Time Domain Analysis
Short duration disturbances are often at the heart of a system problem. We call these short duration events transients. They may involve the system ringdown to a motor startup, or a robot move. The transient may be a velocity ripple that shows up as a disturbance in part of an inspection scan, or the slipping and sliding of a product due to high-G bumps in automated handling. These events must be measured and characterized principally in the time domain to make headway in understanding the problem at hand.
Transient disturbances present more of a challenge than do disturbances that are steady in time. Short duration disturbances are often not well characterized in the frequency domain using FFT analysis techniques because the details of the wave form are what is most important. These details are lost in the magnitude of the FFT. Transients are also often hidden under layers of system vibration and instrumentation noise.
Time Domain Accelerations Supporting Samples During Automated Moves
Yet transient disturbances often have patterns to them that are important. For example, the timing of horizontal to vertical accelerations are critical to evaluating when a part may overcome friction and slip.
Horizontal vs Vertical Acceleration Plot to Estimate When Slip May Occur
The time domain waveform often holds a lot of information that can be used to form a hypothesis for a vibration problem. Below is a picture of a waveform measured on a part that was failing and impacting its supporting frame. At the start of the project our thinking was that the steady operational vibration was suspect. However, we found that, due to a very soft isolation system, the part appeared to impact the frame resulting in very high levels of inertial force. Fatigue often occurs with either many cycles of stress just above the SN curve, or a smaller number of cycles well above the SN curve. Our testing showed that the latter case was most probable and a new mounting system was proposed.
Acceleration Measurement of Non-Linear Impacts During Product Testing to Diagnose Failure Mode
Understanding the instrumentation and the effective time-constants of the instrumentation response is also very important, as is the awareness of non-linear behaviors that include amplitude and frequency dependence, motion history dependence, direction and time dependence.
We have developed techniques to get past these challenges and get at the heart of the problem. To do this well requires a full understanding of the physics of the sensors and the system under test. Give us a call and we can discuss your unique situation in more detail.