other operation, a home position must be established. This is accomplished by pressing 3, which will execute a performance maneuver (rotating from hardstop to hardstop and returning to center). Once this operation is complete, press 7 to define the home position.

From here, the performance maneuver, functional maneuver (60 degrees from center in each direction, returning to center), or lifecycle test operations may be executed, as indicated by the interface.

The data is recorded in a pseudo-csv format, with raw encoder position as the first value, processed encoder position (a correction to show it as a positive value when the motor rotates CCW from the home position), raw encoder index, and processed encoder index (same correction as encoder position). Encoder position is indicative of encoder movement. Encoder index is indicative of full encoder rotations, and only increments or decrements upon completing a full rotation. The delta between encoder index records should be equal to 2048, though 2047 and 2049 are also acceptable values due to the nature of the conversion from encoder counts to degrees of rotation. All data is recorded in hexadecimal format.

In addition to the encoder data, all movement commands are inserted in the data with leading and trailing newline commands. The movement command details are listed in an effort to simplify reviewing data.

In this version, the output for all values was corrected from its original hex type to a decimal type. All additional spaces, tabs, and non-CSV-compliant formatting and outputs were also removed. The overall output format was updated to include all of the pertinent info that was provided previously by the aforementioned non-CSV-compliant entries. Output files now have the following data format:

TIMESTAMP, POSITION_RAW, INDEX_RAW, INDEX_DELTA, DIRECTION_OF_ROTATION, CHANGE_IN_POSITION, CALCULATED_STEPS, PERFORMANCE_CYCLE_COUNT, FUNCTIONAL_CYCLE_COUNT

These data identifiers are also included at the top of each file as headings. In addition to these housekeeping updates, the algorithm used to calculate the home position from the hard stops was updated to store the encoder position of each hard stop after coming in contact with it and holding for the pre-defined 1 second delay. These are then used to calculate the maximum possible movement and find half the distance, then convert it from steps into degrees by dividing the halved distance by 204.8 (as there are 2048 encoder position counts per every 10 degrees of rotation).

To protect against the possibility of missing a hard stop due to an off-nominal starting position, there has been an additional overwatch thread added which does nothing but monitor for one of two cases:

1. A manual abort signal coming from user input
2. A procedural abort signal occuring during the homing sequence With this additional thread in place, the displacement for the first two homing sequence movements have been expanded to be greater than the known maximum angular displacement for the calibrator arm in the assembly (this value is calculated by multiplying the angular displacement from hard stop to hard stop by a statically-defined amplifier - both of which are defined in the config.h file). This currently defines the movement wait time to be equivalent to the time required to move that total displacement, which means the homing sequence takes over one full minute to execute.

The overwatch thread homing sequence asserts that the encoder values will be reset prior to conducting a homing sequence. The reset must be completed prior to every homing sequence. Because of this requirement, the reset is invoked as part of the homing sequence execution to guarantee it occurs.

• update the hold times to start and stop based off of a flag rather than being statically defined for a given movement
• the algorithm to find home from hard stops currently assumes that both hard stops are contacted and does not defensively account for the possibility that one or both was not due to some obstruction or an extreme starting position that prevented the original CCW movement from reaching the CCW hard stop
• this needs to be addressed by adding an overarching watcher that preemptively cuts the existing movement command based on encoder position feedback (which will also reduce the wear on the hard stops by cutting the movement of the motor earlier than is currently explicitly defined)
• once this is in place, the original movement commands for the homing sequence must be updated to originally tell the motor to move beyond the maximum possible angular displacement, thereby guaranteeing contact with the first hard stop
• this will also reduce (or altogether eliminate) the walking of the encoder position readings noted previously