A multitude of tests are necessary during the manufacture of sophisticated tables used to position patients in nuclear-medicine-based medical imaging machines. Some of the tests, such as jerky motion detection, are nonstandard. As a result, an economical, computer-based test system is required to meet these requirements and other manufacturer needs. The resulting test system is fully programmable, compact, and may be used to test additional table types.
Each table has two DC servo motors for the vertical and horizontal motions, numerous limit switches to keep track of the tabletop position for safety, and encoders to provide feedback about the tabletop movement. The test system had to test the motors, limit switches, and encoders, while ensuring that there was no abnormal friction during the motion that could overload the motors or result in jerks. The test system also had to actuate a clutch used for power transfer from one of the motors and measure the slip produced at the clutch.
The table is normally interfaced to the imaging machine's gantry, which houses the motor power supplies and drives as well as the table control system. In order to exercise the table through all its motions, the gantry had to be simulated using automated testing equipment. The power supplies and drives were housed in a panel along with the PC.
The friction and "sticky"; positions during motion could be determined by monitoring the motor current using the analog input channels on a motion control card, in this case the NI PCI-7344. The test was performed by moving the table horizontally or vertically at a constant speed, while monitoring the current drawn by the corresponding motor from the drive using Hall effect current transducers.
The NI UMI-7644 interface was used to bring out the terminals of the PCI-7344 for connection with the table system. The horizontal motion system (motor, encoder, and home, plus two end-limit switches) was connected to the corresponding terminals of Axis 1, and the vertical motion system to Axis 2. An additional encoder was connected to the shaft at the output of the horizontal motor clutch to the encoder of Axis 3.
The table also had additional limit switches, which were connected to the digital input/output (DIO) lines of the card. Their status was polled in NI Lab-VIEW. LabVIEW was also used to interface with custom circuitry that included a buffer to drive the relay that activated the clutch.
The primary tests exercised the motors and checked the motor current, distance between all the limit switches using the encoder output, and repeatability of these readings. Another class of tests was the measurement of the stopping distance. It was important that the table stop smoothly and within the specified distance in a highly repeatable manner when a technologist gave the stop command while positioning a patient.
This parameter was measured by getting the table to move at the specified speed and giving a motion command to stop with a specified deceleration profile and counting the number of pulses from the encoder before the table came to a standstill. In all of these tests, it was assumed that the motion controller was highly repeatable and any variation in measurements was a result of variations between the tables. (Repeatability tests on the same table verified this assumption.)
The clutch slip was measured by counting the output of the two encoders (one on the motor and one on the shaft after the clutch) and calculating the percentage difference for a full length of travel. All of these tests were performed with different speed and load conditions. For each test the pass and fail criteria had been predefined. The test results were stored and were available for post-test review.
ORIGINAL URL - /Briefs/Feb05/ALF21034.html