rdmmdr

Alex

the whole talk about protecting the boards during a flood seems mute because after a flood only an idiot would try to wash the board and reuse. it is a paper weight at that point, i understand the isolation you are trying to accomplish for the hand set comms it would make a repair cheaper after the flood. i would think however if the unit did flood, you would want the board to burn out so it could not be used by someone trying to get cheap.

i really like the idea of the flow control valve due to fact that loss of power/ control gives a fixed oraface, but how are handling a runaway of the valve controller. it seems that you have added multiple modes of failure here not subtracted them and increased the complexity of the part way beyond what it needs to be. the basic problem with the solenoids is corrosion of the plunger. a redesign of the plunger seems to be the best solution to this issue and the simplest. life threating issue with the solenoids are stuck open or failure to open, failure to open is a problem that you have time to deal with. stuck open, while a more critical problem time wise, can be mediate with a properly design flow limiter,a pneumatic fuse and proper training.

you are placing the batteries in the loop, a battery short in a high humidity high o2 environment is a failure i would not be comfortable with, how are you dealing with this issue.

encapsulating, you say that this causes more problems then it solves. i agree that there are problems with encapsulating the electronics, thermal stress and mechanical stress imposed upon the board, but don,t the benefits outweigh the a limited number of damaged boards during manufacture. and since the industry has been doing it for years these problems have been solved.

having talked to both ai and teledyne about the molex connector, what shitty idea those where for a rebreather, any decent molded connector will give you better results. i personally am using the the r-17d for that reason and have asked both manufacturers to look in to gold plating their connectors. i will take the risk of corrosion on the cell side rather then having to replace the harness. i question the protection of the input Chanel does not the load resister handle this and if the load resister is bad then is not the board already shot.

since you have installed a pressure sensor in the unit i assume that you will be including some deco software, what will be the mandatory re-cal requirement for the sensor. since even using two sensors here will not give you redundancy since both sensors are subjected to the same mechanical and thermal stresses. we annually service our regs but how many throw their deco computer in the pot to check the sensor.
rick