Mailing List Archive Mailing List Archive
>Posted on 7 Feb 1996 at 03:01:41 by Dan Volker Dan, would it be possible to forward article #2 to this account ? I never saw it. >3. Some manufacturers are touting nonredundant (three oxygen sensors >do not qualify as redundancy because they are part and parcel of a single >averaging circuit) electronically controlled rebreathers as being safer than > nonelectronic units. Again, hogwash. I'm sure that each and every person >reading this has had a radio, computer, electric clock and or an >electronically controlled appliance fail on him. When it happens in a dry >environment, it's merely an inconvenience. When it happens underwater, it >can very easily become life threatening. This is incorrect analysis. If you look at a competently designed system there is not apriori reason to believe that the electronics as a rule will be the limiting factor as far as reliability is concerned. The examples given have no bearing on the discussion at hand. >Any electronic or electrical failure, including sensor or circuit >readouts, is more likely to occur underwater because of the number >of bulkhead penetrations required to make the transition from a wet >environment to the dry one required for the electronics and/or displays >(see attached drawing 50-004). Electronics do not require bulkhead througholes. You can transfer both power and signals though permanent walls be using magnetic, optic, acoustic etc. means. There is no reason to think electronic throughholes to be more troublesome than mechanical througholes in a proper implementation. >..underwater camera case. Anyone who has used one for any >period of time knows that the case will sooner or later leak through a >strobe or flash wire penetrator, a control shaft penetrator or past >the case lid seal. These are *mechanical* seals failing on you. A good design would anticipate this to happen. This has nothing to do with electronics per se. >Electronic control circuitry and displays don't >work well when wet, and the requirement for heat dissipation prevents >potting many of the components. So design it to not dissipate heat, pot it, seal it and be done. >Batteries don't do well when wet either. Use sealed subcompartments. >Someone on the Internet >recently suggested that the use of electronics would make >mechanically controlled units even safer. Nobody would make a general statement like that. Given a specific problem - yes it's possible. >I would love to see a real life example of this. In light of your stove and radio examples above I think I'll offer this example as a real life example: Dive lights have numerous failure modes. A major one is filament burn-out at turn-on. A current inrush limiter will elimitate filament failures at when the switch is thrown on. Especially valuable in lights used for back-up where a failure to give off light could spell trouble. >In my opinion, the only >time that electronics in a rebreather are justfied is when the high >efficiency of oxygen usage attained by these units is absolutely required by >the intended mission, and then ONLY if full electronic redundancy is >provided, including dual sensor arrays (three each), dual averaging and >control circuitry, dual batteries and dual displays, all in separate >housings. The problem with this is that *one* component failure effectively shuts down *all* the redundancy. In a complicated system this is not good but I agree it's better than no redundancy. As always, anything can fail at any time. Be prepared. john cc015012@br*.br*.ed*
Navigate by Author:
[Previous]
[Next]
[Author Search Index]
Navigate by Subject:
[Previous]
[Next]
[Subject Search Index]
[Send Reply] [Send Message with New Topic]
[Search Selection] [Mailing List Home] [Home]