> This is IMHO something more important to worry about than the
amount
> of radiation that might get in through colony walls. How much load
> can the colony walls take? How much water, dirt and buildings can
we
> put on them per square foot before chunks start pushing out. Esp
> since that dirt and water is going to go where the maximum
> acceleration from the spin is concentrated.
>
> I think that once you have answered this question the wall would
be
> pleanty thick enough, plus with the dirt and water on top, to stop
> even solar flare event radiation.

Very good question. My mind has been pondering that as well. I think
that the structural design of the habitat will have to be designed
specifically for any extraordinary structures within the the
habitat. The ground of the habitat will include dirt, small hollow
hills, lakes, forests, rivers, etc. We can probably estimate the
loading that they would place on the structure beneath them. For
multistorey buildings, all of that weight will have to be designed
for.

I've even toyed with the idea, within a cylinder habitat, of having
a building near each endcap, that has a ground floor on each of the
three land panels and each of the three buildings rise to join in
the center. The buildings would have variable gravity and join at a
minimal-g recreation center. From there, further construction could
progress along the zero-g axis. But as I've been considering this
plan, I've turned a lot of attention to the load factors that all of
that building mass would place on the ground footprint. My gut
feeling is that if these building are designed from the outset, then
the structural design of the habitat could be such as to spread the
load of the building mass.

Just think how cool it would be to go up the elevator in one
building, transfer in the zero-g lobby and go down the elevator in
another building. That lobby, having three elevator orientations,
people exiting at different angles. WOW. That would put the tourist
potential of the Empire State Building to shame.

I fear that it will be impossible to willy-nilly put up muli-storey
buildings in the habitats. Single family homes will probably be
tolerable. Office buildings - probably not.

The habitat as the ultimate planned community. When you reach plan
capacity, build a new habitat. No rezoning allowed. No increasing
densities allowed. It's built the way it's designed.

This whole issue of loading has led me to give a long hard look to
concrete construction for the habitats. Concrete can take impressive
compressive loads, but to directly addresss your question - I'm not
sure at the moment how thick the "floor" would have to be.

My seat of the pants approach for the moment is to work with the
following assumption. Six feet of concrete for the outer shell. Then
a 20 foot high service level where the mag-lev transport is, where
the light industry, shopping, offices, etc are. The interior is
reserved for habitation and parks. Above the service level is
another 4 feet of concrete/dirt/lake etc. For some areas where deep
water is required, we can cut the shielding down to the minimum
required to support the weight of the water above it, i.e. 29 feet
of water, 1 foot of structure. For parks with trees and grass, 3
feet of dirt, 1 foot of concrete support. Is that enough depth for
tree roots? For densely populated city centers, the service level is
minimized and much of that volume is consumed by thick concrete
support structures, effectively giving up to 30 feet of concrete
below the taller structures.

Hope that helps your thinking.

>
> > I think
> > that the structural design of the habitat will have to be
designed
> > specifically for any extraordinary structures within the the
> > habitat. The ground of the habitat will include dirt, small
hollow
> > hills, lakes, forests, rivers, etc. We can probably estimate the
> > loading that they would place on the structure beneath them. For
> > multistorey buildings, all of that weight will have to be
designed
> > for.
>
> Well, don't forget atmospheric pressure. 15 pounds per square
> inch is 10 tonnes per square meter before you start. Typical
> weight of stone is about 6 tonnes per cubic meter. I'd expect
> that you would want a factor of 2 safety factor on your wall
> strength and then add on other 10 tonnes of building.
>
> So the required strength might work out at 40 tonnes.
>
> Buildings would probably need foundations to even out the load.
>
> > I've even toyed with the idea, within a cylinder habitat, of
having
> > a building near each endcap, that has a ground floor on each of
the
> > three land panels and each of the three buildings rise to join
in
> > the center. The buildings would have variable gravity and join
at a
> > minimal-g recreation center. From there, further construction
could
> > progress along the zero-g axis. But as I've been considering
this
> > plan, I've turned a lot of attention to the load factors that
all of
> > that building mass would place on the ground footprint. My gut
> > feeling is that if these building are designed from the outset,
then
> > the structural design of the habitat could be such as to spread
the
> > load of the building mass.
>
> You've missed a trick. If the building goes to the axis, then
> it should hang DOWN from the axis. Tensile structures are much
> cheaper to build than compressive structures; and it avoids
> putting any load on the rim at all.
>
Cool trick if it works. I know that this is frequently cited in the
work of space tethers, space elevators, but I'm not convinced that
it would work in a habitat. Here's why: the force at work on a
planetary body is gravity caused by the mass of the the planet. On a
habitat, we're dealing with a pseudo-gravity effected through
centrifugal force. I think that would be destabilizing because of
the coriolis effect. Isn't there something to that? There really
wouldn't be a hanging down (implied that gravity is pulling as a
vector towards the surface) rather there would be a combination of
vectors to consider in the engineering.

This leads me to the larger question of how much of
standard "reference book" engineering will have to be rewritten to
substitute centrifugal force for the gravitional constant. Will such
reference tables have to be designed for each new habitat design?

>
> > Just think how cool it would be to go up the elevator in one
> > building, transfer in the zero-g lobby and go down the elevator
in
> > another building. That lobby, having three elevator
orientations,
> > people exiting at different angles. WOW. That would put the
tourist
> > potential of the Empire State Building to shame.
>
> Yes, very cool.
>
> > I fear that it will be impossible to willy-nilly put up muli-
storey
> > buildings in the habitats. Single family homes will probably be
> > tolerable. Office buildings - probably not.
>
> Actually I'm thinking that low gravity is a great place for
> um, sleeping. Other things, and sleeping. Mainly sleeping, but
> other things too. I mean, the other things are tiring. ;-)
> Plus there's the view. So the low gravity would be hotels
> and expensive residences.

Oh yeah, I know, wink wink, nudge nudge, say no more. I think that
such a building would offer opportunity for a wide variety of
experiences, sex, recreation, variable-g living environments,
preparing crews for trips to mars or the moon, recovery suites for
mitigating the prolonged stays on those planets or from zero-g
missions.

But this building (or 3 of them) will be huge. Each will be as tall
as the radius of the cylinder, i.e. 1000m - 3000m. That'll put to
shame any of earth's skyscrapers. But do we need such tall
buildings, coolness factor excluded of course. Think of how much
floor space each would have.

Also, any engineers out there want to venture an opinion on how
construction will change when the gravitational constant is no
longer constant. It will be lessening as the building height
increases. On earth, a steel girder will effectively weight the same
at ground level, 100 meters, or 2000 meters. We can't say the same
for a habitat. How can the engineering be relaxed to compensate for
the wieght of stacked masses not growing linearly but on a
diminiishing curve?
>
> > The habitat as the ultimate planned community. When you reach
plan
> > capacity, build a new habitat. No rezoning allowed. No
increasing
> > densities allowed. It's built the way it's designed.
> >
> > This whole issue of loading has led me to give a long hard look
to
> > concrete construction for the habitats. Concrete can take
impressive
> > compressive loads, but to directly addresss your question - I'm
not
> > sure at the moment how thick the "floor" would have to be.
>
> Habitats have large tensile loads; very few compressive loads.
>
Can you explain a bit more. Please.

> On a habitat, we're dealing with a pseudo-gravity effected through
> centrifugal force. I think that would be destabilizing because of
> the coriolis effect. Isn't there something to that? There really
> wouldn't be a hanging down (implied that gravity is pulling as a
> vector towards the surface) rather there would be a combination of
> vectors to consider in the engineering.
The Coriolis force is felt only by objects moving within the rotating
framework. If your building goes up and down like a yoyo, it will
experience Coriolis force. If it stands still like most buildings do,
it will feel only the tug of the centrifugal force, indistinguishable
from real gravity except that the centrifugal force will lessen with
altitude much more rapidly than gravity does.
Fast elevators may need to be designed with the Coriolis force in
mind. Also showers, perhaps.

Jon L. Beck.

> > How much water, dirt and buildings can we put on them
> > per square foot before chunks start pushing out. Esp
> > since that dirt and water is going to go where the maximum
> > acceleration from the spin is concentrated.
That of course depends on how thick and strong the hull is. One
important observation that can give you some more gut feeling (always
important!) is that the atmospheric pressure on the Earth near sea level
corresponds to ten meters of water, or about the weight of three meters
of rock.
So if you design your habitat for, say, half the atmospheric pressure
of the Earth, you'd need to add a meter and a half of rock, or more than
that of dirt, to double the load upon the hull. Since the Earth's
atmosphere is more than seventy percent inert nitrogen, you could remove
most of that and still have something easily breathable but with just
half the total pressure.

> Very good question. My mind has been pondering that as well. I think
> that the structural design of the habitat will have to be designed
> specifically for any extraordinary structures within the the
> habitat. The ground of the habitat will include dirt, small hollow
> hills, lakes, forests, rivers, etc. We can probably estimate the
> loading that they would place on the structure beneath them. For
> multistorey buildings, all of that weight will have to be designed
> for.
I agree with that. If you build a skyscraper in a space habitat,
and the "ground" just there weren't strengthened for that, you'll have
one skyscraper rocketing out of a hole in the hull, taking a lot of
people and atmosphere with it...

> I've even toyed with the idea, within a cylinder habitat, of having
> a building near each endcap, that has a ground floor on each of the
> three land panels and each of the three buildings rise to join in
> the center. The buildings would have variable gravity and join at a
> minimal-g recreation center. From there, further construction could
> progress along the zero-g axis. But as I've been considering this
> plan, I've turned a lot of attention to the load factors that all of
> that building mass would place on the ground footprint. My gut
> feeling is that if these building are designed from the outset, then
> the structural design of the habitat could be such as to spread the
> load of the building mass.
If you build that building such that it supports itself, you won't
have to face that problem to a very high degree. Say you build an
axial-symmetric building, consisting of several cylindrical floors.
They would not need to rest heavily upon the "ground". They would only
have to built together with the "ground" so that they wouldn't rattle
around against it.
Still, I'd like at least one end cap as not a building, but a sloping
park. As the slope increases, weight is reduced, so even when the slope
goes very steeply upwards, the hike up won't be as difficult as it would
on Earth.
Of course, by the time the slope is 45 degrees, your weight is still
reduced only by less than thirty percent.

> This whole issue of loading has led me to give a long hard look to
> concrete construction for the habitats. Concrete can take impressive
> compressive loads, but to directly addresss your question - I'm not
> sure at the moment how thick the "floor" would have to be.
As has been mentioned, the habitat hull would have to take tensile
loads. A concrete foundation can withstand high loads that seek to
crush it. A habitat hull will have to withstand loads that seek to rip
it apart.

> My seat of the pants approach for the moment is to work with the
> following assumption. Six feet of concrete for the outer shell. Then
> a 20 foot high service level where the mag-lev transport is, where
> the light industry, shopping, offices, etc are. The interior is
> reserved for habitation and parks. Above the service level is
> another 4 feet of concrete/dirt/lake etc. For some areas where deep
> water is required, we can cut the shielding down to the minimum
> required to support the weight of the water above it, i.e. 29 feet
> of water, 1 foot of structure. For parks with trees and grass, 3
> feet of dirt, 1 foot of concrete support. Is that enough depth for
> tree roots? For densely populated city centers, the service level is
> minimized and much of that volume is consumed by thick concrete
> support structures, effectively giving up to 30 feet of concrete
> below the taller structures.
The waste heat radiators need have approximately the same area as the
land that is illuminated by sunlight. I made the calcs on the
sci.space.policy NG some time ago. It agrees with my gut feeling, of
course: the Earth radiates from roughly (within an order of magnitude)
the same area as it receives sunlight on. But if you have a certain
land area that receives about 800 W per square meter in the habitat's
daytime (this corresponds to a warm day on the Earth), and you reradiate
this heat from waste heat radiators that are about 25 degrees Celsius,
24 hours a day, and they are the same area as that certain land area,
then those radiators will be enough.
Where do you put the waste heat radiators?
If you simply pump air from the habitat to external coolers, these
coolers need be thick-walled enough to withstand the same air pressure
as the habitat hull. Or you can somehow make the hull itself the waste
heat radiator.

Also consider this: assume a cylindrical habitat with a single metal
hull, and a meter of soil on top. Assume the cylinder axis is pointed
towards the Sun, so that the hull, except for one endcap if that endcap
isn't in the shadow of some other structures, is in perpetual darkness.
Then the outer surface of the hull will be cold indeed. The bottom
layer of soil will be about one degree warmer than the outer surface.
Near the bottom of the soil you will have dry ice temperature.
Twenty centimeters below the topsoil you will have permafrost. Not a
good idea if you try for a habitat with subtropical flora and fauna.
Three options present themselves to prevent the land from becoming
summertime tundra: heat the outer surface of the hull (such as by
shining sunlight on it), add an insulating layer below the soil, or make
a double hull. If at least the outer hull is of metal, you can then
remove waste heat from the hab by pumping air through the interstitial
between the inner and outer hull. I never calculated the approximate
necessary airflow, but I did calculate that if the airflow is big
enough, this would be enough to remove all the waste heat from, say, and
Island Three type cylinder.
Probably you would want to pump air from near the axis at one end
cap, through the interstitial, and out near the axis at the opposite end
cap.
Also useful would be the fact that the air, cooled by heat loss and
then further cooled adiabatically as it rises towards the outlet near
the axis, would precipitate some of its moisture. This could be the
main source of potable water in the colony. Of course, any cooling
surfaces provide this.
The combined thickness of the double hull would not be much different
from the thickness of a single hull. A single hull would need to
contain the load from both the air pressure and the weight of the soil
and structures. With a double hull, the outer hull would contain the
air pressure and the inner hull the weight of soil and structures. Air
pressure and weight would be vectors in the same direction, and hence
can be added just like real numbers.

Of course in the interstitial between the hulls you could build
structures, such as subways and offices, as you proposed above, at least
if the outer hull is thick enough to shield against radiation. And if
it needs to contain air pressure, and the inner hull only needs to
contain a meter of soil and some buildings, the outer hull will be the
thickest one anyway.

Jon L. Beck.

Your well detailed answer to this post was very much appreciated.

> If you build that building such that it supports itself, you
won't
> have to face that problem to a very high degree. Say you build an
> axial-symmetric building, consisting of several cylindrical floors.
> They would not need to rest heavily upon the "ground". They would
only
> have to built together with the "ground" so that they wouldn't
rattle
> around against it.

I don't understand. Would you explain in more detail. Of course, the
building can be situated anywhere along the gradient of the endcap,
but I thought it might be interesting to place it at the boundary of
the cylinder and endcap or even midway between the endcaps, at the
halfway point in the cylinder (but it would sure ruin the vistas.)
Therefore, this building would have an immense height (the radius of
the cylinder - Island 3?)

> Still, I'd like at least one end cap as not a building, but a
sloping
> park. As the slope increases, weight is reduced, so even when the
slope
> goes very steeply upwards, the hike up won't be as difficult as it
would
> on Earth.
> Of course, by the time the slope is 45 degrees, your weight is
still
> reduced only by less than thirty percent.

I agree that the endcaps shouldn't be cluttered with buildings.

>
> > This whole issue of loading has led me to give a long hard look
to
> > concrete construction for the habitats. Concrete can take
impressive
> > compressive loads, but to directly addresss your question - I'm
not
> > sure at the moment how thick the "floor" would have to be.
> As has been mentioned, the habitat hull would have to take
tensile
> loads. A concrete foundation can withstand high loads that seek to
> crush it. A habitat hull will have to withstand loads that seek
to rip
> it apart.
>
> > My seat of the pants approach for the moment is to work with the
> > following assumption. Six feet of concrete for the outer shell.
Then
> > a 20 foot high service level where the mag-lev transport is,
where
> > the light industry, shopping, offices, etc are. The interior is
> > reserved for habitation and parks. Above the service level is
> > another 4 feet of concrete/dirt/lake etc. For some areas where
deep
> > water is required, we can cut the shielding down to the minimum
> > required to support the weight of the water above it, i.e. 29
feet
> > of water, 1 foot of structure. For parks with trees and grass, 3
> > feet of dirt, 1 foot of concrete support. Is that enough depth
for
> > tree roots? For densely populated city centers, the service
level is
> > minimized and much of that volume is consumed by thick concrete
> > support structures, effectively giving up to 30 feet of concrete
> > below the taller structures.
> The waste heat radiators need have approximately the same area
as the
> land that is illuminated by sunlight. I made the calcs on the
> sci.space.policy NG some time ago. It agrees with my gut feeling,
of
> course: the Earth radiates from roughly (within an order of
magnitude)
> the same area as it receives sunlight on. But if you have a
certain
> land area that receives about 800 W per square meter in the
habitat's
> daytime (this corresponds to a warm day on the Earth), and you
reradiate
> this heat from waste heat radiators that are about 25 degrees
Celsius,
> 24 hours a day, and they are the same area as that certain land
area,
> then those radiators will be enough.
> Where do you put the waste heat radiators?
> If you simply pump air from the habitat to external coolers,
these
> coolers need be thick-walled enough to withstand the same air
pressure
> as the habitat hull. Or you can somehow make the hull itself the
waste
> heat radiator.

How do heat pumps work? Don't they extract heat from the air and
can't they concentrate the heat extracted from multiple cycles. I'm
really extending onto weak ice here, but couldn't the heat from
great volumes of air be extracted, multiplyed and transferred to a
liquid, then pump that hot liquid to the cool side of the habitat
and the extreme range of temperature will expedite the radiation of
the heat into the cold of space. Does heat radiate at a uniform rate
or does it radiate more quickly depending on the difference in the
temperatures of the two mediums?

If the hull was used, would you transfer the hottest air to the
portion of the wall that faced space and then move the remaining
residual heat to a portion of the hull more inward, conduct the heat
into the material, then take the remaining residual heat and move it
further inward?

Or should the radiator be external to the habitat? How closely could
you pack the radiator surfaces to each other to allow them to
radiate efficiently in the smallest amount of volume?

>
> Also consider this: assume a cylindrical habitat with a single
metal
> hull, and a meter of soil on top. Assume the cylinder axis is
pointed
> towards the Sun,

This presumes two habitats that are tethered together, rotating in
opposite directions, doesn't it. My understanding is that one
habitat can precess with the sun.

so that the hull, except for one endcap if that endcap
> isn't in the shadow of some other structures, is in perpetual
darkness.
> Then the outer surface of the hull will be cold indeed. The
bottom
> layer of soil will be about one degree warmer than the outer
surface.
> Near the bottom of the soil you will have dry ice temperature.

Yeah, we'd want to avoid that.

> Twenty centimeters below the topsoil you will have permafrost.
Not a
> good idea if you try for a habitat with subtropical flora and
fauna.
> Three options present themselves to prevent the land from
becoming
> summertime tundra: heat the outer surface of the hull (such as by
> shining sunlight on it), add an insulating layer below the soil,
or make
> a double hull. If at least the outer hull is of metal, you can
then
> remove waste heat from the hab by pumping air through the
interstitial
> between the inner and outer hull. I never calculated the
approximate
> necessary airflow, but I did calculate that if the airflow is big
> enough, this would be enough to remove all the waste heat from,
say, and
> Island Three type cylinder.
> Probably you would want to pump air from near the axis at one
end
> cap, through the interstitial, and out near the axis at the
opposite end
> cap.
> Also useful would be the fact that the air, cooled by heat loss
and
> then further cooled adiabatically as it rises towards the outlet
near
> the axis, would precipitate some of its moisture. This could be
the
> main source of potable water in the colony. Of course, any cooling
> surfaces provide this.
> The combined thickness of the double hull would not be much
different
> from the thickness of a single hull. A single hull would need to
> contain the load from both the air pressure and the weight of the
soil
> and structures. With a double hull, the outer hull would contain
the
> air pressure and the inner hull the weight of soil and structures.

Am I correct in understanding that with the outer hull be structural
designed to contain the air presure, the innder hull could be
designed with a number of openings to allow people to move between
the inner surface and the subterranian level. What would support the
inner hull? Wouldn't all of that mass be resting upon the outer
hull? Wouldn't the mass of dirt, lakes, houses, 5 storey building,
50 storey building and the unique 1,000 storey building, indirectly
rest upon the structure of the outer hull?

Air
> pressure and weight would be vectors in the same direction, and
hence
> can be added just like real numbers.
>
> Of course in the interstitial between the hulls you could build
> structures, such as subways and offices, as you proposed above, at
least
> if the outer hull is thick enough to shield against radiation.
And if
> it needs to contain air pressure, and the inner hull only needs to
> contain a meter of soil and some buildings, the outer hull will be
the

> > If you build that building such that it supports itself, you
> > won't have to face that problem to a very high degree. Say
> > you build an axial-symmetric building, consisting of several
> > cylindrical floors. They would not need to rest heavily upon
> > the "ground". They would only have to built together with
> > the "ground" so that they wouldn't rattle around against it.

> I don't understand. Would you explain in more detail. Of course, the
> building can be situated anywhere along the gradient of the endcap,
> but I thought it might be interesting to place it at the boundary of
> the cylinder and endcap or even midway between the endcaps, at the
> halfway point in the cylinder (but it would sure ruin the vistas.)
> Therefore, this building would have an immense height (the radius of
> the cylinder - Island 3?)
Imagine a rotating hab. The hull supports itself, the atmospheric
pressure and soil and structures built on the soil.
Then imagine a ring, coaxial with the hab but with a smaller
diameter, and not touching the soil or the hull at all, and rotating at
the same rate as the hull. Like the hull, this ring would support
itself, and it would not add to the load upon the hull. Of course, if
this ring isn't attached in any way to the hull at all, it will drift
until it hits the hull.
If you break the ring at some point, it will fly apart until bits of
it hits the hull. But of course, like a sensible engineer, you design
the ring to withstand any load that you care to place on it.
A building like you imagine could be built as several concentric such
rings. These rings would be the floors. They would of course be
*attached* to the hull and to each other, but they would not add
significant load to each other nor to the hull. Only a floor or floor
section that is not a complete ring would have to be placed upon the
hull and add to its load, unless you use tethers or long rods to keep
opposite sections of the incomplete ring in place.

> I agree that the endcaps shouldn't be cluttered with buildings.
Of course in a Lagrangia society with hundreds of habitats, they will
not have to be all alike. :-)

> How do heat pumps work? Don't they extract heat from the air and
> can't they concentrate the heat extracted from multiple cycles. I'm
> really extending onto weak ice here, but couldn't the heat from
> great volumes of air be extracted, multiplyed and transferred to a
> liquid, then pump that hot liquid to the cool side of the habitat
> and the extreme range of temperature will expedite the radiation of
> the heat into the cold of space. Does heat radiate at a uniform rate
> or does it radiate more quickly depending on the difference in the
> temperatures of the two mediums?
I suppose heat pumps could work. You compress atmospheric air from
the habitat and it heats up adiabatically. You then transfer this heat
to a fluid that circulates out to the radiators. Alternatively, you
could go without the fluid and simply pass the compressed air through
the radiators. Since the intensity of thermal radiation scales as the
fourth power of the surface temperature, if you double the temperature
from 300 K (which is a pleasant temperature, suitable for a tropical or
subtropical hab) to 600 K (which is hotter than my stove when I bake
bread, though cooler than a lighter flame), you would need one sixteenth
of the radiative surface area to radiate away the same power. Of course
there would be extra heat to be radiated away, because the heat pumps
themselves would generate waste heat. The radiators would therefore
have to be somewhat larger than one sixteenth of the simple design.
If you want to minimize the area of the waste heat radiators, heat
pumps is the way to go, or so I suppose without having done any calcs.
The downside is increased complexity: extra machinery, and the
requirement to handle large volumes of air at rather high pressures.
The simple approach with circulating the air directly to the waste
heat radiators without any compression has of course the benefit that it
is simple (d'oh). But you would either have to build a very large
external radiator, or use the outer hull as a radiator - and this would
preclude using a non-rotating shell of industrial slag as radiation
shielding.
If it turns out with that practical experience that we don't have yet
that metal is fairly easily extracted from lunar or asteroidal material,
and that vacuum depositing of metal vapour is easy, then I suppose that
this will be the favoured design in habitat building. These are of
course two assumptions unfounded in experience.

> If the hull was used, would you transfer the hottest air to the
> portion of the wall that faced space and then move the remaining
> residual heat to a portion of the hull more inward, conduct the heat
> into the material, then take the remaining residual heat and move it
> further inward?
I would pass air through the interstitial. This air would leave the
main volume of the hab at perhaps 25 degrees Celsius, cool down to
perhaps 20 degrees Celsius, and be reintroduced to the main volume at
that temperature. So the outer surface of the *inner* hull wouldn't get
below 20 degrees Celsius, while the topsoil inside the inner hull would
experience normal diurnal variations.
The inner hull would thus be sandwiched between air masses that both
have pleasant temperatures.
What I have not calculated is the rate of air circulation that would
be necessary to remove 800 W/square meter in daytime - averaging perhaps
500 W/square meter over a day and a night, if you want more than twelve
hours of daylight - with this particular temperature change. It should
be straightforward, given the heat capacity of air.
If you reduce the air flow, the outer surface of the outer hull will
become colder because less heat is tranferred to it by the air; what air
passes will be cooled down more than by five degrees, and you would have
lesser volumes of cooled air reintroduced to the main hab, but this air
would be colder. Yet the average temperature within the hab would be
higher, because less heat would be radiated away until the air that is
introduced to the interstitial is warmer than those 25 degrees Celsius.

> Or should the radiator be external to the habitat? How closely could
> you pack the radiator surfaces to each other to allow them to
> radiate efficiently in the smallest amount of volume?
I don't think you need to minimize volume in this particular case.
In your computer, the cooling ribs of your CPU cooler of course need to
take up as small a volume as possible. They lose heat mainly through
conduction to air that flows past them, and it doesn't matter that they
face each other in a very compact manner, as long as air can flow
unimpededly. But a purely radiative cooling surface needs to face the
cold of space, not other cooling surfaces or other warm objects.

> This presumes two habitats that are tethered together, rotating in
> opposite directions, doesn't it. My understanding is that one
> habitat can precess with the sun.
One rotating habitat would keep its axis of rotation fixed, or
precessing only very slowly, unless you force it to precess by
continously firing some sort of thrusters. Which is not acceptable.
You either need to pair them up or pipe light in along the axis with
mirrors.

> Am I correct in understanding that with the outer hull be structural
> designed to contain the air presure, the innder hull could be
> designed with a number of openings to allow people to move between
> the inner surface and the subterranian level. What would support the
> inner hull? Wouldn't all of that mass be resting upon the outer
> hull? Wouldn't the mass of dirt, lakes, houses, 5 storey building,
> 50 storey building and the unique 1,000 storey building, indirectly
> rest upon the structure of the outer hull?
Like I mentioned above with the concentric rings example, the inner
hull would support itself. Beyond being attached to the outer hull so
that the two hulls don't drift relative to each other, it would place no
load upon the outer hull. Only if the inner hull breaks will the load
that it supported until then hit the outer hull.
But since the inner hull would not need to contain air pressure,
nothing prevents it from having openings like you mention. Such
openings would be desirable even if nothing in particular is put into
the interstitial, because easy access to the interstitial for inspection
and maintenance would be a Very Good Idea(tm).

There is one more benefit to the double hull design, with perhaps one
meter thickness of the outer hull to provide shielding as much as to
provide structural strength, and at least room enough in the
interstitial for a several-storey building. If the outer hull is
breached by a large impact, repairs can be made much more efficiently
than if you have a breach in a single hull. In the former case you may
slide a seal or dome in place over the hole - even if the inner hull is
*also* breached - before the atmosphere is all vented from the habitat.
In the latter case you'd have to dig away the soil around the hole first
to make the seal airtight... And an impactor that breaches the outer
hull will likely be partly vapourized, and strike the inner hull with
less force than if the "outer hull" and "inner hull" were built together
as one single hull. I know that the Giotto probe had a dust shield with
this design: a double plate, where the dust from the comet impacted the
outer plate, and what hit the inner plate was fast puffs of vapour.

Jon L. Beck.