APÊNDICE E

DCAP

A full amalgamation of these two diversified and sophisticated technologies will take some time, but major aspects of DCAP philosophy and experience will soon be seen in revisions of Abyss, and in due course new and greatly improved versions of both computational tools will become available.

DCAP:, uma ferramenta sofisticada para pesquisa em descompressão

Aqueles que lidam com descompressão no mergulho técnico já ouviram falar nas tabelas geradas pelo programa DCAP. In technical diving decompression one occasionally hears reference to tables generated by DCAP. Muito do desenvolvimento das técnicas atuais de descompressão no mergulho técnico originaram-se a partir do programa voltado para a geração de tabelas descompressivas chamado DCAP. Much of the current development of technical diving decompression techniques has originated from a general purpose computer program for calculating decompression tables known as DCAP. DCAP é a sigla para is the acronym for Decompression Computation and Analysis Program (Programa para Análise e Cálculo de Descompressão). It is not specific to identify a program as being done by DCAP, since DCAP can do all sorts of tables using all sorts of methods. To try to alleviate some of this potential confusion it might be worthwhile to describe the DCAP program in general, and to be more specific on how it has been used in technical diving.

Origin of DCAP

DCAP dates back to commercial diving in the 1960's when the search for offshore oil found commercial diving companies doing heliox diving; up to that time only navy tables were available, and they did not meet all needs for offshore diving. Dr. Heinz Schreiner led a development team at Ocean Systems, Inc., in both computation and laboratory validation of new heliox tables. This was done in collaboration with others, some of whom were then the leaders in decompression, including Bob Workman, Chris Lambertsen, Val Hempleman, and Albert Bühlmann. Schreiner's computational methods were based on Haldane's concepts, and they relied heavily on feedback from the field and other empirical information.

Following corporate changes the task was picked up by Dave Kenyon and me, and we offered decompression services to others. From this, in about 1975 the need for a number of different tables by a European client led to the idea for Hamilton Research, Ltd., to provide a program to generate decompression tables instead of just providing the tables themselves. That is, to sell the goose instead of just the eggs. The idea was to design a program that could be used by engineers and diving supervisors as well as researchers and would not need a programmer as such. This concept was accepted, and DCAP was eventually acquired by several major diving research laboratories. After we had horrendous struggles with different minicomputers the IBM PC came along and provided an adequate and ubiquitous platform. A project to develop deep air tables by one of the DCAP users, the Swedish Navy, led to a highly reliable matrix that later was found to work well for trimix diving also.

Description of the DCAP program

Thus DCAP itself is more a tool for generating and analyzing tables and profiles than it is a specific model. The User enters a page of instructions, a Basecase file, into a PC running DCAP. This describes the dive or dives to be done and gas mixes, profiles, detailed instructions to the diver, etc. The Basecase also references other files which define other variables such as units, the computational model, a matrix of ascent limits or M-values, the format for the table or tables, and names other output files to be generated such as graphics or gas loadings; a notebook file can keep a record of what has been done. These files use diving and not computer terminology, normally English, but can be translated into other languages by the user.

A number of models or algorithms are used by decompression developers, and they are still evolving; some are better than others. All of them that work rely heavily on empirical experience. Our experience is greatest with the neo-Haldanian Haldane-Workman-Schreiner model designated Tonawanda Iia.

Haldane's concept was that different parts of the body ( compartments ), take up gas at different rates, and this is limited by perfusion, the blood's ability to carry gas from the lungs to and from the tissues (rather than by diffusion around a capillary). Uptake and elimination between lung and tissue follows exponential mathematics, which means simply that the rate of transfer is proportional to the difference; these rates are defined by half times for each compartment (often called tissues but they are not anatomical), the time it takes for half the difference to be equilibrated (half the remaining difference takes another half time, and so on). Gas loadings of inert gases are measured in partial pressures. The original Haldane method provided ascent constraints as limiting ratios of partial pressures (the ratio of current depth to target depth). Because this worked best only for short, shallow air dives, Workman based ascent constraints on differential pressures. A matrix of maximum tolerable gas loadings ( M-values ) for each compartment at each depth defined the ascent limits; the gas loading calculated for each compartment is compared with the limit at each depth, and ascent to the next stop is allowed when the loadings in all compartments are less than the M-values (the loadings decay exponentially as ambient pressure is reduced and gases leave the body). Schreiner made this work efficiently for different inert gases by using different half times for each inert gas and summing the loadings in each compartment. Haldane used 6 half times, others use more, perhaps determined by the number of columns that will fit on a computer printout.

Since many decompression computations are done with Bühlmann's published method, it is relevant to compare it with Tonawanda Iia. Bühlmann uses a similar approach, except that instead of a matrix of M-values it uses factors a and b for each of the half times; these tolerated pressures can be converted algebraically to M-values. In the faster compartments (shorter half times) the inert gases are summed and compared with a calculated limit, and in longer compartments the different inert gases have different a and b values; these are divided proportionally according to the proportion of each inert gas in the mix. Both methods can vary in their conservatism, but in present practice the Tonawanda Iia model using the matrix described, identified as MM11F6, is a bit more conservative that the unmodified Bühlmann method. This combination has worked well with trimix dives.

`Thus to wrap up, DCAP is a practical result of many years of collective experience in preparing and evaluating many types of decompression tables. The DCAP concept does not limit it to calculations with any specific model there is currently a choice of several models, and others are under development. DCAP's main feature is that it facilitates the computational and table production process. It can allow different models, approaches, ascent constraints, and table configurations to be used. A good set of criteria provides a practical approach to trimix decompression in the self-contained, open circuit range with a good track record.