Carlos Lander
Registered
2. Verification of the NDC and configuration control PART 4
As outlined in the preceding paragraphs, the VVal-18 Thalmann Algorithm was already validated with manned diving trials under operationally relevant conditions that demonstrated acceptable PDCS. Testing of the NDC was therefore simply to verify that it was a faithful implementation of the Thalmann Algorithm. This could be done by functional testing of NDCs comparing their behavior to “gold standard” decompression schedules and these gold standards exist in two forms. The gold standard printed VVal-18 Thalmann Algorithm decompression tables are the constant 0.7 atm pO2-in-nitrogen (MK 16 MOD 0) (Thalmann, 1984) decompression tables and MK 16 Mod 1 N2-O2 decompression tables (Johnson et al.,2000) that have appeared in several revisions of the U.S. Navy Diving Manual. The gold standard software implementations are the Thalmann Algorithm Decompression Table Generation Software and the Navy Dive Planner. The latter software package is designed specifically to complement the NDCs and is convenient for generating multilevel dives and decompression schedules of any complexity against which to test the NDC. A sample of 10 to 30 of each configuration of the NDC has been functionally tested by exposing them to simulated dive profiles in a small, flooded test chamber and comparing NDC prescription to gold standard Navy Dive Planner decompression schedules (Southerland, 2000; Gault and Southerland, 2005; Gault, 2006; Southerland et al., 2010). Schedules differ by no more than can be accounted for by the specified pressure sensor tolerance (maximum ±2 fsw (0.61 msw) deviation at maximum operating depth). This type of functional testing is called “black box” testing because the tester has no access to internal data structures and computer code to guide testing. The agreement between the Cochran Undersea Technologies and the U.S. Navy does not extend to sharing such proprietary
information. The outcome of dive computer testing only remains valid while the system remains unchanged and by agreement with the manufacturer, no hardware or software changes are made to any configuration of the NDC after it has passed validation testing at
NEDU. Every NDC unit undergoes a simple functional test of pressure sensor accuracy at purchase and subsequently every 18 months.
Outline of validation of the U.S. Navy Dive Computer
Development, validation, and documentation of the Navy VVal-18 Thalmann Algorithm was a large effort. Consequently, verification of the NDC implementation of the algorithm can be a substantially smaller effort.
The principal requirement of the NDC is implementation of the U.S. Navy-approved VVal18 Thalmann Algorithm. The U.S. Navy maintains gold standard software implementations of the Thalmann Algorithm. VVal-18 Thalmann Algorithm decompression schedules produced by these gold standard implementations have acceptable PDCS as demonstrated in
manned dive trials and estimation of PDCS using probabilistic models. The NDCs are
validated by faithful replication of gold standard decompression schedules when exposed to simulated dives.3
What are the difference between Thalmann and Bühlmann?
Regarding the Cochran VVal-18 computer, one of the things that distinguishes the Thalmann algorithm from the Bühlmann algorithm is that it does not assume that the gas uptake kinetics are the same as the gas elimination kinetics; it is called the EL algorithm. You uptake a gas exponentially and if you are offgasing, you have a sufficient supersaturation that triggers a so-called linear offgasing. You turn from exponential to linear rates and wash out slower.
Does it really matter which computer I use?
Well, yes and no. It’s a matter of preference. There is no right or wrong decompression schedule. What is important is the implementation of the model, the model fundamental in relation of the profile calculations. Because it is possible to fit the wrong model fundamental to a set of data and, so produce the wrong dive schedule , as C. Gutvik describe on the workshop1.
As outlined in the preceding paragraphs, the VVal-18 Thalmann Algorithm was already validated with manned diving trials under operationally relevant conditions that demonstrated acceptable PDCS. Testing of the NDC was therefore simply to verify that it was a faithful implementation of the Thalmann Algorithm. This could be done by functional testing of NDCs comparing their behavior to “gold standard” decompression schedules and these gold standards exist in two forms. The gold standard printed VVal-18 Thalmann Algorithm decompression tables are the constant 0.7 atm pO2-in-nitrogen (MK 16 MOD 0) (Thalmann, 1984) decompression tables and MK 16 Mod 1 N2-O2 decompression tables (Johnson et al.,2000) that have appeared in several revisions of the U.S. Navy Diving Manual. The gold standard software implementations are the Thalmann Algorithm Decompression Table Generation Software and the Navy Dive Planner. The latter software package is designed specifically to complement the NDCs and is convenient for generating multilevel dives and decompression schedules of any complexity against which to test the NDC. A sample of 10 to 30 of each configuration of the NDC has been functionally tested by exposing them to simulated dive profiles in a small, flooded test chamber and comparing NDC prescription to gold standard Navy Dive Planner decompression schedules (Southerland, 2000; Gault and Southerland, 2005; Gault, 2006; Southerland et al., 2010). Schedules differ by no more than can be accounted for by the specified pressure sensor tolerance (maximum ±2 fsw (0.61 msw) deviation at maximum operating depth). This type of functional testing is called “black box” testing because the tester has no access to internal data structures and computer code to guide testing. The agreement between the Cochran Undersea Technologies and the U.S. Navy does not extend to sharing such proprietary
information. The outcome of dive computer testing only remains valid while the system remains unchanged and by agreement with the manufacturer, no hardware or software changes are made to any configuration of the NDC after it has passed validation testing at
NEDU. Every NDC unit undergoes a simple functional test of pressure sensor accuracy at purchase and subsequently every 18 months.
Outline of validation of the U.S. Navy Dive Computer
The principal requirement of the NDC is implementation of the U.S. Navy-approved VVal18 Thalmann Algorithm. The U.S. Navy maintains gold standard software implementations of the Thalmann Algorithm. VVal-18 Thalmann Algorithm decompression schedules produced by these gold standard implementations have acceptable PDCS as demonstrated in
manned dive trials and estimation of PDCS using probabilistic models. The NDCs are
validated by faithful replication of gold standard decompression schedules when exposed to simulated dives.3
What are the difference between Thalmann and Bühlmann?
Regarding the Cochran VVal-18 computer, one of the things that distinguishes the Thalmann algorithm from the Bühlmann algorithm is that it does not assume that the gas uptake kinetics are the same as the gas elimination kinetics; it is called the EL algorithm. You uptake a gas exponentially and if you are offgasing, you have a sufficient supersaturation that triggers a so-called linear offgasing. You turn from exponential to linear rates and wash out slower.
Does it really matter which computer I use?
Well, yes and no. It’s a matter of preference. There is no right or wrong decompression schedule. What is important is the implementation of the model, the model fundamental in relation of the profile calculations. Because it is possible to fit the wrong model fundamental to a set of data and, so produce the wrong dive schedule , as C. Gutvik describe on the workshop1.
where's the smiley for kicking myself? -- I actually missed that bit in all the cheering and thumping.
