Mechanical Dynamics Workbench
Mechanical Dynamics Workbench
Exploring signals, learning dynamics, and understanding product response
By Joseph P. McFadden Sr. | Updated 11 October 2026
Joe McFadden here. I would like to introduce Mechanical Dynamics Workbench, formerly MDSP. I developed this tool to help people explore signals and connect them with physical questions. Students and curious learners can begin without being analysts or bringing field or simulation data: the Workbench can generate signals internally. Engineers can use the same tools to ask what a product experiences, whether qualification testing represents it, and how a component might respond.
The name has changed because the work has grown. Signal processing remains essential, but the tool also supports dynamic response analysis, simulation, environmental comparisons, and learning through experiments. The Workbench is a place to generate or load a signal, examine its meaning, try a model, and preserve the reasoning behind your conclusion.
The updated distribution supplies Mechanical Dynamics Workbench version 7.6.0 as two standalone Windows applications. It includes eight illustrated guides with actual 7.6.0 screenshots, a document reference, and a release guide in Word and PDF, plus launch instructions and checksums. The guides use the current executable names and short screen descriptions to support practical learning.
The engineering questions behind the plots
Imagine that your product has passed its qualification program. It is now installed on a forklift, and a customer reports intermittent failures after operation on rough routes. An engineer returns with a recorder file. There may be hours of ordinary vibration, short impacts at dock plates, machinery tones, and long periods of idle time. Those are different exposures with different implications. A single average or the largest acceleration in the file cannot explain all of them.
The Workbench helps you inspect that record, identify useful intervals, and choose analyses suited to the question. You can compare suitable vibration intervals with a qualification reference, investigate complete shock events separately, and retain the source and processing information needed for a review.
Now consider the analyst asked whether a component will survive. The measured acceleration is an input to that investigation. The component also has natural frequencies, damping, mounting conditions, clearances, and potential failure mechanisms. The Workbench provides tools for exploring the connection between excitation and response. A survival assessment then requires an applicable limit and evidence that the model represents the component well enough for the decision.
This is part of what I mean by combatting engineering mind blindness: learning to see the assumptions behind an apparently convincing answer. When we poke a product, energy moves through a physical system. The signal provides evidence about that interaction. Understanding it means looking beyond the height of a peak to ask what moved, what resonated, and what the measurement actually represents.
Two ways into the Workbench
Actual version 7.6.0 application captures. Displayed records are generated teaching examples.
Intent interface. Follow THE POKE, THE SYSTEM, and THE RESPONSE, with access to the Learning Center, Coach, Challenge, and Journal.
Comprehensive interface. Reach loading, processing, system definition, and analysis directly. Session information and statistics keep the selected record in view.
What you can do at the Workbench
Inspect and organize measured data
The tool supports common tabular data, enDAQ and SlamStick recordings, and workflows for Dewesoft recordings and Abaqus history output. The guides explain the relevant import routes. Confirming the quantity, units, time base, and channel selection is part of the analysis. An acceleration column expressed in millimeters per second squared must not be treated as though its numbers were already in g.
Time histories, Signal Advisory, Signal Survey, and Segment Overview help you examine long records and select events or operating intervals. The Signal Library keeps named signals available for comparison and activation. A derived signal can be retained alongside its source, with notes explaining what changed. Saving an entry and making it active are separate steps; the active signal supplies subsequent analyses.
Work with vibration and shock
Power spectral density, or PSD, describes how mean-square signal content is distributed across frequency. The Workbench includes Welch PSD analysis, spectrograms, PSD specification tools, and signal synthesis from a PSD specification. These support comparisons and controlled experiments, provided the selected interval and analysis settings suit the question.
Vibration response spectra let you compare field and qualification environments through the responses of idealized oscillators. Shock response spectra provide a corresponding way to investigate transient excitation. Rainflow and fatigue damage spectrum tools support fatigue investigations under stated model and exposure assumptions. Each view contributes different evidence; overall vibration level, shock severity, and lifetime exposure are not interchangeable quantities.
Filter with a stated purpose
Filter Lab lets you explore filter families, cutoff, order, and phase treatment while comparing the original and processed signals. Cutoff Explorer examines how selected metrics change across candidate cutoffs. In this version, Apply Selected Cutoff opens the filter process window, where you can review the design, name the result, add notes, and create a new signal. A smoother trace is a visual change. Whether it improves the engineering analysis requires a response comparison and a defensible reason for filtering.
Explore a system and learn from it
Single-degree-of-freedom and multi-degree-of-freedom models let you explore system response. Change a natural frequency or damping assumption and examine the consequences. Generate a sine wave, random vibration, or shock pulse inside the tool, or synthesize an input from a PSD specification. You can control the experiment without collecting measurements or importing simulation results.
The Intent interface organizes the work around THE POKE, THE SYSTEM, and THE RESPONSE. Its Learning Center, Coach, Challenge, and Journal encourage students and curious newcomers to predict, experiment, and explain what happened. Internal filtering, aliasing, and signal-cleaning labs make processing choices visible. Learning is a purpose in itself; you do not need a professional analysis assignment to begin.
Representative analysis tools
Actual version 7.6.0 application captures. Displayed records are generated teaching examples.
Filter Lab. Explore filter family, cutoff, order, and phase while comparing the original signal with the filtered preview.
Power spectral density. Three tones appear at 10, 25, and 50 Hz in this internally generated example. The surrounding skirts are spectral leakage from the finite record and window, not added random noise. This view teaches frequency content; it is not a qualification spectrum.
What is included in the download
The distribution ZIP was updated on 11 October 2026 and retains its original filename. Its 24 files include two 7.6.0 executables, eight illustrated guides, the document reference, and the release guide in Word and PDF, plus a README and checksum manifest. Both applications are standalone; no Python installation is needed.
MechanicalDynamicsWorkbench_Intent_v7.6.0.exe opens the guided experience. Students and curious learners can generate a signal, predict its response, and explore the Learning Center, internal labs, simulation, and PSD synthesis without supplying their own data.
MechanicalDynamicsWorkbench_Full_v7.6.0.exe opens the Comprehensive interface with broader processing and analysis controls. The Guided Tour also offers a route to the Comprehensive GUI. Version 7.6.0 adds stricter paired-signal timing checks for shock transmissibility and updates the Learning Center.
Eight illustrated guides and a document reference
Quick Start Guide. Begin with a generated signal or load a first record, confirm units, inspect the signal, and choose an analysis. Students, curious learners, and first-time users can start here without an external dataset.
Intent Guided Tour User Manual. Explore the guided interface, input, system, and response, plus the Learning Center and internal labs. This is the natural next step for students, instructors, and anyone learning through controlled experiments.
Comprehensive GUI User Manual. Look up the broader processing and analysis controls, Signal Library, spectra, fatigue, PSD modeling and synthesis, and reporting. This detailed reference serves experienced users and learners ready to explore further.
SlamStick Forklift Field Survey Walkthrough. Follow a field recording through import, survey, segmentation, and separate vibration and shock investigations. It helps field engineers handle recordings containing several operating conditions.
VRS Field vs Qualification Walkthrough. Compare qualification vibration with measured field exposure through modeled response. Reliability and test engineers can examine frequency coverage, damping assumptions, and exposure duration.
PRD Field to Spec Guide. Connect measured evidence to a proposed test specification. Product, systems, and quality teams can identify the assumptions, supporting records, and reviews needed before approving a requirement.
Cutoff Explorer Filter Selection Walkthrough. Examine a cutoff sweep, interpret its metrics, and apply a selected cutoff through the filter process window. Review the design, name the new signal, and record why the choice is appropriate.
Abaqus Explicit AutoProcess Walkthrough. Prepare simulation histories with attention to units, irregular time steps, resampling, rate reduction, and filtering. It helps analysts retain the basis for processing and comparison decisions.
Document Reference and Reading Guide. Find a summary of each guide, who benefits, and suggested reading paths, including learning without your own data. Use Mechanical_Dynamics_Workbench_Document_Reference_Distribution.pdf or its Word counterpart. It uses filenames without local folder references.
Getting started with a useful question
Extract the ZIP on your Windows computer. README_START_HERE.txt explains first launch and checksums. Read Mechanical_Dynamics_Workbench_760_Release_Guide.pdf for the version changes, and use the document reference to choose a learning path. Check Help/About for 7.6.0. All eight illustrated manuals use v7.6 filenames and actual 7.6.0 screenshots; Word and PDF copies are included.
Download Mechanical Dynamics Workbench
[Insert your Dropbox download link here]
Distribution file: MD_workbench_for_distribution_10Oct2026.zip
If you are a student or simply curious, start in the Intent application with an internally generated signal. No analyst background, field measurements, or imported simulation data is needed to begin. Predict what a change in damping or filter cutoff will do, change one setting, and compare the result with your prediction. Use the labs to explore the difference.
If you already have field data, begin with the measurement context. Where was the recorder mounted? Which axis corresponds to the qualification direction? What was the product doing? Are there signs of clipping, missing data, or a bandwidth limitation? Keep that information with the record, then use the field survey and VRS guides to organize the investigation.
For simulation output, begin with the Abaqus walkthrough and verify the model's unit system and output timing. A plot that looks plausible can still have an incorrect scale or an unsuitable time base. Preserve the original output and inspect the processing record before using the prepared signal for comparison with a test.
What makes the result useful
My aim is to help you produce an explanation that another engineer can examine. Record the source, selected interval, units, active signal, processing choices, and model assumptions. Retain the relevant plots and state what they support. A measured response above a qualification reference identifies a comparison to investigate; it does not, by itself, establish a particular failure mechanism.
Duration and representativeness also matter. A short recording cannot automatically stand for a product's service life, and equal overall vibration levels can hide differences near a component's resonance. Fatigue calculations require exposure and model assumptions. A component survival conclusion needs an appropriate response-to-limit comparison. The Workbench helps you investigate these questions while keeping the basis of the answer visible.
Use the guides as working companions. Bring a question, generate a signal, and try an experiment. If you have measurements or simulation output, return to them with a clearer sense of what to investigate. If you do not, keep learning with the signals and models inside the Workbench.
If you have a question, a what-if, or a test result that does not match a simulation, I would like to hear about it. You can reach me at mcfadden@snet.net or visit McFaddenCAE.com.
Stay curious. Look for the why. And follow the energy.
Have a wonderful and thoughtful day.
