Did you know that the ABRAVIBE toolbox includes a unique GUI for processing data for impact testing ? This video describes the feature, and shows how you improve the FRF quality. To use it, you simply import your data and put it in the .imptime file format, which is documented in the ABRAVIBE user’s guide, then you can open your files and process them.
This example compares the three currently available options to the frf2ptime command; ‘prony’, ‘lsce’, and ‘ptd’. It can serve as a reference for how to use these three options.
The example uses the same synthetic data for the Plexiglass plate that the Plexi Modal Example uses, but comes with the data to make it independent of the previous example.
The ABRAVIBE example ForcedResponse is illustrating the power of the digital filter forced response method for computing forced response developed by Kjell Ahlin and published in, among other references, “Noise and Vibration Analysis: Signal Analysis and Experimental Procedures,” by Anders Brandt. The method uses a modal superposition formulation using
a ramp invariant transform to formulate the forced response of each mode. Also see inside the ABRAVIBE command timefresp for references to original papers.
The example produces time data simulating a measurement of 35 displacement responses on a Plexiglass plate, when it is excited with four forces in the corners of the plate. Mode shapes and eigenfrequencies from a finite element model are used as a model for the plate.
This is an example of using ABRAVIBE for OMA. It uses synthesized data of the Plexiglas plate, and the modified Ibrahim Time Domain Method (MITD) which is virtually identical to the more known (Covariance-based) Stochastic Subspace Identification, SSI. MITD predated SSI by many years, however.
This is an example of using the Laplace transform to solve the forced response of a mechanical system. In this example the system is excited by a half-sine pulse, which is common in, for example, vibration testing. The supplied zip file contains a pdf file with the problem definition, and a separate zip file with a suggested solution. Like the harmonic SDOF forced response example, the present example is also an example I let all my mechanical vibration students solve.
The ABRAVIBE example PlexiODS is an example using synthesized data
generated from a modal model of a Plexiglas plate (incidentally, by the
example called ForcedResponse) to illustrate operating deflection shapes
(ODS) analysis in ABRAVIBE. This is an advanced example including
multiple-reference ODS to be able to separate two close modes! The
example is divided into two parts, the first a single-reference technique
and the second an extended multi-reference technique.
The ABRAVIBE example PlexiModal is an example of using the ABRAVIBE
toolbox to do experimental modal analysis of a Plexiglass plate using measured data.
The example uses data measured using impact testing, and the
polyreference time domain (PTD) method with two references, to be able to
estimate the first two modes which are strongly coupled on this plate.
The example shows all involved steps such as importing and sorting data,
curve-fitting for poles and mode shapes, and evaluating the results with
a MAC matrix and animation of the modes.
This example is good to use as a template for all modal analysis tasks
with the ABRAVIBE toolbox; just replace the data imported, and the
geometry created before animation, to use the script for your modal
This 25-page technical note presents rather comprehensive information about how modal analysis is implemented in ABRAVIBE. The first part of the note discusses some very important practical aspects of how to perform a good experimental modal analysis (EMA) test. Among the things discussed are choice of measurement points, choice of support, how to check measurement data to ensure data are of good quality, etc. These points are very important. The note states that a good EMA test is at least 80 % measurements, and only maximum 20% parameter extraction. Without good data (frequency response functions) then your EMA test is doomed!
The second part of the note presents the theory for the three methods implemented for MDOF parameter extraction, that is Prony’s method (for a single FRF), the least squares complex exponential (LSCE) method (for many FRFs but only one reference), and the polyreference time domain (PTD) method (for many FRFs and several references).
Finally, the previous example for ABRAVIBE, where the three methods are compared on real measured data, is presented with some key points.
The technical note includes a checklist which will also be published in another post at this site.
This is a one-page checklist in table format which can be used to ensure some important points are not forgotten during a modal analysis test.Continue reading “Checklist for Experimental Modal Analysis”