Sean makes an excellent point.
With that in mind, maybe we can combine
the two ideas? By inserting the canister into a larger tube and then directing
the blower to pressurize the space between the two? That way, the larger scrubbing
surface will be absorbing CO2 at higher partial pressure and the fan will be
operating at higher efficiency.
Maybe I could build one and we could test
it?
Greg
From: owner-
Sent: Monday, August 08, 2011 2:38
PM
To:
Subject: RE: [PSUBS-MAILIST]
Snoopy life support test
Hi Greg,
I had always wondered about which fan
direction was better through the scrubber, and found Sean Stevenson had
once made a good case for drawing into the cylinder instead of blowing out. I
paste from his email:
The scrubber is a radial flow design, with the flow
direction from the outer diameter to the inner one. This makes most efficient
use of the scrubber material, as the flow area is greatest at the outer
diameter when the gas contains more CO2, and reduces as you approach the inner
tube, as CO2 is removed. Also, the inlet flow area of the scrubber (outer
diameter x pi x length) is huge in comparison to the minimum scrubber inlet
diameter, making the velocity (assuming even pressure distribution) almost nil
through the absorbent.
Another thing I like about drawing air out
is that the scrubbed air comes out in a specific direction. The scrubber is way
back against the aft end cap, so I can point the fan outlet forward and get a
slight ventilation effect. If I reversed it, the flow out of the scrubber would
not be directional.
Just two cents...
Alec
From: owner-
Sent: Monday, August 08, 2011
11:45 AM
To:
Subject: RE: [PSUBS-MAILIST]
Snoopy life support test
Hi Alec,
First, great job on the test!
With regard to fan amperage, the photos
look as if your radial fan was set up to draw air through the media as opposed
to blowing into- is that correct?
Generally speaking, if a fan draws a
partial vacuum, amperage will drop (because the load is going down).
Also, efficiency of the media should go up
with an increase in pressure. It would be very interesting to conduct the same
test with the fan “blowing” into the media instead.
Greg
From: owner-
Sent: Sunday, August 07, 2011 4:02
PM
To:
Subject: RE: [PSUBS-MAILIST]
Snoopy life support test
Hi everyone,
OK, let's try to answer some of all these
questions. First, my daughter is twelve but she's extremely tall for a
twelve year old, being already the same height as her mother. She
would still be on the light side for an adult, but she's definitely
more than half an adult. If I were diving with a crewmember of about my
size, my guess (and that is all it is) initial O2 setting would be 1.25
lpm instead of the 1 lpm used in the test with my daughter.
I agree the initial concentrations of
CO2 are high, and I think they are due to the time taken up by
getting into position and dogging down the hatch. I was working in the sub up
to about ten minutes before the first test, so there was probably some lingering
CO2 from my earlier presence too. The getting-ready process was
probably about 4 minutes, the last minute or so being with the
hatch closed. Just to make sure the instrument is not reading high, I
tested it outside and it reads 0.02% CO2 and 20.2% O2.
In terms of the permissible
concentrations of CO2, the air at the end of the unassisted dive did not
yet feel at all stale. We were at 1.6% CO2. But according to Phil's
life support white paper, the Navy's standard for manned submersibles
is to keep CO2 under 2.5%, while NASA's standard is 1% for
indefinite exposure and 3% for up to one hour. I could not tell any
difference in air quality between the three dives, other than by looking at the
instruments.
Yes, my O2 meter did track the Sub Aspida
readings, which is reassuring for when I ship the latter back to Jon! Problem
is, mine has one percent discrimination while the Sub Aspida goes to
a hundredth of a percent. On the unassisted dive mine went down to 19% and
then 18%, at which point its lower alarm went off, but on the two life support
dives mine didn't have the precision to show anything meaningful at all, and
was at 20% the whole time. Its functional enough to tell you if you're in
trouble, with a backlit display and low-high alarms, but useless for a test of
any accuracy. By the way, with something as accurate as the Sub Aspida you have
to be a bit careful in the measurements. If you aim a breath in its direction,
the reading shoots way up. If you raise it into the dome, it shoots way up.
Just walking from outside into my home, the CO2 increases by 300%.
Regarding the radial fan current versus
battery capacity, I've a new bit of information. Yesterday I noticed that the
fan's spec sheet lists the current draw as 0.22 amps, while the fan has
0.36 amps printed on it. Given the conflicting figures, I put down the
higher one. But Cliff's email prompted me to go out and measure the
current draw to settle the matter. It turns out the fan pulls only 0.2
amps with the scrubber full, so I guess the spec was right and the motor
sticker wrong. Another detail here is that Snoopy, being a little K250, is
a 12 volt boat with a single battery bank. The original K250 design
has three batteries wired in parallel. In the recent redesign I was
able to expand that with a fourth battery for a total of 316 amp hours,
but I still have the original single-bank configuration. Of course,
without a separate hotel battery, my reserve for the fan will depend on
the battery charge condition.
If anyone wants the specifics, the radial
fan is a Delta Electronics model BFB0712H. Here are some photos attached
of scrubber and fan.
Final point... I was not surprised by the
fan results. What really did surprise me was the stability of the cabin
pressure, in particular in view of the precision to which the Sub Aspida was
measuring it (1 millibar), and how easy it was to use a system with no feedback
loop regulating the flow. I wonder if that is typical or lucky?
Thanks,
Alec
The
contents of this e-mail are intended for the named addressee only. It contains
information that may be confidential. Unless you are the named addressee or an
authorized designee, you may not copy or use it, or disclose it to anyone else.
If you received it in error please notify us immediately and then destroy it.
From: owner-
Sent: Sunday, August 07, 2011 4:25
AM
To:
Subject: Re: [PSUBS-MAILIST]
Snoopy life support test
The CO2 readings are definitely high, especially in the test
without life support. I know that when I used the monitor out in the open
the CO2 percentage was between .01 and .03 so I don't expect a calibration
error is the cause of the high readings unless something happened to the device
during shipment. That should be easy enough for Alec to test however by
simply going to an open area and checking the CO2 reading.
Because the oxygen levels are also low it sounds feasible that the initial
environment was the result of respiration while getting both people settled in
the cabin and starting the test, although it doesn't really explain the drop in
CO2 from 1.6 percent at the end of the no-life-support test to the beginning of
the radial-fan test.
It looks to me like even the radial fan had a problem keeping up with the CO2
levels. It rises from .2 to .3 percent in the first four minutes and then
goes back down to .2 percent at 15 minutes and then starts rising again.
Given the CO2 data from both fan tests it might appear that the scrubber is not
efficient enough to handle two people in the cabin. Perhaps another
scrubber is necessary or a larger one, or there is something wrong with the
airflow over the sofnolime (packed too hard, obstruction, or fan is not
powerful enough). I would have expected the scrubber to handle not only
the respiration of the occupants but also reduce the CO2 concentration of the
existing atmosphere from the time the test started.
I find the no-life-support CO2 levels very interesting given how quickly the
levels accumulate. At about 43 minutes the CO2 level would have been at
3% or the maximum short term exposure allowed by OSHA. Simultaneously,
the O2 level would have hovered around 16%, borderline survivable and certainly
requiring immediate surfacing. What it really illustrates is that getting
into a situation where you cannot surface at the end of your planned dive, is
really going to ruin your day. Without life support, there's no room for
error, accident, or circumstance.
Jon
On 8/7/2011 2:37 AM, Cliff Redus wrote:
Alec, thanks for posting the life support test for
Snoopy. I quite enjoyed seeing actual test data. Your results of axial
vs. radial fans mirrored my results and my conclusions were the same; i.e.,
radial fans are the way to go. Have you done any test or calcs to see how
long a 0.36 amp current load will take to burn your backup batteries? My
guess is that at this current, you would not make 72 hours. IF they don
you may have to find a radial fan that draws less current.
I note from your test data that your CO2 readings look high. Atmosphric
air has about 300-400 ppm of CO2. As there are 10,000 ppm per 1%, this translates
to 0.03% to 0.04% . At the point you initiated each test, you were seeing
0.385, 0.25 and 0.27% for the base case, radial fan and axial fans,
respectively or in terms of ppm, 3800, 2000 and 2700. This could be
becuase your CO2 sensor is out of calibration or there was a lot of exhalation
of CO2 in the boat prior to the start of each test. When I did my in the
garage life support test, I found that it was necessary to use my air
compressor with a nozzle to purge the cabin to get the CO2 level back to normal
air for the start of each test. If the CO2 sensor readings are correct,
then with the axial fan, you reached the ABS maximum allowed CO2 level of
5000 ppm (0.5%) reading at 4 minutes. If the CO2 sensor had a 1600-1800
ppm bias error, then the axial fan would have also kept the cabin below the
5000 ppm for the duration of the test.
Thanks again for posting.
Cheers Cliff