2025-11-28T04:58:18.870988

Theory of a dynamic plasma flow pressure sensor

Kolesnikov, Kotelnikov, Prikhodko
The problem of reconstructing the time dependence of the dynamic pressure of a plasma jet impinging on one end of a solid rod based on the measured displacement of the opposite end has been solved. This solution allows for a reduction in the size of the dynamic pressure sensor proposed and later improved in the works [1, 2].
academic

Theory of a dynamic plasma flow pressure sensor

Basic Information

  • Paper ID: 2511.07446
  • Title: Theory of a dynamic plasma flow pressure sensor
  • Authors: Evgeny Kolesnikov, Igor Kotelnikov, Vadim Prikhodko
  • Institution: Budker Institute of Nuclear Physics, Novosibirsk State University, Russia
  • Classification: physics.plasm-ph
  • Publication Date: November 12, 2025
  • Paper Link: https://arxiv.org/abs/2511.07446

Abstract

This paper addresses the problem of reconstructing the time-dependent dynamic pressure of a plasma jet based on displacement measurements at the end of a solid rod. The solution enables significant size reduction of the dynamic pressure sensor previously proposed and improved by researchers from the Belarusian Academy of Sciences.

Research Background and Motivation

1. Research Problem

In plasma experiments conducted at the Budker Institute of Nuclear Physics using the Gas-Dynamic Trap (GDT), it is necessary to measure the pressure of plasma flow injected from a coaxial plasma accelerator (Marshall gun). This plasma flow exhibits the following characteristic parameters:

  • Density: n ~ 10¹⁵–10¹⁶ cm⁻³
  • Velocity: u ~ 10⁷ cm/s
  • Pressure: reaching several atmospheres
  • Instantaneous power density: P > 0.5 GW/m²

2. Problem Significance

Measuring accelerated plasma parameters is crucial for maintaining plasma mass balance in open-ended GDT. However, conventional measurement methods face enormous challenges due to high instantaneous power density and severe interference from pulsed currents.

3. Limitations of Existing Methods

The interferometric pressure sensor method developed by Belarusian physicists contains critical errors:

  • Incorrect Formula: Uses p(t) = cρv(t), which is the pressure-velocity relationship in a traveling wave
  • Actual Situation: The measurement end experiences superposition of incident and reflected waves, not a single traveling wave
  • Over-design: Requires rod length satisfying l ≥ cτ/2 to avoid reflected wave interference, resulting in oversized sensors

4. Research Motivation

To correct the theoretical errors in the original method and provide a correct inverse problem solution, thereby allowing the use of shorter measurement rods and enabling sensor miniaturization.

Core Contributions

  1. Correction of Theoretical Error: Points out that the formula p(t) = cρv(t) used in the original method is inapplicable to boundary conditions with wave reflection
  2. Direct Problem Solution: Rigorously solves the wave equation using Laplace transform method, obtaining exact expressions for rod-end displacement and velocity
  3. Inverse Problem Solution: Proposes the correct formula for reconstructing plasma pressure from measured end velocity: p(t)=12ρc[v(t+l/c)v(tl/c)]p(t) = \frac{1}{2}\rho c[v(t+l/c) - v(t-l/c)]
  4. Sensor Optimization: Demonstrates that accurate measurement is possible without using long rods, overturning the original method's requirement of l ≥ cτ/2
  5. Stability Analysis: Proves that the inverse problem solution is robust to measurement errors and is not an ill-posed problem

Detailed Methodology

Problem Definition

Input: Displacement x(t) or velocity v(t) at the left end of the measurement rod (far from the plasma end) Output: Pressure f(t) or p(t) acting on the right end (facing the plasma) Constraints:

  • Rod initially at rest: x(z,0) = 0, v(z,0) = 0
  • Left end free: ∂x/∂z(0,t) = 0
  • Right end loaded: kl∂x/∂z(l,t) = f(t)

Theoretical Framework

1. Wave Equation

Acoustic wave propagation in the rod follows the one-dimensional wave equation: 2xt2=c22xz2\frac{\partial^2 x}{\partial t^2} = c^2 \frac{\partial^2 x}{\partial z^2}

where acoustic velocity c=E/ρc = \sqrt{E/\rho}, E is Young's modulus, and ρ is density.

2. Laplace Transform Solution

Applying Laplace transform to the wave equation: X(s,z)=0estx(z,t)dtX(s,z) = \int_0^\infty e^{-st}x(z,t)dt

Considering boundary conditions, we obtain: X(s,z)=X(s)cosh(sz/c)X(s,z) = X(s)\cosh(sz/c)

Combining with the right-end boundary condition, we establish the relationship between force and displacement transform functions: F(s)=sX(s)sinh(sl/c)F(s) = sX(s)\sinh(sl/c)

3. Convolution Representation

Introducing the Green function Gv(s)=1/sinh(s)G_v(s) = 1/\sinh(s) (dimensionless form), velocity can be expressed as: v(t)=2j=0(t1)/2f(t12j)v(t) = 2\sum_{j=0}^{\lfloor(t-1)/2\rfloor} f(t-1-2j)

This series contains only a finite number of terms at any finite time.

Technical Innovations

1. Explicit Solution of Inverse Problem

Through difference operations, we obtain the explicit solution of the inverse problem: f(t)=12[v(t+1)v(t1)]f(t) = \frac{1}{2}[v(t+1) - v(t-1)]

or restoring dimensions: p(t)=12ρc[v(t+l/c)v(tl/c)]p(t) = \frac{1}{2}\rho c[v(t+l/c) - v(t-l/c)]

Key Corrections:

  • The original method lacks the factor 1/2
  • Requires velocity at two time instants, not just a single instant
  • The factor 1/2 originates from the superposition of incident and reflected waves

2. Physical Interpretation

Within the time interval 0 < t < 2l/c, the second term vanishes, and the formula simplifies to: f(t)=12lmcv(t+l/c)f(t) = \frac{1}{2l}mcv(t+l/c)

This appears similar to the original method in form but differs fundamentally:

  • Correctly includes the factor 1/2
  • Valid only in the early time period
  • Does not require excessively long rods to avoid reflections

3. Alternative Solution Method (Appendix A)

Using the partial wave method, the solution is expressed as two counter-propagating waves: x(z,t)=ξ+(tz)+ξ(t+z)x(z,t) = \xi_+(t-z) + \xi_-(t+z)

By tracking multiple wave reflections, we obtain results consistent with the Laplace method, verifying the theoretical correctness.

Mathematical Rigor

Characteristic Frequency Analysis

For a rod with both ends free, characteristic frequencies are: ωj=πclj,j=0,1,2,...\omega_j = \frac{\pi c}{l}j, \quad j=0,1,2,...

Corresponding standing wave modes: xj(z,t)=Ajsin(ωjt+αj)cos(qjz)x_j(z,t) = A_j\sin(\omega_j t + \alpha_j)\cos(q_j z)

Causality Discussion

Although formula (40) formally uses velocity at the "future" time t+l/c, due to wave reflection, v(t-l/c) can actually be expressed through velocities at subsequent times, thus not violating causality.

Experimental Setup

Measurement System

This is primarily a theoretical work, but based on the experimental configuration from reference 2:

Sensor Components:

  • Copper cylindrical rod (acoustic element)
  • Helium-Neon laser (λ = 632.8 nm)
  • Photomultiplier tube

Operating Principle:

  1. Polished surface at the rod's left end reflects laser back into the cavity
  2. Plasma impact on the right end generates compression waves
  3. Waves propagate to the left end causing displacement
  4. Displacement causes laser phase modulation
  5. Light intensity modulation frequency is proportional to end-face velocity

Experimental Parameters

Based on data from reference 2:

  • Rod length: l = 0.84 m
  • Acoustic velocity in copper: c ≈ 5000 m/s
  • Acoustic wave propagation time: l/c = 170 μs
  • Pressure pulse duration: τ ≈ 50-70 μs (approximately 0.1-0.2 l/c)
  • Interference signal duration: T ≈ 900 μs

Velocity Measurement

Half a laser wavelength displacement corresponds to one complete power modulation cycle: v(t)=0.5λ/Tv(t) = 0.5\lambda/T where T is the light intensity modulation period.

Experimental Results

Theoretical Verification

1. Numerical Examples

The paper constructs an exponentially decaying pulse force: f(t)=tτexp[tτ]θ(t)f(t) = \frac{t}{\tau}\exp\left[-\frac{t}{\tau}\right]\theta(t)

Its Laplace transform is: F(s)=τ(1+sτ)2F(s) = \frac{\tau}{(1+s\tau)^2}

2. Displacement and Velocity Calculations

Through Hurwitz-Lerch transcendent function representation (formulas 30-31) or more concise form (formula 42), we calculate:

  • Figure 3: Left-end displacement x(t) versus time for different cτ/l = 0.1, 0.2, 0.5, 1
  • Figures 4-5: Time dependence of left-end velocity v(t)

Key Observations:

  • Displacement grows approximately linearly over ~500 μs, corresponding to sinusoidal interference signal
  • Different pulse widths produce different response characteristics
  • Cases with cτ/l = 0.1 and 0.2 match experimental signals from reference 2

3. Pressure Reconstruction

Using the correct formula (38) to reconstruct pressure from velocity: f(t)=12[v(t+1)v(t1)]f(t) = \frac{1}{2}[v(t+1) - v(t-1)]

Figure 6 Results:

  • Successfully reconstructs the original pulse shape
  • Verifies theoretical self-consistency
  • For cτ/l = 0.1 case, reconstruction precision is limited by 15-bit arithmetic precision

Method Comparison

MethodFormulaApplicable ConditionsError Source
Original method 1,2p = cρvInfinite medium traveling waveIgnores reflected wave, lacks factor 1/2
This paper's methodp = ½ρcv(t+l/c)-v(t-l/c)Finite rod, considers reflectionCorrect
Simplified formp = ½ρcv(t+l/c)Valid only for 0<t<2l/cTime window limited

Stability Analysis

The inverse problem solution (40) exhibits good robustness to measurement errors:

  • Involves only velocity differences, not differentiation operations
  • Although converting from displacement x(t) to velocity v(t) requires numerical differentiation (ill-posed problem)
  • Converting from v(t) to pressure p(t) is a stable linear operation
  • No regularization methods needed

1. Original Methods

  • Kostukevich (2002): First proposed the concept of optical pulse pressure sensor
  • Astashynski et al. (2014): Improved sensor design using coaxial He-Ne laser configuration
    • Advantages: Single optical axis design, easy alignment, vibration resistant
    • Disadvantages: Incorrect theoretical formula, requires excessively long rod

2. Rod Vibration Theory

  • Landau & Lifshitz: Classical foundations of elasticity theory and fluid mechanics
  • Milstein (2024): Critical analysis of finite-mass spring problems, inspiring the Laplace transform method used in this paper
  • Manzhosov & Martynova (2001): Rod motion under constant pressure using multiple reflection method

3. Plasma Diagnostics

  • Ivanov & Prikhodko (2017): GDT experiment review, demonstrating importance of pressure measurement
  • Morozov (2008): Introduction to plasma dynamics, providing CPA operating principles

Innovation of This Paper

Compared to related work, this paper:

  1. First rigorously solves the inverse problem of wave equation with reflecting boundaries
  2. Corrects the fundamental error in formulas applied for years in the field
  3. Provides two independent solution methods (Laplace transform and partial wave method) for mutual verification
  4. Demonstrates that sensor miniaturization is possible

Conclusions and Discussion

Main Conclusions

  1. Core Formula: The correct pressure reconstruction formula is p(t)=12ρc[v(t+l/c)v(tl/c)]p(t) = \frac{1}{2}\rho c[v(t+l/c) - v(t-l/c)] rather than the original method's p(t) = cρv(t)
  2. Physical Meaning: The factor 1/2 originates from the simultaneous presence of incident and reflected waves at the measurement end, each contributing half the velocity
  3. Sensor Design: No need to satisfy l ≥ cτ/2 requirement; shorter rods can be used, enabling sensor miniaturization
  4. Mathematical Properties: The inverse problem solution is well-posed and robust to measurement errors

Limitations

  1. Idealized Assumptions:
    • Ignores effects of rod suspension structure
    • Assumes constant stiffness coefficient (actual density changes during deformation, ~10% error)
    • Neglects acoustic dispersion effects
  2. Measurement Precision:
    • Limited special function calculation precision in numerical computation (sharp peaks in Figure 4)
    • Numerical differentiation from displacement to velocity is itself an ill-posed problem
  3. Theoretical Scope:
    • Only considers both ends free
    • Suspended rod case requires separate treatment (promised in next article)

Future Directions

  1. Theory Extension: Consider effects of rod suspension structure, modify Green function
  2. Experimental Verification: Verify corrected theory with known pressure pulses
  3. Design Optimization: Optimize sensor dimensions and configuration based on new theory
  4. Error Analysis: Detailed analysis of error propagation through measurement chain

In-Depth Evaluation

Strengths

1. Theoretical Rigor

  • Dual Verification: Laplace transform method (main body) and partial wave method (appendix) independently solve and yield consistent results
  • Mathematical Completeness: Logic chain from wave equation, boundary conditions, initial conditions to inverse problem solution is complete
  • Physical Clarity: Clearly explains the physical origin of factor 1/2 (incident + reflected waves)

2. Practical Value

  • Error Correction: Identifies and corrects fundamental errors in published methods
  • Sensor Optimization: Allows use of shorter rods, reducing cost and complexity
  • Engineering Guidance: Provides directly applicable formula (40)

3. Methodological Innovation

  • Explicit Solution: Obtains analytical solution of inverse problem, avoiding numerical inversion
  • Stability: Proves solution stability, no regularization needed
  • Generality: Method generalizable to more complex boundary conditions

4. Writing Quality

  • Clear structure, progressing from problem statement to solution
  • Detailed but concise mathematical derivations
  • Effective figures supporting arguments

Shortcomings

1. Insufficient Experimental Verification

  • Primarily theoretical work, lacking new experimental data
  • Only qualitative comparison based on reference 2 data
  • No quantitative comparison experiments with original method

2. Limited Applicability Range

  • Only addresses ideal case of both ends free
  • Actual sensors necessarily have suspension structures requiring theory extension
  • Neglects material nonlinearity, dispersion, and other real effects

3. Room for Numerical Method Improvement

  • Special function calculation precision issues (peaks in Figure 4)
  • Could explore more efficient numerical algorithms
  • Displacement-to-velocity differentiation handling could be more refined

4. Insufficient Error Analysis

  • Claims solution is stable but lacks quantitative error bounds
  • Does not discuss specific impact of measurement noise on reconstruction
  • How does the 10% density change error propagate to pressure reconstruction?

Impact

1. Academic Contribution

  • Theory Completion: Fills gap in dynamic pressure measurement theory
  • Methodology: Demonstrates powerful application of Laplace transform in boundary value problems
  • Inspiration: Reminds researchers of importance of wave reflection effects

2. Practical Value

  • Direct Application: GDT plasma experiments can immediately adopt new formula
  • Sensor Improvement: Promotes next-generation sensor miniaturization design
  • Cost Reduction: Shorter rods mean lower manufacturing and maintenance costs

3. Reproducibility

  • Detailed mathematical derivations fully reproducible
  • Provides Mathematica code calculation examples
  • Numerical results can verify independent implementations

4. Potential Impact

  • May influence design of other pressure sensors using similar principles
  • Method generalizable to other inverse problems in wave dynamics
  • Provides new tools for plasma diagnostic technology

Applicable Scenarios

Ideal Application Scenarios

  1. High-Power Plasma Experiments:
    • GDT, tokamak, and other fusion devices
    • Plasma accelerator testing
    • Pulsed power experiments
  2. Extreme Environment Measurement:
    • High-temperature, strong electromagnetic interference environments
    • Situations where conventional sensors cannot operate
    • Non-contact measurement requirements
  3. Fast Transient Processes:
    • Microsecond-scale pressure pulses
    • Shock wave measurement
    • Explosion experiments

Limiting Conditions

  1. Rod length and pulse width must be reasonably matched (cτ/l ~ 0.1-1)
  2. Requires precise optical measurement system
  3. Demands uniform and stable rod material properties
  4. Suspension structure effects require further theoretical treatment

References

Key Citations

  1. Kostukevich (2002): Original sensor method proposal
  2. Astashynski et al. (2014): Improved sensor design and experimental results
  3. Milstein (2024): Theoretical work inspiring this paper's Laplace transform method
  4. Landau & Lifshitz: Classical textbooks on elasticity theory and fluid mechanics
  5. Ivanov & Prikhodko (2017): GDT experiment background

Mathematical Methods

  • Tikhonov & Arsenin (1977): Solution of ill-posed problems (this paper proves the problem is well-posed)
  • Kolokolov et al. (2013): Mathematical methods in physics problem collection

Summary

This is a high-quality theoretical physics paper that successfully corrects an important theoretical error in a field measurement method. The paper's main value lies in:

  1. Theoretical Contribution: Provides rigorous mathematical derivation and physical interpretation
  2. Practical Significance: Directly improves existing sensor design
  3. Methodological Demonstration: Shows application of classical mathematical physics methods to modern problems

Although lacking new experimental verification, the theoretical rigor and clarity make it an important reference in the field. The promised follow-up work (considering suspension structure) will further enhance its practical value.