Gas Pressure Regulators: Why Surge Screening Depends on the Valve Operating Point

Published on 11.17

Gas Pressure Regulators: Why Surge Screening Depends on the Valve Operating Point

Abstract

Outlet pressure can respond differently to the same load disturbance when a gas regulator starts from a different operating point. Li and colleagues derived a local stability model for a direct-acting regulator and compared measured and simulated outlet-pressure records at valve openings of 0.6 mm and 4 mm. The first case oscillated, while the second settled in both types of record. The experiment shows where a linear model can help classify response and where it can misrepresent the eventual size of an oscillation. This article explains how to weigh operating conditions, model assumptions, and test traces in a technical review.

Background

An outlet pressure reading taken after a system has settled gives little information about the path it followed after a demand change. Depending on the regulator and downstream volume, a pressure disturbance can settle, oscillate briefly, or persist. Li and colleagues studied a direct-acting MBN series regulator. They treated pressure response as the product of a mechanical balance and a gas-flow feedback loop. Their test provides a way to distinguish stability near one working point from performance across an installation’s full duty range.

Research question

The authors examined whether a local model can explain why the selected regulator surges near one working point but settles near another. For a buyer examining a regulator report, this becomes a practical question about whether the recorded test conditions cover the operating point of interest. The source’s two points were run with the same reported inlet pressure and outlet setpoint. Their valve opening and the reported effective port parameter differed, which affects how the comparison should be interpreted.

Model and experimental method

The source describes the pressure sensed at the outlet reaching the diaphragm via a sensing line. Spring load and pressure forces act through the valve stem. This moves the plug relative to the seat and changes the flow into the regulated side. Figure 1 locates the parts used in the simplified model. The source assumes ideal gas behavior, uniform conditions inside each modeled chamber, negligible line resistance and outlet pressure equal to the sensed chamber pressure. Those assumptions are part of the evidence when comparing the model with a different installation.
Schematic of the seat and plug, spring, diaphragm assembly and outlet sensing line
Figure 1. English redraw of Li et al., original Figure 1, p. 19. Component positions are schematic.
The authors then linearized the force and chamber-flow equations around the equilibrium state. In source Equation (12), the forward factors G₁ = κp₂/(V₂s) and G₂ = A/(ms² + Ds + k) are linked by the port feedback term H = 2ζ₁p₁πr. In the stated notation the closed-loop relationship is Δx/Δq₃ = G₁G₂/(1 + G₁G₂H). The symbols p₁ and p₂ denote inlet and outlet pressures, V₂ the outlet-side volume, A the effective diaphragm area, m the moving mass, D damping, k spring stiffness, and ζ₁ the valve-port shape coefficient. The Laplace variable is s. The Routh condition in Equation (14) combines these terms. Its use in this paper is local because the linearization depends on the selected equilibrium.
Small-signal block diagram with forward gas-chamber and mechanical factors and a negative feedback path
Figure 2. English redraw of original Figure 2, p. 20. The source's block expressions and feedback sign are retained.
The experiment used a reported 0.25 MPa inlet pressure and 16 kPa outlet setpoint. The source describes a downstream demand step of 36 m³/h at time zero. Table 2 reports the valve openings and gives values of 89 and 39 for the parameter written as 2πrp₁ζ₁, with units carried over from the source.
Operating case
Starting valve opening
Port parameter 2πrp₁ζ₁ (m²/s, as printed)
Observed outlet response
Case 1
0.6 mm
89
Sustained oscillation over the recorded interval
Case 2
4 mm
39
Settling over the recorded interval

Results and discussion

In Case 1 the measured outlet pressure oscillated within a bounded range. The simulated oscillation grew across the plotted interval. The authors reported broadly comparable oscillation frequency and attributed the difference in the amplitude envelope to finite valve travel in the test versus the idealized model. This is a meaningful difference for a design review. A local model can flag the onset of instability without establishing peak pressure after the physical motion reaches a limit.
Approximate digitization of measured and simulated outlet pressure showing bounded oscillation versus a growing envelope
Figure 3. English redraw of original Figure 3, p. 21. The traces were digitized from the printed graph and are approximate.
In Case 2 both traces settled after the step. They followed a different early trajectory and reached slightly different terminal levels in the plotted record. The original report links stability to outlet-side volume, pressure conditions and the shape of the port. Because the tested openings correspond to different effective port parameters, the comparison does not isolate opening as a universal predictor. For another regulator, a review should identify the expected operating range and examine transient measurements at representative starting points.
Approximate digitization showing measured and simulated outlet pressure returning to nearly steady levels in the second case
Figure 4. English redraw of original Figure 4, p. 22. The traces were digitized from the printed graph and are approximate.

Conclusion

The two test cases demonstrate different dynamic outcomes in one regulator configuration. The local calculation aligns with the broad distinction between oscillation and settling. The measured trajectory remains the basis for assessing the magnitude and timing of the physical response. For a technical inquiry, provide the expected inlet and outlet pressure ranges, representative demand changes and available information on downstream volume. Request the corresponding test traces and conditions.

Limitations and outlook

The study covers two operating points on a selected regulator and one stated demand step. A cross-check on other duty points requires additional tests. A model intended to predict the bounded amplitude would need further evidence about stroke limits and other nonlinearities. The original cases are a way to read the evidence for one tested setup.

Notes

Units and the reported port parameter in the table follow the printed source. The digitized plots do not add measurements or imply a precision beyond the plotted curves.

References and Image Sources

  1. Li Haomin, Yue Ming, Yang Wenliang, Chen Shaojie and Chen Hao (2026). Linearization and Stability Simulation Analysis of Surge Phenomenon Model for Gas Pressure Regulators. City Gas, issue 5, pp. 18–22, Sections 2–5. https://doi.org/10.3969/j.issn.1671-5152.2026.05.004
  2. Figures 1–4 are English redraws of reference 1, pp. 19–22. Figures 1–2 reconstruct the depicted system and block diagram. Figures 3–4 plot values digitized from the published curves.
  3. Cover: NASA archive image of Eugene Kranz. Editorial crop and typographic layout. https://www.nasa.gov/image-article/happy-birthday-mr-kranz/
  4. NASA media usage guidance. https://www.nasa.gov/nasa-brand-center/images-and-media/