# Breathin’ on the edge – the hidden complexity of pilot breathing regulators

**Authors:** Karoliina Messo, Heikki Mansikka

PMC · DOI: 10.3389/fphys.2026.1745844 · Frontiers in Physiology · 2026-02-17

## TL;DR

This study examines how a breathing regulator in military aircraft performs at low pressures, revealing issues that could affect pilot breathing and mission performance.

## Contribution

The study provides new empirical insights into regulator performance at low inlet pressures, which were previously not well understood.

## Key findings

- Regulator flow delivery continued during early expiration at low inlet pressures, causing elevated mask pressures.
- User expiratory load increased due to a flow termination failure in the regulator at low pressures.
- High-resolution monitoring is needed to ensure life support systems remain under pilot control and reduce respiratory strain.

## Abstract

The life support system of high-performance military aircraft is designed to protect aircrew from all adverse respiratory conditions. Many of its critical functions depend on an on-demand regulator, required to deliver breathing gas flow solely in response to pilot’s respiratory demands. However, the performance of the regulator when operated at the lower bound of, or outside, its specified operational range is not well known, understood, or fully characterized. To address this gap, we examined the performance of a CRU-103A/P safety pressure on-demand regulator when connected to a Gentex 5400 flight mask and supplied by a GGU-12/A on-board oxygen generating system–a life support system configuration employed in a variety of aircraft platforms. Regulator inlet pressures were controlled within and below the specified operating range (5–120 psig), specifically at 10, 6, 4, and 2 psig, producing a range of resting inspiratory flow demands in the participant. A dedicated, in-house–developed measurement system was used to capture high-resolution mask and regulator outlet pressures, as well as inspiratory and expiratory flows. Our results showed that an increase in peak inspiratory flow demand sufficient to produce mask pressures approximately 1 mbar or more below typical resting minimum values occurred at regulator inlet pressures below 10 psig. This led to continued regulator flow delivery during the early phase of expiration, resulting in elevated regulator outlet and mask pressures. This demonstrated that at regulator inlet pressures near or below the lower limit of the operational range, high inspiratory flow demand delayed regulator closure at the onset of expiration. Consequently, a brief period of continued breathing gas delivery occurred during early expiration. Further, the findings indicated that the user’s expiratory load increased at the onset of expiration due to a user-triggered flow termination failure in the regulator. This occurred as expiratory pressure propagated through the open inhalation valve to the regulator outlet and further to the rear of the mask exhalation valve, adding additional backpressure required to open it. The present study highlights the need for accurate, high-resolution monitoring of life support system’s performance. Such monitoring system would ensure that the life support system remains under the pilot’s control and imposes minimal respiratory loading. This is of vital importance for reducing dyspnea and allowing the pilot to allocate cognitive resources to mission-focused tasks, while also mitigating hypo- or hypercapnia that could compromise physiological normoxia.

## Full-text entities

- **Diseases:** Dyspnea (MESH:D004417), PPB (MESH:D018467), AGSM (MESH:D013180), hypercapnia (MESH:D006935), compromised respiratory state (MESH:D012131), hypoxic (MESH:D002534), hypoxia (MESH:D000860), ILS (MESH:D054082), hypocapnia (MESH:D016857), adversely affect cognitive function (MESH:D003072), respiratory or cardiovascular conditions (MESH:D018376)
- **Chemicals:** oxygen (MESH:D010100), CRU-103A/P (-), argon (MESH:D001128)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12953129/full.md

## Figures

1 figure with captions in the complete paper: https://tomesphere.com/paper/PMC12953129/full.md

## References

19 references — full list in the complete paper: https://tomesphere.com/paper/PMC12953129/full.md

---
Source: https://tomesphere.com/paper/PMC12953129