Re: [PSUBS-MAILIST] life support method?

["Sean T. Stevenson" <ststev@uniserve.com>]

Although the ambient CO2 level is useful information, I don't think it
needs to be an integral part of the scrubbing system, simply due to the
fact that CO2, in the volume of a sub, will build up very slowly, most
scrubber systems will be able to handle the volume even with
intermittent operation, and that the preliminary effects of hypercapnia
can be noticed by the astute vehicle occupant.  It just seems to me to
be unnecessary electronic control.  How reliable are these CO2 sensors?
 It makes more sense to me to calculate, based on estimated metabolic
production, how much scrubber material is necessary and with what flow
rate of cabin air through the material.  I think the better application
of a CO2 sensor would be simply as a passive display, to indicate
possible scrubber or environmental control failure, without actually
relying on the sensor to turn the scrubber fan on.  In my conceptual
PSub design, I have two circulation fans, one of which is running at
all times.  Passing air through the scrubber that doesn't really need
scrubbing does no harm to the scrubber material, apart from the
degrading effects of moisture, but I have a dehumidifier upstream of
the scrubber to deal with that.  Everything in this vehicle is PID
control (environmental controls included) with manual overrides for all
functions.  As for oxygen addition, I went with automatic addition via
a redundant (self diagnostic) sensor array, as I stated in a previous
message.  Now, I should point out that I am a fan of safety by
simplicity, and like to avoid unnecessary electronics or other
unwarranted complexity, but given that the rate of oxygen metabolism
changes with workload and/or stress, and that this vehicle can operate
at varying internal pressures, this seemed a better way to go than
passive addition.  If you are going to go with automatic oxygen level
control, I would not trust that to a single O2 sensor.

-Sean


On Sat, 05 Aug 2000 10:01:31 -0400, Chris Horne wrote:

>OK, after absorbing everything in this thread about life support and doing a
>little research out there and feel a bit more aware; I even got the "Life Support
>in Small 1ATM Sytems" white paper from Nuytco Research  yesterday.
>
>It seems several things are true:
>    - We need a way to "trade off" CO2 that we exhale for O2
>    - It needs to be reliable and efficient
>    - It could be electronically controlled or manually controlled.
>    - Electronic control could measure barometric pressure, which should only
>change in an enclosed environment when a person used up O2 and the CO2 was
>converted to a solid, or could measure O2 directly in the air by percentage.
>    - Manual O2 mechanism would be needed whether a electronic control existed or
>not, just in case
>
>So why not just build a system comprised of a embedded controller, an electronic
>solenoid or valve made for O2, an O2 sensor, and a CO2 sensor? I'm sure I'm not
>the only one on the list that can work with embedded controllers (PIC is what I
>had in mind, but something as simple as BASIC Stamp would do the job), O2 sensors
>are simply create 0.488mV/percent O2, or around 10.2 millivolts for our
>application, so the interface would be no more than a simple A/D conversion, with
>possibly an opamp (I'm still checking into the CO2 sensor, but I know it's very
>similar). The logic for the controller would be very simple- continuously monitor
>the O2 and CO2 and barometric pressure. Constantly allow a small amount of O2
>into the air at all times (maybe .2lpm/person, as we consume around .25lpm), and
>the controller adds a small amount when the O2 level dips. Monitor CO2 and
>barometric pressure simply for reference. Create two of these systems, or imply
>interface 2 or more sensors into the same unit.  Ideas? Comments?
>
>-chris
>
>