any limits on mechanical seals?

Joe Strout wrote:ß
-- docking more than one ship at a time is only one.
If you have a wheel with empty space in the middle you can
use both sides of the axle.
Depending on the size of the ships and the station you can set up
lots more docking stations along a noncontinuous axle.

I'm afraid I'm not quite following you here. Can you elaborate?
Perhaps you're picturing an axle like this (ASCII art ahead!):

_____ _____
| | | |
___| |______| |___ ...

so that you can dock one small ship in each bend of the axle?
That would work for small ships but what I had in mind was
something different:

Picture a wheel of a bycicle, nave, lots of spokes, tyre and all.
People live in the tyre and, say, 8 spokes (four on each side)
point into the middle towards the nave.
Make the nave really long and slim and attach the spokes
not at two disks at either end of the nave but at various points
along the axis of the nave.
Still with me? You now have 8 spokes, say one of them being
the shortest because it points straight to the nave, following
the radius of the wheel. Two are the longest one because they
are attached to the outer ends of the nave, everything else lies
somewhere in between. The exact layout depends on whether you have
an even or odd number of spokes.

Now remove the nave, having the spokes end in nothing at the
theoretical rotational axle, pointing into emptiness.
The docking ports point along the rotational axle, so any ship
can fly inside in between the spokes, find one end point, go
stationary with respect to the station and match rotational velocity
easily.

Now, with 8 spokes, each spoke having two ports you can serve
16 ships in parallel. Put the spokes in the right order and
you can have a nice spiral approach pattern for the ships.

Here's a picture with two spokes and two ships ("S") parked
at each spoke. You are looking at the wheel from the side:

Wheel: O Tyre. People live in there.
|\
| \ Spoke
| \
S:S | S:S
\ |
\ |
Spoke \|
O same Tyre.



Lots of Greetings!
Volker
--
For email replies, please substitute the obvious.

On Fri, 7 Jul 2006, Jeff Findley wrote:
"William Elliot" wrote in message
On Thu, 6 Jul 2006, Joe Strout wrote:
William Elliot wrote:
On Wed, 5 Jul 2006, Joe Strout wrote:

Suppose you have a large space station (OK, let's say a colony)
with a rotating portion and a stationary portion, both
pressurized, and with constant traffic back and forth between
them. Obviously you need a large mechanical seal between them,
and I have some questions about that which I hope someone can
answer:

Lets say the stationary protion is in the center of a wheel.
Instead of this dubious mechincal arrangement, it would be easier
landing.

It's not about arriving cargo. We need a rotating and a stationary
section for lots of reasons -- docking more than one ship at a time is
only one.

Make a parking structure, five decks for five arrivals. Even better is
that a craft can land on both sides of a deck. Thus three decks for six
arrivals and with just a single deck you can have more than one craft at a
time.

Without a rotating section, the only place you can "land" incoming ships is
at the axis. You do this by spinning the incoming ship at the same rate as
the station and attach yourself somehow to the center axis.

The geometric center of each deck is at the center of rotation.

However, once you've docked, berthed, or otherwise attached yourself somehow
to the station, it's conceivable that you could then move the ship around,
say to a lower deck with "gravity". This would be similar to using an
elevator on an aircraft carrier to move aircraft from the flight deck to a
lower deck.

In article ,
Joe Strout wrote:
Has anyone already published such an arrangement? Or, have there been
other studies of very large rotating pressure seals?

The issue has been looked at a little bit, here and there, but I don't
know of anybody doing a real proper engineering study of it. The ideas I
presented are original with me as far as I know, but I couldn't swear to
that -- it's just possible I read about them somewhere in the dim past.
--
spsystems.net is temporarily off the air; | Henry Spencer
mail to henry at zoo.utoronto.ca instead. |

In article ,
Joe Strout wrote:
...sci.space.tech is a much better forum for serious technical debate.

It would be, except that the moderation time there is so long as to
effectively squelch any discussion at all. I've seen it take as long as
a week for messages to appear.

Unfortunately so. That's one reason why I've given up on it (the other
being that my postings in particular have a bad habit of just disappearing
altogether -- possibly George's spam filters are causing trouble).
--
spsystems.net is temporarily off the air; | Henry Spencer
mail to henry at zoo.utoronto.ca instead. |

I've been doing some work on seals around a 2mm shaft, and these are
watertight no problem. It could be scaled up, but the rubber component
would need to slide across a totally flat ceramic plate, which may be
problematic on a larger scale.

In general, if you can keep your corridor as narrow as possible, this
helps. You have two further problems not yet raised in this group:

1. You will need bearings further out from the seal to handle bending
moments. All the seal designs need to be kept stationary (except in
rotation) but provide limited support for bending moments. It would be
difficult to provide for bending on seal, so bearings need to ensure
that no bending moment is applied to it.

2. As well as people, you need to transfer fluids (water, air in,
carbon dioxide out), and electricity, and data. Data can probably done
with laser links between the two sections. Electricity with very large
commutators. Fluids are difficult. It could be that in the centre of
the corridor is a fluid pipe with its own seals. Indeed, it could be a
pipe (water) within a pipe (CO2) within a pipe (air/oxygen) within a
pipe (people).

Given all above, it might make sense to limit the apperture to, say, 8m
diameter. This still allows 1000s of people per hour to transfer. Large
objects could be left to air locks.

If you go to a very large structure, with a million people or so, with
10,000s tranferring per hour, you can have a radial "train" that
matches speed with the cylinder, and transfers people and cargo. Then
it matches speed with the stationary bit and does the same. This would
surround the "light axis" (where light enter) and so have a diameter of
about 1km, making it 3km long, and so capable of carrying several
thousand passenger. Fluids would be transferred as cargos.

In article .com,
Alex Terrell wrote:
2. As well as people, you need to transfer fluids (water, air in,
carbon dioxide out), and electricity, and data. Data can probably done
with laser links between the two sections. Electricity with very large
commutators. Fluids are difficult...

Data you can definitely do optically or by microwave. Power can go by
commutator, by rotary transformer (a transformer with one winding on each
side of a narrow air gap), or even by microwave beaming. Fluids are a bit
more challenging, but there are ways...

Notably, consider a car ti it has a U-shaped cross-section, and holds
air because the ends of the U are sealed to the central wheel. A tire is
fixed to its wheel, but with suitable seals, it could rotate while the
wheel was held stationary.

As I suggested earlier, build the joint between rotating and stationary
sections as a pair of concentric cylinders, say the rotating one outside
and the stationary one inside. The passageway is through the middle; the
annular space between the cylinders is where all the engineering is
located -- bearings, seals, etc. Most of the engineering hardware is
fixed to the outer cylinder, and slight pseudo-gravity is available there
because of the rotation. The stationary inner cylinder "rotates"
overhead.

Consider a trough, open side facing inward, running all the way around the
engineering space, fixed to the outer cylinder, mounted overhead on struts
so people and equipment can pass under it (it's the "tire"). Its upper
edges seal against the inner cylinder (the "wheel"). Blow air into it
from an air duct opening into its bottom, and pull air out of it through a
duct opening out of its top, and you can pass air from one section to the
other. Add another such assembly to pass air the other way. (The ducts
are *not* concentric cylinders -- just ordinary square-ish ducts at a few
places around the circumference -- so they don't have any problem going
around each other.)

Similar setups will work for water, pressurized gases, even sewage -- just
make sure the seals are good. :-)
--
spsystems.net is temporarily off the air; | Henry Spencer
mail to henry at zoo.utoronto.ca instead. |

Andrew Nowicki wrote:

You would seal the rotating joint with a hydrodynamic
bearing filled with vacuum oil (red). Grooves in the
bearing can be designed to push the oil away from the
vacuum, so that it does not leak out...

Henry Spencer wrote:

As I understand it, it's likely to be hard to make
hydrodynamic bearings work well at low rotation rates.
They rely on significant relative velocity between the
rotating and stationary surfaces.

True. Note that my drawing depicts a bearing made of 3
parts, so the angular velocity of the middle part can
be kept arbitrarily high by artificial means. Also note
the flexible corrugated pipe/passageway. The flexibility
is necessary to eliminate enormous forces due to
misalignment and catastrophic collisions. I believe that
ball bearings would be too weak for the rotating joint.

I do no know if my solution is the most elegant, but I
agree that the problem of air leaking out is not severe,
so many other solutions may be good enough.

__________________________________________________ _______


Andrew Nowicki wrote:

...sci.space.tech is a much better forum for serious
technical debate.

Joe Strout wrote:

It would be, except that the moderation time there is so
long as to effectively squelch any discussion at all.
I've seen it take as long as a week for messages to appear.

That is not a bug but a necessary feature to keep the
space cadets away.

Henry Spencer wrote:

Unfortunately so. That's one reason why I've given up on it
(the other being that my postings in particular have a bad
habit of just disappearing altogether -- possibly George's
spam filters are causing trouble).

Some of my posts disappeared as well.

Volker Hetzer wrote:

Btw, has anyone ever thought about simply ionizing the escaping
air and catching the created plasma magnetically? In order to catch
the little bits of air escaping the seals maybe it's realistic. All
you need is a radiation or heat source, right?

This is called plasma window:
http://www.islandone.org/LEOBiblio/SPBI1PW.HTM

Plasma window would be an overkill in my opinion.
It would be much cheaper to make air-tight
bearing.

Andrew Nowicki wrote:

Andrew Nowicki wrote:

...sci.space.tech is a much better forum for serious
technical debate.

Joe Strout wrote:

It would be, except that the moderation time there is so
long as to effectively squelch any discussion at all.
I've seen it take as long as a week for messages to appear.

That is not a bug but a necessary feature to keep the
space cadets away.

No, it's a bug.

D.
--
Touch-twice life. Eat. Drink. Laugh.

-Resolved: To be more temperate in my postings.
Oct 5th, 2004 JDL

"Jeff Findley" wrote in message
...


With a suit like the Orlon, with that nice door in the back, this seems
like
a good idea, but you might still need an intermediate airlock for a suit
like this, since I don't think Orlon works at sea level pressure. You'd
need a small lock that would take the pressure down to Orlon pressure
before
you opened the back of the Orlon. Still, it would be very easy to recycle
air from this intermediate lock, since you don't have to pump the pressure
down to near vacuum anymore.

Right, I was thinking Orlon specifically but genera idea... i.e. make the
suit the airlock.



Jeff
--
"They that can give up essential liberty to obtain a
little temporary safety deserve neither liberty nor
safety"
- B. Franklin, Bartlett's Familiar Quotations (1919)