During zero-compression fatigue testing of equine bone samples I encountered some troubles with the displacement recordings in recent samples. There is a large noise component in the displacement reading of recent tests which was not present in previous tests which were run under the same conditions. While in some samples the noise is larger during the second half of the test, there is also one sample in which the noise is largest in the middle third of the test. Actuator displacement is recorded through the inbuilt Dynacell loadcell.
I am trying to find the source of the large variations in displacement readings (still talking about displacements of less than 0.2 mm).
So far I considered the following:
a) True changes in length of the samples: unlikely. Why would a bone lengthen when it is under a compressive load?
b) External vibrations: unlikely. This would require an almost constant source of vibration from 10 pm to 8 am to explain the noise in one test and shorter periods during day time for other tests.
c) Machine related ‘miss-registration’: 1) loosening of the crosshead resulting in change of frame compliance, but there are cycles with little noise after periods of much noise. 2) Poor loadcell performance – is this the most likely option?
d) Other?
How can I differentiate one from the other? Are there some points/issues which should be considered while trying to find the source of this problem? Is the use of an external extensometer the only possibility to monitor performance of the actuator position recordings of the inbuilt loadcell? Due to the small sample size it is difficult to attach an extensometer to the sample. Would it be valid using an external extensometer between the upper and lower platen?
Thanks for any ideas!
I am trying to find the source of the large variations in displacement readings (still talking about displacements of less than 0.2 mm).
So far I considered the following:
a) True changes in length of the samples: unlikely. Why would a bone lengthen when it is under a compressive load?
b) External vibrations: unlikely. This would require an almost constant source of vibration from 10 pm to 8 am to explain the noise in one test and shorter periods during day time for other tests.
c) Machine related ‘miss-registration’: 1) loosening of the crosshead resulting in change of frame compliance, but there are cycles with little noise after periods of much noise. 2) Poor loadcell performance – is this the most likely option?
d) Other?
How can I differentiate one from the other? Are there some points/issues which should be considered while trying to find the source of this problem? Is the use of an external extensometer the only possibility to monitor performance of the actuator position recordings of the inbuilt loadcell? Due to the small sample size it is difficult to attach an extensometer to the sample. Would it be valid using an external extensometer between the upper and lower platen?
Thanks for any ideas!
Comment