An automated, low complex end to end resistance measurement system for multi-channel slip-rings

摘要

Contact resistance is the key factor influencing the performance of a slip-ring, which seriously affects the reliability of the signal conduction [1]. End to end resistance of a slip-ring is contact resistance plus interior cable resistances in the slip-ring. This paper describes the electrical model (low and mid frequency) and implementation of automated end to end resistance measurement system for 158-channel slip-ring intended for low speed signal conduction. Automated test is performed by using PXI-based dynamic signal analyzer, LXI-based switching unit, a turntable, two power supplies (one for driving the turntable, the other for applying test current through the contact under test) and test cables. Slip-ring is mounted on a turntable that will turn at its nominal speed during the tests. Loopback cables that make all contacts shorted are connected to the rotating part of the slip-ring. During the tests these cables can rotate together with the rotating part of the slip-ring continuously enabling a fully-isolated structure between rotating and fixed parts. All the other test devices are connected to the fixed side of the slip-ring. This kind of isolated measurement structure enables continuous and unsupervised test scenarios such as automated end-to-end resistance tests and life tests since cable twisting or damages are not problems for this test system anymore. On the other hand, test cable resistances must be measured before or after the tests in the same test environment conditions to be subtracted from the total resistance values measured during the tests to find the aimed end to end resistance values. It is also possible to perform this operation automatically for all channels after only one manual connector replacement operation. Although the end to end resistance values of all channels cannot be measured in one shot due to the multiplexed nature of the proposed test system, using the same measurement devices in the loop will enhance the measurement uncer- ainty compared to the distributed parallel hardware architectures. In addition, performing the measurements in a conditioned test environment will further reduce the variations of the measurements taken from the system at different time instances. Although contact noise results in electrical consequences, the generation nature of it is mechanical. The contact noise can be estimated from an AC signal generated by the change in dynamic contact resistance. In order to estimate the contact noise for the frequency band of interest, fast acquisition of the dynamic contact resistance is performed by the proposed test system. The estimated noise voltage signal is proportional to the change of contact resistance when small currents are applied, namely less than 100 milliamps. For optimal contact noise estimation results, DC test current of 100 milliamps is applied to each contact under test.