Author Topic: Power options for a Mars settlement  (Read 136174 times)

Offline lamontagne

  • Full Member
  • ****
  • Posts: 1484
  • Liked: 1959
  • Likes Given: 314
Re: Power options for a Mars settlement
« Reply #560 on: 04/29/2018 11:08 pm »
This may be difficult in terms of agreements with NASA, even on Mars... But a Kilopower without its Stirling engines, or power generation could use extended heat pipes as a direct water mining tool. Maybe the unit will need too much shielding to make it useable, but conversely secondary heat pipes may possibly be made 10's of metres long to provide heat for mining water at a distance from the reactor itself. This is using the power directly without all that electric mumbo-jumbo....
Mirrors could also be set up to heat a cathode using concentrated sunlight, and a high capacity heat pipe used to transfer this heat energy to mine large quantities of water.
Others have suggested habitats in ice caverns so made.
It would be something of a waste not to produce electricity.  Kilo power is likely no more than ...30% efficient? so it already produces  70% of its output as heat.  And the electricity will be used to light plants, and 95% of that 30% will also turn into heat.  That should also be recoverable to melt ice.




Online Robotbeat

  • Senior Member
  • *****
  • Posts: 28825
  • Minnesota
  • Liked: 8975
  • Likes Given: 5773
Re: Power options for a Mars settlement
« Reply #561 on: 04/29/2018 11:16 pm »
Yall are still missing the point: energy for mining is tiny compared to the energy needed for electrolyzing that into fuel and oxygen.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Joseph Peterson

  • Full Member
  • *
  • Posts: 180
  • Pittsburgh, PA
  • Liked: 110
  • Likes Given: 1501
Re: Power options for a Mars settlement
« Reply #562 on: 04/30/2018 08:08 am »

SNIP

You get three per synod if you can turn off the methane generation during the night.

SNIP


Electrolysis, not methane generation.  Methane generation can continue as long as there is enough hydrogen left in the buffer.  The hydrogen buffer is there to allow steady state operations, reducing catalyst degradation.

The whole process from whatever it takes to gather resources on. If you can modulate down to match insolation, without peak/averages costing you more than 1/3 in terms of mass, means you can mostly skip batteries.

I was being less careful in my wording as I was correcting a three orders of magnitude implication (500MW vs  around 500kW) for near-term power use.

It's easy to make a minor mistake on a secondary point.  I do it far more often than I'd like.

1/3 of what in terms of mass? 

If I'm doing my math right, we need 77 kg of hydrogen per day to refuel one BFS per synod.  If we can run on battery regulated electrolysis 50% of the day, we need approximately 1 tonne of hydrogen buffer mass.  We'll need the fudge factor to account for maintenance, dust storms, and the like, so 5 tonnes is a more realistic number.  Also needed are the sun shield, waste heat disposal system(potentially heating water prior to electrolysis), and the wiring harnesses needed to connect and power the sensor net and control system.

Hopefully this is enough for speculation purposes.  I'm nowhere near considering mass optimization.  This is an extrapolation from my don't-blow-myself-up proof-of-concept design work.
I hate feeling like I'm playing chess when everyone else is playing checkers.  Reality proving my feelings are correct is even worse.

Offline speedevil

  • Senior Member
  • *****
  • Posts: 3225
  • Fife
  • Liked: 1674
  • Likes Given: 1998
Re: Power options for a Mars settlement
« Reply #563 on: 04/30/2018 09:14 am »
The whole process from whatever it takes to gather resources on. If you can modulate down to match insolation, without peak/averages costing you more than 1/3 in terms of mass, means you can mostly skip batteries.

I was being less careful in my wording as I was correcting a three orders of magnitude implication (500MW vs  around 500kW) for near-term power use.

It's easy to make a minor mistake on a secondary point.  I do it far more often than I'd like.

1/3 of what in terms of mass? 

Total landed mass.
If your extraction hardware (whatever is the hardest part to scale) weighs more as it has to deal with instantaneous solar power, not average daily power from the solar panels, then batteries plus continual extraction hardware may be lighter.

From the ebay thread, you can fit around 750kW into 150 tons, or 500kW with batteries to smooth it over the whole day.

Offline Joseph Peterson

  • Full Member
  • *
  • Posts: 180
  • Pittsburgh, PA
  • Liked: 110
  • Likes Given: 1501
Re: Power options for a Mars settlement
« Reply #564 on: 04/30/2018 07:19 pm »

SNIP

1/3 of what in terms of mass? 

Total landed mass.
If your extraction hardware (whatever is the hardest part to scale) weighs more as it has to deal with instantaneous solar power, not average daily power from the solar panels, then batteries plus continual extraction hardware may be lighter.

From the ebay thread, you can fit around 750kW into 150 tons, or 500kW with batteries to smooth it over the whole day.

I'm not understanding the logic of using total landing mass.  I'll do a quick estimate even though I don't yet see the point.

We'll use Robobeat's 7.5kWh per kg of water as our placeholder.  Hydrogen is close enough to 1/9th of the total mass of water.  To produce 77 kg of hydrogen per day we will need ~5200kWh.  Assuming our panels are producing at the rated capacity for 12 hours each day, we need 430kW of generation.  Using your 5kW/t figure we need roughly 86t of solar panels.  Since our panels won't be producing at the rated capacity for 12 hours each day we should use the 1t hydrogen buffer mass as our mass to compare.  We're already down to 1/87th of the landed mass and haven't accounted for the Sabatier reactor, electrolysis system, water collection and processing, or the assorted little bits.  I'm not going to estimate the rest of the system mass.  Keeping the buffer under 1/3 of landed mass is no problem.

I would think that a better mass to compare is the mass of batteries needed to run all night.  Using the same basic assumptions we'll need  ~16t of batteries to replace the 1t hydrogen buffer.  We don't need to fiddle with fractions because this is basically a like for like replacement.  The hydrogen buffer is clearly superior to batteries from a landed mass perspective.

The $64B question is, is the hydrogen buffer preferable to replacement catalysts and the tooling to refurbish the Sabatier reactor?  The best answer I can give you at this point is maybe.  In theory the buffer is better mass-wise but we all know what happens to theory once real world testing begins in earnest.  If I can minimize startup degradation so that the reactor can be restarted 100+ times, we can potentially end up with total landed mass savings. 

We might not be looking at actual saving though.  How much landed mass is one person-hour of Martian labor worth?  We are only going to have so many workers on Mars.  It could easily be worthwhile to send an extra tonne so human labor can do other things.
I hate feeling like I'm playing chess when everyone else is playing checkers.  Reality proving my feelings are correct is even worse.

Offline speedevil

  • Senior Member
  • *****
  • Posts: 3225
  • Fife
  • Liked: 1674
  • Likes Given: 1998
Re: Power options for a Mars settlement
« Reply #565 on: 04/30/2018 08:47 pm »
I'm not understanding the logic of using total landing mass.  <snip>

We might not be looking at actual saving though.  How much landed mass is one person-hour of Martian labor worth?  We are only going to have so many workers on Mars.  It could easily be worthwhile to send an extra tonne so human labor can do other things.

I have not carefully read the above post, and need to, but as a more general point, what else would you be optimising for?
Surely the goal is to get capabilities on Mars, and the lighter you can do this (if it does not make it too much more expensive), the more capabilities you can fit in 150 tons. (modulo density limits).

To start with, at least, those initial capabilities would be improvements in landing safety, life support capacity and the capacity to manufacture propellant, and maintenance,but what you want after the next synod is an interesting question.


'What you should optimise for' is a fun question for subsequent synods, at least initially it's fairly simple.

Offline DistantTemple

  • Full Member
  • ****
  • Posts: 426
  • England
  • Liked: 325
  • Likes Given: 514
Re: Power options for a Mars settlement
« Reply #566 on: 04/30/2018 09:10 pm »
I'm not understanding the logic of using total landing mass.  <snip>

We might not be looking at actual saving though.  How much landed mass is one person-hour of Martian labor worth?  We are only going to have so many workers on Mars.  It could easily be worthwhile to send an extra tonne so human labor can do other things.

I have not carefully read the above post, and need to, but as a more general point, what else would you be optimising for?
Surely the goal is to get capabilities on Mars, and the lighter you can do this (if it does not make it too much more expensive), the more capabilities you can fit in 150 tons. (modulo density limits).

To start with, at least, those initial capabilities would be improvements in landing safety, life support capacity and the capacity to manufacture propellant, and maintenance,but what you want after the next synod is an interesting question.


'What you should optimise for' is a fun question for subsequent synods, at least initially it's fairly simple.
I think one thing to optimise for especially early on is:

  "likely-to-work-out-of-the-box-first-time and continue-reliably-with-no-input-or-servicing and overall-minimal-manpower-input"

 So a battery that just works, period... despite its mass, is better than something else that requires development and tending! Especially as all activity is compromised by power problems. (However even lots of batteries eventually run out so they are only part of a solution)

This as we all frequently repeat is very much one of EM's principals. (The BFS is not optimal mass for every mission... it is optimal amortization of design and manufacturing costs, to reduce individual mission cost, and so increase overall transport capability, (and hasten HSF to Mars) )
We can always grow new new dendrites. Reach out and make connections and your world will burst with new insights. Then repose in consciousness.

Offline Joseph Peterson

  • Full Member
  • *
  • Posts: 180
  • Pittsburgh, PA
  • Liked: 110
  • Likes Given: 1501
Re: Power options for a Mars settlement
« Reply #567 on: 05/01/2018 03:12 am »
I'm not understanding the logic of using total landing mass.  <snip>

We might not be looking at actual saving though.  How much landed mass is one person-hour of Martian labor worth?  We are only going to have so many workers on Mars.  It could easily be worthwhile to send an extra tonne so human labor can do other things.

I have not carefully read the above post, and need to, but as a more general point, what else would you be optimising for?
Surely the goal is to get capabilities on Mars, and the lighter you can do this (if it does not make it too much more expensive), the more capabilities you can fit in 150 tons. (modulo density limits).

To start with, at least, those initial capabilities would be improvements in landing safety, life support capacity and the capacity to manufacture propellant, and maintenance,but what you want after the next synod is an interesting question.


'What you should optimise for' is a fun question for subsequent synods, at least initially it's fairly simple.

I am optimizing for the terrestrial market.  That means low costs, specifically maintenance costs, are critical.  I have to make the money to get to Ceres* somehow and there is free renewable electricity being curtailed.  In some cases, people are even paying others to take electricity away.  This could result in a carbon neutral fracking replacement at a lower price.

My system will eventually be mass optimized.  That comes after proving theory works in practice.  My reaction chamber is currently at least 4 times what it needs to mass for terrestrial use.  The reasoning is that if I find yet another obvious-but-somehow-unpatented way to increase operating pressure without increasing costs I don't have to redesign the entire test rig.  The hydrogen buffer is modular, and since I plan to start testing at 1.5 bar, knowing that its current mass fraction is a fraction of a percent doesn't add anything to this thread. 

What is really important for this thread is a ballpark estimate of the hydrogen buffer mass.  5 tonnes per BFS, plus sun shield and connections, is enough to determine whether a hydrogen buffer or batteries are preferable for continuous operation.  Whether continuous operation is the right choice is a fact I have yet to grok in fullness.

* My goal is to start Ceres Organic Chemicals so I can provide things like fertilizer to the farmers, who are feeding the shovel makers.
I hate feeling like I'm playing chess when everyone else is playing checkers.  Reality proving my feelings are correct is even worse.

Offline Joseph Peterson

  • Full Member
  • *
  • Posts: 180
  • Pittsburgh, PA
  • Liked: 110
  • Likes Given: 1501
Re: Power options for a Mars settlement
« Reply #568 on: 05/01/2018 04:15 am »
I'm not understanding the logic of using total landing mass.  <snip>

We might not be looking at actual saving though.  How much landed mass is one person-hour of Martian labor worth?  We are only going to have so many workers on Mars.  It could easily be worthwhile to send an extra tonne so human labor can do other things.

I have not carefully read the above post, and need to, but as a more general point, what else would you be optimising for?
Surely the goal is to get capabilities on Mars, and the lighter you can do this (if it does not make it too much more expensive), the more capabilities you can fit in 150 tons. (modulo density limits).

To start with, at least, those initial capabilities would be improvements in landing safety, life support capacity and the capacity to manufacture propellant, and maintenance,but what you want after the next synod is an interesting question.


'What you should optimise for' is a fun question for subsequent synods, at least initially it's fairly simple.
I think one thing to optimise for especially early on is:

  "likely-to-work-out-of-the-box-first-time and continue-reliably-with-no-input-or-servicing and overall-minimal-manpower-input"

 So a battery that just works, period... despite its mass, is better than something else that requires development and tending! Especially as all activity is compromised by power problems. (However even lots of batteries eventually run out so they are only part of a solution)

This as we all frequently repeat is very much one of EM's principals. (The BFS is not optimal mass for every mission... it is optimal amortization of design and manufacturing costs, to reduce individual mission cost, and so increase overall transport capability, (and hasten HSF to Mars) )

Agree and disagree but we are drifting toward off-topic so I will keep this short.  If we save enough mass using a repairable system that we afford to send two extra people, having a person to repair the system is preferable.

Perhaps a thread discussing the value of human labor on Mars is in order.
I hate feeling like I'm playing chess when everyone else is playing checkers.  Reality proving my feelings are correct is even worse.

Offline speedevil

  • Senior Member
  • *****
  • Posts: 3225
  • Fife
  • Liked: 1674
  • Likes Given: 1998
Re: Power options for a Mars settlement
« Reply #569 on: 05/01/2018 10:03 am »
I am optimizing for the terrestrial market.  That means low costs, specifically maintenance costs, are critical.  I have to make the money to get to Ceres* somehow and there is free renewable electricity being curtailed.  In some cases, people are even paying others to take electricity away.  This could result in a carbon neutral fracking replacement at a lower price.

My system will eventually be mass optimized.
Ah!
That makes sense.
My above posts were in the context of near-term mass trades on how much you can pack onto BFS, trying to base on current commercially available hardware as a realistic floor.

This makes costs to a first degree irrelevant. (for initial missions if they're under $1000/kg or so)

Offline Joseph Peterson

  • Full Member
  • *
  • Posts: 180
  • Pittsburgh, PA
  • Liked: 110
  • Likes Given: 1501
Re: Power options for a Mars settlement
« Reply #570 on: 05/01/2018 10:45 pm »
I am optimizing for the terrestrial market.  That means low costs, specifically maintenance costs, are critical.  I have to make the money to get to Ceres* somehow and there is free renewable electricity being curtailed.  In some cases, people are even paying others to take electricity away.  This could result in a carbon neutral fracking replacement at a lower price.

My system will eventually be mass optimized.
Ah!
That makes sense.
My above posts were in the context of near-term mass trades on how much you can pack onto BFS, trying to base on current commercially available hardware as a realistic floor.

This makes costs to a first degree irrelevant. (for initial missions if they're under $1000/kg or so)

My posts are in the context of near-term mass trades based on the commercially available components I am using to build my prototype.  If the University of Michigan thermal conductive plastic was commercially available then I would have lowered my hydrogen buffer mass estimate by whatever mass savings that material allows.  The same applies for any unobtainium you want to suggest.

I'm not sure how to make this clear without posting the design work I am planning on patenting.  The hydrogen buffer is a subsystem.  The plumbing for the subsystem is designed for the prototype subsystem because hydrogen storage is expensive.  I simply can't give you a good mass fraction because it is other subsystems, and the plumbing that ties them together, that are designed to go from the initial sub-kg/day testing at 1.5 bar to kg/hr tests at 50 bar(currently requires unobtainium).  The entire system is also designed so that any force from explosions is directed away from people and expensive hardware.  It is literally a test rig.  What I can do, and have already done, is to provide a mass estimate for a hydrogen buffer subsystem scaled to 77 kg/day without the excessive shielding.  I could reduce mass by doing things like replacing steel brackets with titanium, but this will push the cost above your $1000/kg or so(What is this based on?) figure.

Because I am designing using modular subsystems it is really easy to make macroeconomic comparisons once we have a buffer mass estimate.  Installing the buffer in an existing Sabatier system design is as simple as cutting the pipe between the electrolysis subsystem and the Sabatier reactor subsystem, then installing a pair of fittings(on the order of a kg or less, so insignificant).  We don't need to know the entire system mass to determine whether a hydrogen buffer makes sense.  All we need to compare are the subsystems that can replace the functionality of the buffer.  What we should be comparing is 1 tonne of hydrogen buffer to 16 tonnes of batteries or ?? tonnes of tooling, spare parts, and mass to support the worker who is refurbishing Sabatier reactors.  These options replace the functionality of the hydrogen buffer.  Knowing the total mass of solar panels needed is irrelevant at this level of detail.

Note: I haven't touched on storage losses.  While storage losses will be far greater using batteries before electrolysis(no need to store the energy lost in electrolysis) than hydrogen buffers, batteries already look bad enough.
I hate feeling like I'm playing chess when everyone else is playing checkers.  Reality proving my feelings are correct is even worse.

Offline LMT

  • Lake Matthew Team
  • Full Member
  • ****
  • Posts: 711
    • Lake Matthew
  • Liked: 92
  • Likes Given: 0
Re: Power options for a Mars settlement
« Reply #571 on: 05/07/2018 03:31 am »
Some recent Mars ISRU power and production numbers, in Mars ISRU:  State-of-the-Art and System Level Considerations

See esp.:  "Mars ISRU Pathfinder Demo Payload Options"

Production:  .48 kg/hr O2 & .12 kg/hr CH4 using 5.75 kW (ISRU power only, excluding other systems)

Application example:  Operating continuously without losses, that ISRU plant would fill one pair of SpaceX ITS tanks, using 41 TJ, over 228 Earth years.
« Last Edit: 05/07/2018 03:35 am by LMT »

Online Robotbeat

  • Senior Member
  • *****
  • Posts: 28825
  • Minnesota
  • Liked: 8975
  • Likes Given: 5773
Re: Power options for a Mars settlement
« Reply #572 on: 05/07/2018 04:41 am »
The ISRU pathfinder is not terribly efficient. But right order of magnitude. Gonna need 100 times that amount of power.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline LMT

  • Lake Matthew Team
  • Full Member
  • ****
  • Posts: 711
    • Lake Matthew
  • Liked: 92
  • Likes Given: 0
Re: Power options for a Mars settlement
« Reply #573 on: 05/07/2018 11:56 am »
The ISRU pathfinder is not terribly efficient.

Which design changes might best improve efficiency, quantitatively?

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6250
  • Liked: 4116
  • Likes Given: 5646
Re: Power options for a Mars settlement
« Reply #574 on: 05/07/2018 01:07 pm »
Some recent Mars ISRU power and production numbers, in Mars ISRU:  State-of-the-Art and System Level Considerations

See esp.:  "Mars ISRU Pathfinder Demo Payload Options"

Production:  .48 kg/hr O2 & .12 kg/hr CH4 using 5.75 kW (ISRU power only, excluding other systems)

Application example:  Operating continuously without losses, that ISRU plant would fill one pair of SpaceX ITS tanks, using 41 TJ, over 228 Earth years.

What supports this as 'state-of-the-art' except for NASA's PowerPoint title?
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Online Robotbeat

  • Senior Member
  • *****
  • Posts: 28825
  • Minnesota
  • Liked: 8975
  • Likes Given: 5773
Re: Power options for a Mars settlement
« Reply #575 on: 05/07/2018 01:09 pm »
The ISRU pathfinder is not terribly efficient.

Which design changes might best improve efficiency, quantitatively?
Things like propellant production tend to gain efficiency at large scale due to less heat loss.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline speedevil

  • Senior Member
  • *****
  • Posts: 3225
  • Fife
  • Liked: 1674
  • Likes Given: 1998
Re: Power options for a Mars settlement
« Reply #576 on: 05/07/2018 01:58 pm »
Application example:  Operating continuously without losses, that ISRU plant would fill one pair of SpaceX ITS tanks, using 41 TJ, over 228 Earth years.
Or with the above mentioned 500kW(av) power supply as fits into one BFS, about 2 years, around a quarter of what I guesstimated using 50% efficiency as implied by SpaceX.
If the equipment weighs 150 tons (including harvesting) that would be a BFS full of fuel per synod, per pair of landed BFSs.

At least it's pretty much the right order of magnitude.

Offline deruch

  • Senior Member
  • *****
  • Posts: 2407
  • California
  • Liked: 1968
  • Likes Given: 5250
Re: Power options for a Mars settlement
« Reply #577 on: 05/08/2018 02:45 pm »
Some recent Mars ISRU power and production numbers, in Mars ISRU:  State-of-the-Art and System Level Considerations

See esp.:  "Mars ISRU Pathfinder Demo Payload Options"

Production:  .48 kg/hr O2 & .12 kg/hr CH4 using 5.75 kW (ISRU power only, excluding other systems)

Application example:  Operating continuously without losses, that ISRU plant would fill one pair of SpaceX ITS tanks, using 41 TJ, over 228 Earth years.

What supports this as 'state-of-the-art' except for NASA's PowerPoint title?

I think the "state of the art" in the title was saying that the presentation was going to give an overview on the state/status/developments of ISRU research.  Not that any particular demonstration/project represented the best currently possible with today's technology.
Shouldn't reality posts be in "Advanced concepts"?  --Nomadd

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 7883
  • Everyplaceelse
  • Liked: 1334
  • Likes Given: 8916
Re: Power options for a Mars settlement
« Reply #578 on: 08/14/2018 09:42 pm »
The ISRU pathfinder is not terribly efficient.

Which design changes might best improve efficiency, quantitatively?
It's true going bigger usually helps but there is also work in the chemical industry on "process intensification" this shrinks process units, usually by things like photo etched metal channels diffusion bonded together, and merging multiple units so (for example) the outflow of one unit is cooled by the inflow of other unit and vice versa, allowing heat to be scavenged  and reused.

I'd suggest the unit NASA has built is more of a Proof of Concept that it can function on Mars, not that it's particularly optimized for making a lot of propellant fast.
BFS. The worlds first Methane fueled FFORSC engined CFRP stainless steel structure A380 sized aerospaceplane tail sitter capable of flying in Earth and Mars atmospheres. BFR. The worlds biggest Methane fueled FFORSC engined CFRP stainless steel structure booster for BFS. First flight to Mars by end of 2022. Forward looking statements. T&C apply. Believe no one. Run your own numbers. So, you are going to Mars to start a better life? Picture it in your mind. Now say what it is out loud.

Offline RobLynn

  • Full Member
  • ****
  • Posts: 477
  • Per Molestias Eruditio
  • NZ
  • Liked: 245
  • Likes Given: 143
Re: Power options for a Mars settlement
« Reply #579 on: 08/20/2018 12:16 am »
No mention yet of solar + flywheel. For storage during long dust storms, seems like a high-reliability, high-density, overall easy option compared to large quantities of batteries, stored heat or stored fuel. No risk of leaks, fires, hopefully little chance of RUD. Additionally it could provide bursts of high current for things like welding, hot water, ovens, without causing a brownout.

Flywheel energy storage makes a lot of sense.  With ultra-high cost and time to ship to Mars long life is critical, and manufacturing cost is unimportant.  A flywheel can last for at least 10's of thousands of full power cycles with high round-trip efficiencies of >90% and very high power levels possible (for things like electric launch catapults) without impacting overall life.

Mars has near-vacuum atmosphere making necessary vacuum casing very thin and light - and can pack dirt around it for RUD safety.  Minimal vacuum pumping is required (and near-infinite life magnetically levitated turbo molecular pump can exhaust straight to atmosphere), and lowered gravity reduces magnetic bearing lift requirements.

7GPa T1100G carbon fibres in unidirectional layup enables almost 2500000m/s specific strength = ~300Wh/kg thin-cylindrical flywheels, but more realistically perhaps 150-200Wh/kg, which is pretty competitive with state of the art lithium ion batteries (that can't be cycled too deeply anyway).

Big flywheels have lowered accelerations at rim, and so reduced sensitivity to small defects that might otherwise lead to delaminations or breakages, and less sensitivity to gas friction.
The glass is neither half full nor half empty, it's just twice as big as it needs to be.