Design and optimisation of linear variable differential transformers and voice coil actuators using finite element analysis: a methodical approach to enhance sensor response and actuation force

This study presents a systematic, FEMM-based design and optimization methodology for unified Linear Variable Differential Transformer (LVDT) sensors and Voice Coil (VC) actuators that enhances performance and minimizes heat dissipation under strict geometric and thermal constraints for high-precision applications like gravitational wave detectors, with results validated by experimental measurements.

The Gist

Imagine you are trying to build a super-sensitive robotic hand that can feel the tiniest touch (like a feather landing) and also push back with just the right amount of strength, all while fitting inside a tiny, delicate box.

This is exactly what the scientists in this paper did, but instead of a robotic hand, they built a high-tech device for Gravitational Wave Detectors (machines that listen to the ripples of the universe). These machines need to be incredibly quiet and precise.

Here is the story of their work, broken down into simple concepts:

1. The Problem: The "Swiss Army Knife" Dilemma

The scientists needed a device that does two jobs at once:

• Job A (The Sensor): It needs to be a LVDT (Linear Variable Differential Transformer). Think of this as a "super-ear" that listens to exactly where a part is moving without ever touching it.
• Job B (The Muscle): It needs to be a Voice Coil (VC) Actuator. Think of this as a "super-muscle" that can push or pull that same part with great precision.

Usually, engineers design these two things separately. But for these space-like experiments, they needed to cram both the "super-ear" and the "super-muscle" into one tiny package. The challenge? Making the ear hear better and the muscle push harder, without the device getting too hot or running out of space.

2. The Solution: A Digital "Playground" (Simulation)

Instead of building hundreds of expensive, heavy metal prototypes and testing them one by one (which is slow and costly), the team built a digital playground using a computer program called FEMM.

Think of this program like a video game physics engine. They could build a virtual version of their device, change the size of the wires, move the magnets, or shrink the gaps, and instantly see how it would perform.

• Analogy: Imagine you are baking a cake. Instead of baking 50 real cakes to find the perfect recipe, you use a computer to simulate the ingredients. You can instantly see, "If I add more sugar, it gets sweeter but might burn."

3. The Method: The "Tightrope Walk"

The paper introduces a step-by-step recipe to find the perfect balance. They didn't just guess; they followed a strict order, like tuning a radio to find the clearest signal.

Here is their recipe, explained with metaphors:

• Step 1: The Outer Ring (Secondary Coils): They started by arranging the outer rings of wire.
  • The Trade-off: If you put the rings too close together, the "muscle" gets stronger, but the "ear" starts hearing things that aren't there (it becomes less accurate). If you put them too far apart, the "ear" gets quiet. They had to find the "Goldilocks" distance—just right.
• Step 2: The Gap (Radial Space): They had to decide how much empty space to leave between the inner and outer parts.
  • The Trade-off: A smaller gap makes the signal stronger (better hearing), but if it's too small, the parts might rub against each other like a door jamming in a frame. They had to leave just enough room for the parts to wiggle without crashing.
• Step 3: The Height: They made the coils taller.
  • The Result: Making the coils taller was like giving the "ear" a bigger dish to catch signals and giving the "muscle" more surface area to push. It helped both jobs without causing problems.
• Step 4: The Magnet: They made the magnet inside as big as the box would allow.
  • The Result: A bigger magnet is like a stronger engine; it made the "muscle" much stronger without messing up the "ear."
• Step 5: The Wire Thickness: Finally, they played with the thickness of the copper wire.
  • The Trade-off: Thinner wire allows for more loops (more turns), which boosts the signal. But thinner wire is like a narrow pipe; it gets hot easily if you push too much electricity through it. They had to pick the wire thickness that gave the best signal without melting the device.

4. The Result: A Super-Device

After running thousands of these digital simulations, they built a real-life prototype based on their "perfect recipe."

• The "Ear" (Sensor): It became 2.8 times more sensitive. It could hear whispers it couldn't hear before.
• The "Muscle" (Actuator): It became 2.5 times stronger. It could push with much more authority.
• The Best Part: It didn't lose its accuracy. The "ear" was still perfectly straight and true, and the "muscle" was still steady.

5. Why This Matters

Before this paper, engineers were like chefs guessing recipes. This paper gave them a structured cookbook.

Now, anyone building these high-tech devices for particle accelerators, earthquake-proof buildings, or space telescopes can use this same "digital playground" method. They can save time, save money, and build devices that are stronger, more sensitive, and more reliable.

In a nutshell: The scientists used a computer to play "what-if" games with the design of a high-tech sensor/motor combo. By carefully balancing the size of the parts and the gaps between them, they created a device that is nearly three times better at its job than the old version, all while keeping it cool and compact.

Technical Summary

Here is a detailed technical summary of the paper "Design and optimisation of linear variable differential transformers and voice coil actuators using finite element analysis: a methodical approach to enhance sensor response and actuation force."

1. Problem Statement

Linear Variable Differential Transformers (LVDTs) and Voice Coil (VC) actuators are critical components in high-precision instrumentation, particularly for gravitational wave (GW) detectors like the Einstein Telescope (ET) and its R&D precursor, ETpathfinder.

• Limitations of Commercial Designs: Standard industrial LVDTs are optimized for robustness and cost, not the ultra-high vacuum (UHV), low-noise, and high-sensitivity requirements of GW detectors.
• Specific Challenges: GW detectors require a unified sensor-actuator system where the LVDT (for position sensing) and VC (for actuation) share a compact geometry. Existing design methodologies often treat these as isolated problems or rely on trial-and-error parameter studies.
• The Gap: There is a lack of a systematic, scalable framework to simultaneously optimize the trade-offs between sensor linearity/response, actuation force, thermal dissipation, and strict geometric constraints (e.g., limited radial gaps for transverse motion).

2. Methodology

The authors propose a systematic, unified simulation pipeline based on the Finite Element Method Magnetics (FEMM), implemented via a custom Python extension (pyFEMM).

A. Simulation Framework

• Modeling: The system is modeled in a 2D axisymmetric $r-z$ plane.
• Boundary Conditions: Dirichlet boundary conditions are applied to prevent magnetic flux leakage.
• Modes:
  • LVDT Mode: Primary coil excited with 10 kHz AC (20 mA); secondary coils measure differential voltage.
  • VC Mode: Secondary coils driven with DC current (1 A); force is calculated using the Maxwell stress tensor on the permanent magnet inside the primary coil.
• Key Metrics:
  1. LVDT Response: Sensitivity ($V_{out} / mm \cdot A_{in}$).
  2. LVDT Linearity: Deviation from a linear fit (target >99%).
  3. VC Force: Actuation force normalized to input current ($N/A$).
  4. VC Force Stability: Consistency of force across the displacement range.

B. The Methodical Optimization Sequence

The paper introduces a step-by-step design procedure starting from the secondary coils and moving inward, ensuring constraints are met at each stage:

1. Secondary Coil Distance ($D_s$): Optimized to balance linearity (favors larger $D_s$) and response (favors smaller $D_s$).
2. Radial Gap ($\Delta R_{space}$): Optimized by adjusting the secondary radius ($R_s$) or primary radius ($R_p$). Reducing $R_s$ improves both LVDT response and VC force without the thermal penalty of increasing $R_p$.
3. Coil Heights ($H_s, H_p$): Maximized within the mechanical envelope ($H_{env}$) to increase flux linkage and force.
4. Magnet Dimensions: Maximized within the primary coil to boost VC force (diameter has a stronger impact than length).
5. Wire Configuration: Optimizing wire diameter ($d_p, d_s$) and layers ($N_p, N_s$). Thinner wires increase turns and flux but raise resistance/heat; this requires balancing voltage-normalized response against thermal limits.

3. Key Contributions

• Unified Workflow: Unlike previous studies that optimize sensors and actuators separately, this work provides a single framework to optimize the combined LVDT+VC system, revealing how geometric changes affect both functions simultaneously.
• Trade-off Analysis: The study explicitly quantifies critical trade-offs, such as:
  • Reducing the secondary coil distance improves response but degrades linearity.
  • Increasing the primary coil radius improves response but increases heat dissipation significantly.
  • Reducing the secondary coil radius offers a "safe" improvement for both sensing and actuation without thermal penalties.
• Scalable Design Process: The methodology starts from a reference geometry (ETpathfinder Type-A) but provides a generalizable algorithm applicable to any high-precision sensor-actuator design.

4. Results

The methodology was applied to optimize an LVDT+VC assembly for the ETpathfinder seismic isolation system. A prototype was built and experimentally validated.

A. Simulation vs. Experiment

• LVDT Response: The optimized design achieved a 2.8-fold increase in sensitivity compared to the initial design.
  • Simulated: $0.0485 \pm 0.0006$ V/mm/V.
  • Measured: $0.0460 \pm 0.0009$ V/mm/V.
  • Agreement: Within 5% (dominated by systematic uncertainties in excitation voltage and mesh resolution).
• VC Force: The optimized design achieved a 2.5-fold increase in actuation force.
  • Simulated: $1.33$ N/A.
  • Measured: $1.38$ N/A.
  • Discrepancy: The measured force was slightly higher than simulations using manufacturer magnet data, likely due to variations in actual magnet coercivity, but the trend and magnitude were confirmed.
• Performance Metrics:
  • Linearity: Maintained at $>99\%$ (meeting the strict requirement).
  • Force Stability: Maintained at $>98\%$ across the $\pm 2.5$ mm range.

B. Specific Optimization Outcomes

• Secondary Radius ($R_s$): Reduced from 20 mm to 17 mm. This significantly boosted response and force while keeping linearity above the 99% threshold.
• Secondary Height ($H_s$): Increased from 10 mm to 25 mm, maximizing flux linkage.
• Secondary Distance ($D_s$): Reduced from 40 mm to 35 mm to enhance coupling.

5. Significance

• Advancement in GW Instrumentation: This work directly addresses the stringent requirements of next-generation gravitational wave detectors (like the Einstein Telescope), enabling more precise seismic isolation and control.
• Validation of Simulation-Driven Design: The close agreement between the FEMM-based simulation and physical prototype measurements validates the custom pipeline as a reliable tool for reducing development time and cost.
• General Applicability: While focused on GW detectors, the methodology is applicable to other high-precision fields, including particle accelerators and industrial robotics, where integrated sensing and actuation are required under tight spatial and thermal constraints.
• Open Source Tool: The authors have made the simulation toolkit available, fostering reproducibility and further development in the field.

In conclusion, the paper successfully demonstrates that a structured, finite-element-based optimization approach can significantly enhance the performance of combined LVDT+VC systems, achieving substantial gains in sensitivity and force without compromising linearity or stability.