On 2007-10-08 18:17:58 -0400, Mark Borgerson said:

Getting rid of the disssolved gases in the headspace and as bubbles
forming on the sides of the tube is going to be a major headache.

Not a headache, an impossibility (they're not really dissolved at that
point though) :-) That, and the increase in pressure due to water vapor
will make this an oscillating, self-quenching system. It'll require
more and more heat as the partial pressures of the non-condensables
increases, and the column heights will drop as the pressure goes up,
with the diffusion path increasing the whole time.

I agree with that part---except for the oscillation part. I think the
processes are slow enough and the thermal and physical masses are high
enough that the oscillations will be damped out and you will see a slow
change to equilibrium with little or no overshoot.

Though I consider this whole discussion impractical, I haven't seen
anyone mention that the fresh-water side will be drawn down fairly
regularly. And, of course, the sea water side will be replenished from
time to time.

Suck hard enough on the fresh-water side and you get even better
"vacuum" at the top. (Dissolved gasses are likely to be a problem,
though.) Cool the fresh-water side and water vapor will condense there
-- the whole point of the exercise.

Thinking only momentarily on a problem that I have little interest
in... if the fresh-water side is evacuated to the point that the
salt-water side is slightly below the top, every once in a while (or
perhaps often), the fresh-water side will be empty and only the
previously-dissolved gasses evacuated.

The required evacuation pumps and one-way valves sound like the problem
at the moment.

--
Jere Lull
Tanzer 28 #4 out of Tolchester, MD
Xan's new pages: http://web.mac.com/jerelull/iWeb/Xan/
Our BVI pages: http://homepage.mac.com/jerelull/BVI/

Brian Whatcott wrote:
On Mon, 08 Oct 2007 10:13:17 -0700, Keith Hughes
wrote:

...I proposed filling the tubes
completely with water so that the initial head space would be zero.

No, it won't be zero. It can't be. If it is, then you have a solid
liquid stream, and it's just a siphon.
...
A solid liquid loop will not separate into two separate columns.
...
Keith Hughes

Perhaps it would be better for you to check what is the maximal rise
(head) of a syphon. Can you guess what it might be?

Uhmmm, maybe you should follow the thread. If you *pump* the water up
the columns, evacuating the headspace as you go, until the columns meet,
you can siphon pretty much any height you want. Until outgassing
creates a headspace. If the bases of the columns are at differing
heights, you have a siphon until that happens.

Keith Hughes

Mark Borgerson wrote:
In article ,
says...
Mark Borgerson wrote:
In article ,
says...
SNIP
You need to get back to the gas law to see where this error lies. You
have to *create* the vacuum. That requires a HUGE increase in volume
for whatever the initial headspace is. For this to happen you need a
much longer tube to start with.

You seem to have missed the fact that I proposed filling the tubes
completely with water so that the initial head space would be zero.
No, it won't be zero. It can't be. If it is, then you have a solid
liquid stream, and it's just a siphon. You have to have headspace. And
it has to be sufficient to maintain separation of the seawater and
freshwater to prevent contamination when filling the tubes. And it has
to be large enough to prevent percolation carryover when boiling is
initiated.

At that point you release the pressure on the water and it falls
to the point where water weight plus vapor pressure equals 1ATm.
A solid liquid loop will not separate into two separate columns. They
have to be separated by a headspace. You can heat the seawater side and
create a headspace by liberating dissolved gases, then let the columns
drop to create vacuum, but you will have contaminated the freshwater side.


The head space is generated by the evaporation (or boiling) of some of
the water in a column. It's exactly the same principle that you get it
you fill a closed tube full of mercury and then invert it, placing the
end in a reservoir of mercury.

You seem to be forgetting that the whole purpose is to Purify/desalinate
the water. No initial headspace = single process stream = contamination
on the distillate side.

(We call these things barometers.)

Except when we call them Mcleod gauges...

You start with no head space, but when you invert it, VOILA!
head space appears as the mercury sinks to a level where the weight
of the mercury equals the atmospheric pressure. You get a much
better vacuum with mercury, since it has a much lower vapor pressure
at room temperature.

A column of water will behave the same way. The column just has
to be much taller.

True, but you need to keep the context - water purification. The
contamination control features are as crucial to the operational
constraints as are the physical parameters. Thus, you have to *Start*
with headspace. Sure, you could purge the freshwater side until the
contaminants are removed, but by then most, if not all, of your
production will be wasted.


Some of the historical references on water barometers mention that,
despite precautions, the water in the barometer eventually got
contaminated with dissolved gases and they lost their accuracy.

Yes, you can only deaerate so far prior to filling. Personally, I've
never seen an absolute pressure water barometer. IME they are primarily
used in an inverted u-tube configuration for DP measurements. BTW,
mercury barometers suffer the same fate, primarily through oxidation of
the mercury, changing the density. Just look at that almost black film
layer on any old barometer.


I agree with that part---except for the oscillation part. I think
the processes are slow enough and the thermal and physical masses
are high enough that the oscillations will be damped out and you
will see a slow change to equilibrium with little or no overshoot.

You may be right, but I doubt it. Unless you control the temperature
versus pressure relationship, which is virtually impossible with any
passive heating process, then I'd expect self quenching would result in
an oscillating system.

Keith Hughes

On Mon, 08 Oct 2007 17:36:59 -0700, Keith Hughes
wrote:


... If you *pump* the water up
the columns, evacuating the headspace as you go, until the columns meet,
you can siphon pretty much any height you want.
...
Keith Hughes


No, you cannot syphon to any height.
Yes, you can pump to any height, within reason.
There is a difference.
If you pump to 40 feet head, quite filling an inverted u-tube, for
instance, when the pump is halted, the water falls until a particular
maximum height is no longer exceeded, if the foot of the u-tube is
opened to water whose surface is at atmospheric pressure..

Brian W

Brian Whatcott wrote:
On Mon, 08 Oct 2007 05:33:39 -0700, Keith
wrote:

Boiling Point Elevation
The boiling point of a solution is higher than that of the pure
solvent. Accordingly, the use of a solution, rather than a pure
liquid, in antifreeze serves to keep the mixture from boiling in a hot
automobile engine.....
From:
http://www.chemistryexplained.com/Ce...roperties.html


Actually, no. Ethylene glycol in its pure liquid state boils near
200 degC
http://www.dow.com/ethyleneglycol/about/properties.htm

It is usually cut to 50% dilution for use as an antifreeze.

Brian Whatcott Altus OK

Since he used the term "solvent", the assumption, at least on my end, is
that he's talking about a solvent/solute system, not a solution of
miscible liquids.

Keith Hughes

Jere Lull wrote:
On 2007-10-08 18:17:58 -0400, Mark Borgerson said:

snip

Though I consider this whole discussion impractical,

I don't know that I'd call the *discussion* impractical; the device
certainly. Kind of the point of the discussion.

I haven't seen
anyone mention that the fresh-water side will be drawn down fairly
regularly. And, of course, the sea water side will be replenished from
time to time.

Suck hard enough on the fresh-water side and you get even better
"vacuum" at the top. (Dissolved gasses are likely to be a problem,
though.) Cool the fresh-water side and water vapor will condense there
-- the whole point of the exercise.

Thinking only momentarily on a problem that I have little interest in...
if the fresh-water side is evacuated to the point that the salt-water
side is slightly below the top, every once in a while (or perhaps
often), the fresh-water side will be empty and only the
previously-dissolved gasses evacuated.

The required evacuation pumps and one-way valves sound like the problem
at the moment.

The whole exercise was to get a passive system. If you're going to add
a vacuum pump, then you just provide continuous evacuation on the
freshwater side, using a demister that drains into the freshwater pool,
to separate the water vapor from the non-condensables. But if you
accept the need for a pump, why use this rather byzantine approach at all?


Keith Hughes

In article ,
says...
Brian Whatcott wrote:
On Mon, 08 Oct 2007 10:13:17 -0700, Keith Hughes
wrote:

...I proposed filling the tubes
completely with water so that the initial head space would be zero.

No, it won't be zero. It can't be. If it is, then you have a solid
liquid stream, and it's just a siphon.
...
A solid liquid loop will not separate into two separate columns.
...
Keith Hughes

Perhaps it would be better for you to check what is the maximal rise
(head) of a syphon. Can you guess what it might be?

Uhmmm, maybe you should follow the thread. If you *pump* the water up
the columns, evacuating the headspace as you go, until the columns meet,
you can siphon pretty much any height you want. Until outgassing
creates a headspace. If the bases of the columns are at differing
heights, you have a siphon until that happens.

If you pump both the fresh and salt water to the top of the U-Tube,
then switch from the pumps to the reservoirs at the bottom,
you won't get a siphon. The boiling of water at the top will
break the siphon action.

Mark Borgerson

In article , betwys1
@sbcglobal.net says...
On Mon, 08 Oct 2007 05:33:39 -0700, Keith
wrote:

Boiling Point Elevation
The boiling point of a solution is higher than that of the pure
solvent. Accordingly, the use of a solution, rather than a pure
liquid, in antifreeze serves to keep the mixture from boiling in a hot
automobile engine.....
From:
http://www.chemistryexplained.com/Ce...roperties.html


Actually, no. Ethylene glycol in its pure liquid state boils near
200 degC
http://www.dow.com/ethyleneglycol/about/properties.htm

It is usually cut to 50% dilution for use as an antifreeze.

True---when mixing liquids, the boiling point is somewhere between
the boiling points of the two.

Radiators and cooling systems are pressurized so that the system
can have an elevated boiling point.


Mark Borgerson

In article ,
says...
Mark Borgerson wrote:
In article ,
says...
Mark Borgerson wrote:
In article ,
says...
SNIP
You need to get back to the gas law to see where this error lies. You
have to *create* the vacuum. That requires a HUGE increase in volume
for whatever the initial headspace is. For this to happen you need a
much longer tube to start with.

You seem to have missed the fact that I proposed filling the tubes
completely with water so that the initial head space would be zero.
No, it won't be zero. It can't be. If it is, then you have a solid
liquid stream, and it's just a siphon. You have to have headspace. And
it has to be sufficient to maintain separation of the seawater and
freshwater to prevent contamination when filling the tubes. And it has
to be large enough to prevent percolation carryover when boiling is
initiated.

At that point you release the pressure on the water and it falls
to the point where water weight plus vapor pressure equals 1ATm.
A solid liquid loop will not separate into two separate columns. They
have to be separated by a headspace. You can heat the seawater side and
create a headspace by liberating dissolved gases, then let the columns
drop to create vacuum, but you will have contaminated the freshwater side.


The head space is generated by the evaporation (or boiling) of some of
the water in a column. It's exactly the same principle that you get it
you fill a closed tube full of mercury and then invert it, placing the
end in a reservoir of mercury.

You seem to be forgetting that the whole purpose is to Purify/desalinate
the water. No initial headspace = single process stream = contamination
on the distillate side.

(We call these things barometers.)

Except when we call them Mcleod gauges...

You start with no head space, but when you invert it, VOILA!
head space appears as the mercury sinks to a level where the weight
of the mercury equals the atmospheric pressure. You get a much
better vacuum with mercury, since it has a much lower vapor pressure
at room temperature.

A column of water will behave the same way. The column just has
to be much taller.

True, but you need to keep the context - water purification. The
contamination control features are as crucial to the operational
constraints as are the physical parameters. Thus, you have to *Start*
with headspace. Sure, you could purge the freshwater side until the
contaminants are removed, but by then most, if not all, of your
production will be wasted.

With the proper placement of the check valves, I think you could
start with the initial boiling happening in the freshwater side---
after all it is going to boil at a lower temperature.

The procedure might look like this:

1 Pump both fresh and salt water to near the top.
2. Shut offf the salt water side pump, but keep the
tube closed at the bottom.
3. Pump a bit more fresh water into the tube---where
it overflows to the sal****er side, displacing
the rest of the air out the check valve.

You now have no air in the tube and a small layer of
fresh water on top of the salt water.

4 Release the pressure at the bottom, and the fresh
water at the top will boil and create your head space
with little or no contamination of the freshwater
side.

5 Apply your heat differential and remove distilled
fresh water as it overflows the reservoir at the
bottom.

This should work until the dissolved gas problem lengthens
the vapor path to the point where you have to start over
at step 1.




Some of the historical references on water barometers mention that,
despite precautions, the water in the barometer eventually got
contaminated with dissolved gases and they lost their accuracy.

Yes, you can only deaerate so far prior to filling. Personally, I've
never seen an absolute pressure water barometer. IME they are primarily
used in an inverted u-tube configuration for DP measurements. BTW,
mercury barometers suffer the same fate, primarily through oxidation of
the mercury, changing the density. Just look at that almost black film
layer on any old barometer.

That is mostly due to contaminants trapped in the glass and impurities
in the mercury. Production barometers didn't use glass that was
heated with a vacuum to remove contaminants.


I agree with that part---except for the oscillation part. I think
the processes are slow enough and the thermal and physical masses
are high enough that the oscillations will be damped out and you
will see a slow change to equilibrium with little or no overshoot.

You may be right, but I doubt it. Unless you control the temperature
versus pressure relationship, which is virtually impossible with any
passive heating process, then I'd expect self quenching would result in
an oscillating system.

What do you mean by "self-quenching"?


Mark Borgerson

On Mon, 8 Oct 2007 18:22:14 -0700, Mark Borgerson
wrote:
....
If you pump both the fresh and salt water to the top of the U-Tube,
then switch from the pumps to the reservoirs at the bottom,
you won't get a siphon. The boiling of water at the top will
break the siphon action.

Mark Borgerson


I am regretting this already.
But If I repeat this little test, pumping mercury up an inverted
U-tube to 35 inches, when I stop the pump and open the tubes to a
mercury pool, the mercury levels in the two tubes drop to a 29.92
inch column each side. The mercury does not boil.

29.92 inches is 760 mm of mercury, by the way. So boiling is not
essential to breaking a syphon.
Excess height is all that is needed.

Brian W