Dive computers and eCCRs: How accurately must the oxygen setpoint be validated?
Modern eCCRs and dive computers are technically sophisticated – yet, there are surprising gaps when it comes to verifying the decompression algorithms they use. A poster by Sofia Glashoff and colleagues from the Blekinge Institute of Technology, Sweden, at the 2026 EUBS (European Baromedical Society) Congress deals with the question of whether the existing European standards for dive computers and rebreathers are even adequate for this purpose.
The authors’ answer is quite clear: No.
The EN 13319:2000 standard for dive computers deals primarily with time and depth measurement. The EN 14143:2013 standard for rebreathers, on the other hand, defines requirements for key components of the device, such as overall functionality based on breathing resistance. What is lacking, however, are standardised procedures for verifying whether a decompression algorithm implemented in a dive computer operates correctly and delivers valid results under realistic conditions.
Small ppO₂ deviation – surprisingly large effect
One simulation carried out by the authors is particularly interesting. They used the Bühlmann ZHL-16C model with GF 70/70 and compared a planned eCCR setpoint of 1.3 bar with an actual ppO₂ of 1.2 bar. According to the EN 14143:2013 standard, such a deviation still falls within the accepted setpoint tolerance.
Nevertheless, it had a measurable influence on the calculated decompression. In the simulations using the LEMGEM model – which the US Navy uses to calculate DCS risks from dive profiles – there was a relative increase in the predicted DCS risk of:
16.2% at 30 m / 100 min with nitrox,
12.8% at 60 m / 45 min with nitrox,
23.5% at 80 m / 30 min with heliox.
The total decompression time also increased noticeably.
It is important to note the interpretation here: 23.5% does not mean that the absolute DCS risk increases by 23.5 percentage points. It refers to a relative change in the risk predicted by the model used.
Why this is of interest to CCR divers
This example highlights a fundamental problem: a dive computer may calculate its decompression based on a ppO₂ of 1.3 bar, whilst the rebreather actually only delivers 1.2 bar. Both devices may function correctly within the applicable technical specifications – and yet the decompression calculation is based on a physiological exposure that does not actually occur.
Particularly during long and deep CCR dives, such deviations – which may initially seem minor – can accumulate over the course of the entire dive. For example, we know that the Bühlmann model, with a GF pair of 50/85, indicates a DCS risk of approximately 4 per cent for a 30-minute dive to 120 metres. This would rise to 5 per cent or more simply because of the less accurate validation. A completely unacceptable level.
However, this also means that the increase in risk for shallower, shorter dives hardly matters.
Conclusion
The authors call for standardised procedures for the verification and validation of dive computers and their decompression algorithms. It is not enough for a computer to measure depth and time correctly, or for a rebreather to maintain its setpoint within the permissible tolerances. Ultimately, what matters is whether the entire system – comprising the rebreather, actual ppO₂ and the decompression computer – functions reliably as a whole.
For CCR divers, the message is simple: a set point is not necessarily the actual ppO₂ being breathed – and for the decompression calculations of deep, long dives, this difference may be more significant than one might initially assume.
Further reading on dekoblog.ch:
Kommentare