Dynamic torsional stiffness testing measures the ability of a lead screw or ball screw to resist angular deflection under oscillating torque, delivering frequency-dependent stiffness data that is essential for servo tuning, feed-drive design, and precision positioning performance. This article explains the test principle, sample requirements, measurement procedures, applicable standards such as ISO 3408, JIS B 1192, and GB/T 17587, and the typical engineering applications of the results.
Technical Overview
A lead screw or ball screw is a mechanical power transmission element that converts rotary motion into linear motion, and it is widely used in CNC machine tools, precision stages, industrial robots, and electric actuators. Torsional stiffness quantifies the resistance of the screw system to angular deformation when a torque is applied, and it is defined as the ratio of the applied torque to the resulting angular deflection, expressed in N·m/rad or N·m/deg.
Static torsional stiffness is measured under slowly applied or quasi-static torque, whereas dynamic torsional stiffness is evaluated under oscillating torque excitation across a range of frequencies. Because inertia, contact nonlinearity, lubricant film behavior, and material damping all participate under dynamic conditions, the dynamic torsional stiffness of a screw assembly is frequency-dependent and is generally described by a complex quantity containing a storage (elastic) component and a loss (damping) component.
The total angular compliance of a screw drive originates from several sources: torsional deflection of the screw shaft itself, Hertzian contact deformation between rolling elements and raceways (or sliding thread contact in lead screws), deformation of the nut body, compliance of the bearing supports and couplings, and micro-slip within the loaded interfaces. Dynamic torsional stiffness testing captures the combined behavior of these elements under realistic service conditions, providing data that static testing alone cannot reveal, such as resonant frequencies, phase lag, and amplitude-dependent softening.
Testing Samples
Ball Screws
Precision ground ball screws and economical rolled ball screws, including single-nut and double-nut designs with defined preload classes, are the most common test subjects. Samples are typically supplied as complete screw-and-nut assemblies with machined end journals so that they can be clamped in the test fixture in a manner representative of the actual installation. Precision grades and acceptance characteristics are generally classified with reference to the ISO 3408 series, JIS B 1192, and GB/T 17587.
Lead Screws and Trapezoidal Screws
Sliding lead screws with trapezoidal (ACME) or similar thread profiles can be tested with the same methodology. Because these screws rely on sliding contact, their dynamic torsional behavior is more strongly influenced by lubrication condition and friction, which makes the dynamic characterization particularly valuable for life and positioning performance prediction.
Planetary Roller Screws
Planetary roller screws, used in high-load aerospace and heavy machinery applications, may also be evaluated. Their higher contact stiffness and greater load capacity generally require higher torque excitation levels and more robust fixturing, but the measurement principle remains identical.
Complete Support Units
In some programs, the screw is tested together with its bearing support units, couplings, and nut bracket so that the measured stiffness reflects the entire drive train rather than the screw alone. This approach is preferred when the data will be used directly for servo controller tuning of a complete axis.
Specimen Preparation Requirements
- Specimen length, nominal diameter, lead, and thread direction clearly documented, with end journals or center holes provided for repeatable clamping.
- Nut preload level and measured friction torque (preload torque) recorded before testing, since preload is a dominant factor in screw stiffness.
- Threads cleaned of preservative and lubricated according to the intended service condition, with the lubricant type and quantity documented.
- Surface condition free of corrosion, dents, or transport damage, and a runout check of the journals performed prior to mounting.
- Sufficient sample quantity or repeated mounting cycles to assess repeatability, typically recommended for statistically meaningful results.
Testing Items
A comprehensive dynamic torsional stiffness test program generally includes the following measured and derived quantities:
- Dynamic torsional stiffness as a function of excitation frequency, Kt(f), expressed in N·m/rad.
- Static torsional stiffness measured by incremental torque loading, used as the low-frequency baseline.
- Torsional natural frequencies of the mounted assembly and identification of dominant vibration modes.
- Damping ratio and loss factor of the torsional mode, commonly extracted by the half-power bandwidth method or by curve fitting of the frequency response function.
- Hysteresis loops from loading and unloading cycles, revealing lost motion and energy dissipation.
- Stiffness variation under different nut preload levels and different superimposed axial loads.
- Amplitude dependence of stiffness, obtained by repeating the sweep at several torque amplitudes.
- Temperature influence on stiffness and damping where operation near heat sources is expected.
- Measurement uncertainty evaluation and repeatability assessment.
Testing Methods
Fixturing and Boundary Conditions
The specimen is mounted on a rigid test bench with one end of the screw connected to the torque excitation system through a calibrated torque transducer. The nut is axially and rotationally restrained against a stiff reaction mass, and the far end of the screw is either left free or supported in a fixture that replicates the bearing arrangement of the target application. The chosen boundary condition must be reported with the results, because the same screw can exhibit markedly different dynamic stiffness when mounted fixed-free versus fixed-fixed. All fixturing is verified for stiffness well above the expected specimen stiffness to avoid corrupting the measurement.
Static Baseline Measurement
Before dynamic testing, torque is applied in incremental steps in both directions, and the resulting angular deflection is recorded. The slope of the linear portion of the torque-angle curve defines the static torsional stiffness, while the separation between loading and unloading curves quantifies hysteresis and lost motion. This baseline is used to normalize the dynamic data and to confirm the health of the assembly.
Dynamic Excitation Techniques
Dynamic characterization is performed by applying controlled oscillating torque and measuring the angular response. Common excitation approaches include:
- Sinusoidal sweep excitation over a defined frequency band, typically from a few hertz up to several hundred hertz or approximately 1 kHz depending on the application, which provides high signal-to-noise ratio at each frequency.
- Broadband random excitation using a servo-driven actuator, which shortens test time and is suitable for modal identification.
- Torsional impact excitation, in which a calibrated angular impulse is applied and the response is measured, following the mobility measurement philosophy of the ISO 7626 series for mechanical frequency response determination.
Applied torque is measured with a precision torque transducer mounted in series with the specimen. Angular response is captured with high-resolution rotary encoders, angular laser interferometry, or calibrated angular accelerometers placed at defined positions along the screw. All sensor chains are calibrated traceably before the campaign, and cross-checking with redundant sensors is recommended at critical frequencies.
Data Processing and Evaluation
The acquired time histories are transformed into the frequency domain to obtain the angular compliance frequency response function, defined as angular deflection divided by applied torque. The dynamic torsional stiffness is the reciprocal of this function and is reported as magnitude and phase versus frequency. Around each torsional resonance, a single-degree-of-freedom or multi-degree-of-freedom model is fitted to extract the natural frequency and damping ratio. Results are typically reported as stiffness values at specified reference frequencies, the minimum stiffness within the operating band, and the identified resonance characteristics, together with a statement of measurement uncertainty and the mounting condition.
Reference Standards
While the ISO 3408 series, JIS B 1192, DIN 69051, and GB/T 17587 primarily define vocabulary, acceptance conditions, and static axial rigidity for ball screws, dynamic torsional characterization is commonly performed in accredited third-party laboratories by combining these product standards with general vibration and mobility measurement practice from the ISO 7626 series. This combination ensures that results are traceable, comparable between laboratories, and suitable for use in simulation and servo engineering.
Applications
Dynamic torsional stiffness data is used throughout the life cycle of precision drive systems:
- Feed drive and servo design: engineers use the measured stiffness and resonant frequencies to set servo gains, predict the bandwidth limit of the position loop, and avoid controller excitation of structural resonances in CNC machine tools and semiconductor equipment.
- Positioning accuracy analysis: torsional deflection under dynamic torque contributes to lost motion and following error, so stiffness data supports error budgeting for precision stages and measuring machines.
- Simulation model validation: measured dynamic stiffness and damping are used to calibrate finite element and multibody models of screw-driven axes, improving the reliability of virtual prototyping and digital twins.
- Product development and comparison: manufacturers evaluate the effect of nut design, preload class, ball count, and support configuration on dynamic performance, supporting design optimization.
- Incoming inspection and supplier qualification: verified stiffness values serve as objective acceptance criteria in procurement, verified by accredited third-party laboratories when independent confirmation is required.
- Troubleshooting and condition monitoring: a shift in dynamic stiffness or damping can indicate preload loss, wear, or contamination, enabling predictive maintenance decisions before functional failure occurs.
- Emerging actuation fields: electric cylinders in robotics, exoskeletons, medical devices, and aerospace actuation systems rely on the same data to guarantee responsiveness and stability under cyclic load.
FAQ
What is the difference between static and dynamic torsional stiffness?
Static torsional stiffness is the slope of the torque-angle curve under slowly applied load, while dynamic torsional stiffness is the torque-to-deflection ratio measured under oscillating excitation at a specific frequency. The dynamic value varies with frequency, torque amplitude, and preload because inertia, contact behavior, and damping participate in the response, and it is generally the more relevant figure for servo-controlled applications.
What units are used to report the results?
Dynamic torsional stiffness is normally expressed in N·m/rad, and occasionally in N·m/deg for engineering convenience. Damping is reported as a dimensionless damping ratio or loss factor, and resonant frequencies are reported in hertz.
Which standards govern this test?
The ISO 3408 series, JIS B 1192, DIN 69051, and GB/T 17587 cover ball screw vocabulary, acceptance conditions, and static rigidity, while frequency response and mobility measurement techniques follow the general methodology of the ISO 7626 series. Test reports should always state the specific procedures and boundary conditions used.
Why does the mounting condition matter so much?
Bearing supports, couplings, and the nut bracket contribute significant compliance in series with the screw. A screw measured free-free, fixed-free, or fixed-fixed will show different stiffness and resonant frequencies, so the fixture must replicate the intended installation or the condition must be clearly documented alongside the data.
How does preload affect the measured stiffness?
Higher nut preload increases contact stiffness and therefore raises both static and dynamic torsional stiffness, at the cost of increased friction torque, heat generation, and wear. Testing at several preload levels allows designers to select the best trade-off for the application.
What frequency range is typically investigated?
Most feed-drive applications require data from a few hertz up to several hundred hertz, and high-speed or high-bandwidth systems may extend the sweep toward approximately 1 kHz. The range is selected to cover the servo crossover frequency and any expected structural resonances of the axis.
Can the same method be applied to complete axis assemblies?
Yes. Measuring the screw together with its support units, coupling, and nut bracket yields the drive train stiffness directly, which is especially useful for controller commissioning. The same laboratory equipment and data reduction methods apply, provided the fixture stiffness and sensor placement are adapted to the larger assembly.
Why Choose ZKGX?
- State-of-the-art analytical equipment
- Highly qualified scientific team
- Fast turnaround time
- Competitive pricing