Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Zero-G Still

0 views
Skip to first unread message

Earl Colby Pottinger

unread,
Nov 17, 2004, 11:43:00 AM11/17/04
to sci-spa...@moderators.isc.org
How do you make a still that can process fluids in Zero-G?

My first guess is that the evaporator uses some kind of mat to how the liquid
being processed, but if there are lots of solids produce that the mat will
need changing often.

Suggestions?

Earl Colby Pottinger

--
I make public email sent to me! Hydrogen Peroxide Rockets, OpenBeos,
SerialTransfer 3.0, RAMDISK, BoatBuilding, DIY TabletPC. What happened to
the time? http://webhome.idirect.com/~earlcp

Tommy

unread,
Nov 20, 2004, 12:15:57 AM11/20/04
to sci-spa...@moderators.isc.org
could you ionize the gas then accelerate it through a magnetic field to
seperate it by molecular weight?

Hephaestus

unread,
Nov 22, 2004, 5:00:36 PM11/22/04
to sci-spa...@moderators.isc.org
Earl Colby Pottinger <ear...@idirect.com> wrote in message news:<9dednccKE_w...@look.ca>...

> How do you make a still that can process fluids in Zero-G?

What's the difficulty? You need a little force to keep the liquid
down and make the vapor bubbles rise, but that isn't hard to arrange.
Spinning the container is sufficient; this is often done anyway to
ensure constant mixing and a nice smooth boil.

Cameron Dorrough

unread,
Nov 21, 2004, 5:29:24 PM11/21/04
to sci-spa...@moderators.isc.org
"Earl Colby Pottinger" <ear...@idirect.com> wrote in message
news:9dednccKE_w...@look.ca...
> How do you make a still that can process fluids in Zero-G?
>
> My first guess is that the evaporator uses some kind of mat to how the
liquid
> being processed, but if there are lots of solids produce that the mat will
> need changing often.
>
> Suggestions?

If the fluid was at least partly volatile (eg. alcohol.. ;-) you could
circulate it through a cross-flow heat exchanger to heat it up past boiling
point.

Once up to temperature, then spray it into one end of a baffled, rotating,
distillation column, heated at one end and cooled at the other. The
condensate should(*) collect at the appropriate location along the column
depending on the temperature at which it condenses. The column needs to
rotate to get the condensate droplets to contact the outer wall, where it
can be collected - but not so fast that nothing ever gets to the far end.
The baffles prevent the condensate remixing.

(*) - but then again, never having tried this for real, I wouldn't really
know. ;-)

Cameron:-)


Anvil

unread,
Nov 22, 2004, 12:46:34 PM11/22/04
to sci-spa...@moderators.isc.org
Earl Colby Pottinger:

> How do you make a still that can process fluids in Zero-G?
*snip*
> Suggestions?
>
Spin the evaporator and condenser. The evaporator could be
spherical but the condenser/collector end would be better
disk like with the condenser like a stack of thin plates
with thin gaps (think Tesla turbine). The remainder of the
cylindrical area would be for collection. Adding pumps,
pipes, and rotary unions a known process could be designed
as continuous.
--
Anvil*

Mike Miller

unread,
Nov 22, 2004, 8:59:57 AM11/22/04
to sci-spa...@moderators.isc.org
Earl Colby Pottinger <ear...@idirect.com> wrote in message news:<9dednccKE_w...@look.ca>...
> How do you make a still that can process fluids in Zero-G?

Use a pump to circulate the liquids in a heated cylinder; don't let in
enough liquid to fill the cylinder. You can add inlet and outlet ports
to sustain continuous circulation. (Alternately, instead of a separate
pump, just spin the cylinder.)

Centripetal force will pin the liquid to the walls. Normal
differential evaporation will carry the substance with a lower
evaporation temperature to the hollow center of the cylinder, where
vapor pressure (created by the boiling) and/or fans can drive the
vapor out through another outlet port.

Further plumbing requirements are dependent on the subtances involved.
You can probably arrange the circulating liquid in the cylinder to
carry off any solids to a filter (or centripetal separator or
something). If the distilled vapors (e.g., alcohol) need cooling, you
can run them through some chilled piping and either depend on vapor
pressure and/or pumps to keep the stuff moving.

Mike Miller, Materials Engineer

Perplexed in Peoria

unread,
Nov 20, 2004, 11:09:04 PM11/20/04
to sci-spa...@moderators.isc.org

"Earl Colby Pottinger" <ear...@idirect.com> wrote in message news:9dednccKE_w...@look.ca...
> How do you make a still that can process fluids in Zero-G?
>
> My first guess is that the evaporator uses some kind of mat to how the liquid
> being processed, but if there are lots of solids produce that the mat will
> need changing often.
>
> Suggestions?

Spin it. Get pseudo g from the centrifugal force.

Henry Spencer

unread,
Nov 23, 2004, 6:26:17 PM11/23/04
to
In article <hqAnd.361199$wV.242200@attbi_s54>,

Tommy <tlsn...@mchsi.com> wrote:
>> How do you make a still that can process fluids in Zero-G?
>
>could you ionize the gas then accelerate it through a magnetic field to
>seperate it by molecular weight?

Unfortunately, the mass flow through such a system is tiny. Space-charge
problems limit the beam current -- when the ion density in the beam gets
high enough, not only does it start to spread of its own accord, but it
distorts the fields used to accelerate the ions -- even with the use of
some clever tricks. And it's difficult to ionize a large fraction of the
material, so non-ionized gas gets everywhere and messes things up if you
try for too strong a flow.

The "calutron" isotope separator built for the Manhattan Project, the size
of a small car, processed a few grams of material per day. Energy use was
huge and maintenance loads were high. That was a harder problem, mind
you, but you'd hit the same fundamental limits. Doing any sort of useful
materials processing that way is basically impractical.

(The Manhattan Project filled a huge building with over a thousand
calutrons, and used them to do the uranium enrichment for the Hiroshima
bomb. Once the first gaseous-diffusion enrichment plant came on line,
most of the calutrons were immediately shut down and scrapped.)
--
"Think outside the box -- the box isn't our friend." | Henry Spencer
-- George Herbert | he...@spsystems.net

Henry Spencer

unread,
Nov 24, 2004, 9:25:51 AM11/24/04
to
In article <cnr4s5$dso$1...@news-02.connect.com.au>,

Cameron Dorrough <cdor...@nortonconsultants.com> wrote:
>If the fluid was at least partly volatile (eg. alcohol.. ;-) you could
>circulate it through a cross-flow heat exchanger to heat it up past boiling
>point.

One problem that arises is that some of the fluids you'd like to distill,
e.g. urine, don't take well to serious heating -- they tend to leave
deposits behind.

Tommy

unread,
Nov 24, 2004, 5:21:16 PM11/24/04
to sci-spa...@moderators.isc.org
Why would you need to heat them, just lower the pressure to the vapor
point and then compress them back to a liquid after seperation.

Cameron Dorrough

unread,
Nov 24, 2004, 6:05:25 PM11/24/04
to sci-spa...@moderators.isc.org
"Henry Spencer" <he...@spsystems.net> wrote in message
news:I7otF...@spsystems.net...

> In article <cnr4s5$dso$1...@news-02.connect.com.au>,
> Cameron Dorrough <cdor...@nortonconsultants.com> wrote:
> >If the fluid was at least partly volatile (eg. alcohol.. ;-) you could
> >circulate it through a cross-flow heat exchanger to heat it up past
boiling
> >point.
>
> One problem that arises is that some of the fluids you'd like to distill,
> e.g. urine, don't take well to serious heating -- they tend to leave
> deposits behind.

True - I was thinking the OP had analytical experimentation in mind..

Urine would need some serious filtration - even as a superheated vapour - to
get the muck out of it. Is it only the water that would be required or are
there other useful compounds in there as well??

Cameron:-)


Paul F. Dietz

unread,
Nov 24, 2004, 6:54:32 PM11/24/04
to sci-spa...@moderators.isc.org
Henry Spencer wrote:

> One problem that arises is that some of the fluids you'd like to distill,
> e.g. urine, don't take well to serious heating -- they tend to leave
> deposits behind.

The obvious answer there is to distill at reduced pressure. Also,
organic compounds could be oxidized before distillation (reaction
with ozone, perhaps?)

Paul

Henry Spencer

unread,
Nov 24, 2004, 9:42:22 PM11/24/04
to
In article <C8adnRyIUpi...@dls.net>,

Paul F. Dietz <di...@dls.net> wrote:
>> One problem that arises is that some of the fluids you'd like to distill,
>> e.g. urine, don't take well to serious heating -- they tend to leave
>> deposits behind.
>
>The obvious answer there is to distill at reduced pressure.

Indeed so.

>Also, organic compounds could be oxidized before distillation (reaction
>with ozone, perhaps?)

Unfortunately, one of the reasons you might want to heat human wastes is
for the absolutely definitive oxidizing process: SCWO, SuperCritical
Water Oxidation. Feed most anything -- human waste, plastic, bodies :-),
whatever -- into an SCWO reactor, and out comes CO2, N2, H2O, and sterile
metal-oxide ash. But it does mean getting the stuff hot, and the deposits
problem makes it difficult to recover the waste heat, so it's dreadfully
energy-intensive.

Dan DeLong

unread,
Nov 26, 2004, 1:41:04 PM11/26/04
to sci-spa...@moderators.isc.org
>
> Unfortunately, one of the reasons you might want to heat human wastes is
> for the absolutely definitive oxidizing process: SCWO, SuperCritical
> Water Oxidation. Feed most anything -- human waste, plastic, bodies :-),
> whatever -- into an SCWO reactor, and out comes CO2, N2, H2O, and sterile
> metal-oxide ash. But it does mean getting the stuff hot, and the deposits
> problem makes it difficult to recover the waste heat, so it's dreadfully
> energy-intensive.

I've been away from it for 9 years, so my information may be out of
date, but at the time, energy consumption was the big problem with
SCWO. I got a small IR&D budget and built a SCWO that minimized energy
consumption. The mass throughput was sized for 4 people assumed on the
Space Station and it processed all the water streams except fecal
matter (I was working for the SSF prime contractor and we used the
official water model). I don't recall the flowrate, but it was a few
liters per minute. Anyway, I got the process down to 250 Watts
including both thermal and pumping power. Probably could have gotten
another factor of two lower but neither NASA nor the SSF prime
contractor were interested, as there was already a baseline hardware
architecture.

The project didn't address accumulation of metals and salts in the
apparatus. The design was not particularly good or bad in that
respect, just not considered. So I won the battle of proving that SCWO
is not necessarily energy intensive, but lost the war to have it
developed into a useful apparatus. I agree with Henry, the nice thing
about SCWO is that it doesn't kill the bugs or trap them in a filter,
it turns them into CO2 and H2O. Including viruses.

Dan DeLong

BTW, urine at 3200 psi and 400 C is extremely corrosive!

Henry Spencer

unread,
Nov 27, 2004, 1:53:00 AM11/27/04
to
In article <cf9b38fa.0411...@posting.google.com>,
Dan DeLong <del...@earthlink.net> wrote:
>[SCWO] Anyway, I got the process down to 250 Watts

>including both thermal and pumping power. Probably could have gotten
>another factor of two lower but neither NASA nor the SSF prime
>contractor were interested...

Hmm, interesting, I'd never heard of that one -- not surprising, I guess.
Did this get written up anywhere accessible?

dave schneider

unread,
Dec 3, 2004, 7:07:06 PM12/3/04
to sci-spa...@moderators.isc.org
"Perplexed in Peoria" <jimme...@sbcglobal.net> wrote:
[...]

> Spin it. Get pseudo g from the centrifugal force.

>From the caption at <http://mediaarchive.ksc.nasa.gov/detail.cfm?mediaid=23072>,
the word "rotational" sounds like that is what NASA is getting ready
to do.

The water processing release in my s.s.s/s.s.m thread is no longer
available at the msfc site; go to the "real" one at ames, instead:
<http://www1.nasa.gov/centers/ames/news/04_104AR.html>

/dps

Cra...@gmail.com

unread,
Dec 6, 2004, 7:26:45 AM12/6/04
to sci-spa...@moderators.isc.org

Henry Spencer wrote:
> But it does mean getting the stuff hot, and the deposits
> problem makes it difficult to recover the waste heat, so it's
dreadfully
> energy-intensive.

I can't see my last reply, so...

Could the heat exchange occur away from a deposit filter/trap, so that
you're only capturing heat from a relatively pure liquid or gas?
Mike Miller, Materials Engineer

Henry Spencer

unread,
Dec 13, 2004, 6:36:48 PM12/13/04
to
In article <co343p$9h2$1...@news-02.connect.com.au>,

Cameron Dorrough <cdor...@nortonconsultants.com> wrote:
>Urine would need some serious filtration - even as a superheated vapour - to
>get the muck out of it. Is it only the water that would be required or are
>there other useful compounds in there as well??

The water is most of what you care about. The rest is only really
relevant when you start to think about recycling food.

Alfred Montestruc

unread,
Dec 15, 2004, 12:51:14 AM12/15/04
to sci-spa...@moderators.isc.org

Earl Colby Pottinger wrote:
> How do you make a still that can process fluids in Zero-G?

My first cut would be to use centripital force to seperate the
different densities.

So you make a container that can be rotated that has a balance arm that
you can adjust on the fly as the liquid evaporates to keep it in
balance, a swivel pipe connector that allows the vapor to pass through
a rotating connector to another pipe system that can condense it, a
cooling liquid that goes on a runaround to an outside radiator to dump
heat, and an electric heater to add heat to the liquid to be distilled.
One can add a second pipe swivel to add more liquid.

It should work just fine, and I doubt you will need a whole 1 gee at
the outside to keep the liquid from the vapor.


>
> My first guess is that the evaporator uses some kind of mat to how
the liquid
> being processed, but if there are lots of solids produce that the mat
will
> need changing often.

And cleaning, that sounds really labor intensive and so expensive.

Ian Woollard

unread,
Jan 9, 2005, 10:53:38 PM1/9/05
to Alfred Montestruc, sci-spa...@moderators.isc.org
Alfred Montestruc wrote:
> Earl Colby Pottinger wrote:
>>How do you make a still that can process fluids in Zero-G?
>
> My first cut would be to use centripital force to seperate the
> different densities.

Yes. Accelerations, particularly rotational acceleration seems to be the
only way to go.

> So you make a container that can be rotated that has a balance arm that
> you can adjust on the fly as the liquid evaporates to keep it in
> balance, a swivel pipe connector that allows the vapor to pass through
> a rotating connector to another pipe system that can condense it, a
> cooling liquid that goes on a runaround to an outside radiator to dump
> heat, and an electric heater to add heat to the liquid to be distilled.
> One can add a second pipe swivel to add more liquid.

I would think it should be possible to form a vortex boiler i.e. use a
propeller to swirl the heated liquid you're trying to still around the
rim of a fixed cylindrical vessel, and pipe the vapour out the central
axis and bubble it/blow it into/onto a fixed, cooled liquid/cylindrical
surface.

If you do it right, the distilled liquid will spin around too, and you
can collect it without worrying too much about foaming.

>> Earl Colby Pottinger

D Schneider

unread,
Jan 10, 2005, 2:15:39 PM1/10/05
to sci-spa...@moderators.isc.org
Ian Woollard <junk...@wolfkeeper.plus.com> wrote:

[...]


> Yes. Accelerations, particularly rotational acceleration seems to be the
> only way to go.

Could you tell anything about what type of rotational apparratus NASA is
looking at from the photo of the columnar part? (see link earlier in this
thread).

/dps

--
Using Opera's revolutionary e-mail client: http://www.opera.com/m2/

0 new messages