i'm redoing the notes on the TESS model on Sketchfab

i looked over the document about the heat losses, and have been thinking about a comment on the space show blog that such a system might be insufficient for power generation

i believe i'd had the idea that after the pit was functional, a trench could be dug around it to limit heat loss, but then i left that

now i'm back at the spot where i wonder if i should return the design to being a huge above-ground tank, or i should model in the trench

or maybe even a little of both. a trench saves on building the tank, which is no small feat, but what do you do about the bottom? i have thought about making it a cone, and inserting supports as the ground is dug away.

complicated... but then again, so is a huge tank...

let me whip up a sketch

@kimholder I think we need to have another round about the energy situation again. The question is important and not completely answered yet.

it sure isn't

Do you have some idea about the thermal situation of the base itself? Does it require heating? All the time? Only at night? Or perhaps radiators instead?

the earlier days can be resolved with nuclear reactors, for lalande. the solar thermal could work well if the technical problems have decent solutions proposed for them.

A thermal heat pool is necessary in almost any system as a buffer for storage. But it sure is lossy.

@Hohmannfan with the number of light tubes the galleries have, it takes a large mass of stone to act as a heat sink to keep temps even. i postulated that for most of the gallery length, the floor at least is in contact with rock

and by sinking shafts in that rock to get the surface area and penetration that is adequate, the heat coming in with the sun can be dissipated by that rock so that temperatures stabilize at a point that is comfortable and then it is very hard to shift temps away from that

As far as I remember (I think you had a diagram for that), the lunar rock temperature is pretty stable and close to ideal. That is good then.

well, they figure it is about -20 C

Very nice.

Also, covering the base partly with reflective aluminium should be a very easy thing to do in case of overheating. That stuff is basically 0 kg/m²

@Hohmannfan my hope was that once again, the vacuum could be used to limit losses. But you have to isolate the whole mass almost completely.

@Hohmannfan the overlay of regolith needed for radiation protection makes that unnecessary

the only place where that is an issue is where sun is being brought in for plants, in the atriums

in the galleries, where light tubes are used, you just turn the lenses off axis a bit to dilute the light if it is getting hot

Does not regolith have some 12% reflectivity or something? That is not very good.

yes, but because it is mostly very small particles, it doesn't conduct heat very well

the measurements on the moon showed that temperatures are almost constant under about 20 cm of regolith

No problems then.

Do you perhaps have some thermal data for loose regolith in your collection?

the document the whole solar thermal idea rests on has a bunch of figures and calculations

ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19920005262.pdf

Very good.

After modelling most parts of the base now, do you have some estimate on the energy demand? I guess the most power hungry part is the industrial equipment, but that is just spikes when it is active.

that is so complicated and totally above my math abilities i've left it. my hope was to simply size things large enough to have a good margin for error.

i am a bit more concerned about getting the sizing of individual pieces a bit more into proper scale. i have no idea how big the heat reserve and the radiators should be on the solar thermal system

there is so much to do, putting things together all properly sized has to come together slowly.

the TESS is for an advanced colony, and unless it really has problems, it would be nice to keep it.

The ISS has about 100kW for 3-6 persons, so let us say somewhere between 10 and 100kW/person for an order of magnitude.

mmm, i don't know if a valid comparison can be made that way though

the moon colonies conduct industry

1 to 1000 kW/person then :P

Yeah, we don't really know.

:] i have tried to use direct solar heat as much as possible, so it doesn't need to be converted to electricity

But a continuous supply of 1MW is useful, and checking out what is required to provide that may be interesting.

on the scale and format Cernan's Promise is done on, some things that the ISS has to worry about shouldn't be a problem - fans, air scrubbers, pumps

@Hohmannfan yes, that sounds like a good starting point

Sorry, but are not all those things needed?

well, much less. the point on the timeline where a solar thermal system is even possible is advanced

the atriums will be full of plants - reducing need for air scrubbers

the living spaces are far more open, reducing need for fans

heat control should be largely passive, reducing need for pumps

coming up with ways to distribute plants widely, and with bio-processors for human waste - these things would be research priorities, and success would also reduce power needs.

Perhaps not CO2 scrubbers directly, but a working ventilation system is definitely needed.

you are dealing with large open areas instead of a lot of small tubes

sketchfab.com/models/1be68b1079b54315809e2be3ea8d1a6d

i have been editing these notes, and i am about to simply delete reference to oxygen collection. The temperatures needed are just too high. They create too many materials issues with the heat pipes running through the mirror troughs.

Let us first rule out the chemical batteries. If you have solar power only during the day, you need storage for 330 hours of darkness. That is 330MWh. Good batteries are about 200 Wh/kg, so that is about 1,650,000 kg of batteries. Not gonna happen.

flywheels?

Can try to look up some numbers.

According to the wiki, JET has two 775 ton flywheels each storing 3.75GJ (1.04 MWh). That means 250,000,000 kg of flywheels for our system. Nope.

There may be better flywheels though, but there are way too many orders of magnitude here.

i have this feeling we have had this conversation before... that has been happening to me more lately

getting old i guess

but i think we settled on heat reserves because it was the least bad option

We have had this conversation before. Kinda. But no solution.

@kimholder Yeah, least bad.

That was also the strongest argument for nuclear reactors, they also work at night.

other that nukes, which are great, and maybe it would even be reasonable to say that the colony can make fuel themselves

beat me

the good thing about a solar thermal system is that the lava is useful for other reasons.

if the heat dissipation issue could be handled, it would be a sweet system

and if the pipes and the working fluid and all that - if that has a good solution

The pipes need to run at something like 1400 C

A Lava Lake! Or even Nuclear Lava Lake!!

welcome to nuclear lava lake park. enjoy your stay.

i recall that on the forum, you were very concerned about the 4th power heat loss thingy

but because reflective foil is so effective in vacuum, you felt it could be controlled.

if a trench could be sunk around a large enough volume, there could also be a heat gradient included within that volume.

wolfram alpha lava pool size

ooooo - how sweet is that

About 300 m³ of liquid basalt. Totally "manageable"

Though a significant fraction of the energy would be in the gradient in the insulation material.

so, then it could also be reasonable to put a buffer zone around the volume, so the heat that reaches the walls is lower

by the way, the unit as it stands is for 6 MW

i don't know how i guesstimated that, but that number is there

something is going to be a buffer zone anyway, it is not like the entire universe is 1700K

okay, let me throw around some numbers like the total engineer wannabe i am...

Hey, what about a permanent nuclear meltdown keeping the lava hot?

you have a tiny evil genius in you...

let's see, volume of a cone...

base times height divided by 3

Putting "cone volume base 10 m^2 height 18 m" into wolframalpha even draws a visualisation of the volume.

okay, a cone with a radius of 10 and a height of 18 gives us the desired volume, more or less

i really have to learn to use wolfram alpha

Put whatever you want into the box. It probably understands what you mean.

That is a really sharp cone by the way.

20 by 18?

That is still a base diameter of only 5 meters

oh - i mean the radius is 10 m.

Ouch, I did not see that. Sorry.

Yeah, 10m radius, height 18m is a nice shape.

if you add 3m for a buffer zone around everything, the volume is then 3700 m3

by the way, how do you get that nice superscript 3?

Like this? m³

yeah

m ³

that doesn't work, so i was wondering

Shift, ^ and then 3

oh, nice... gee, they think of everything around here

This works everywhere though...

it does? i know it is standard notation, but i haven't seen it properly processed into a superscript in a forum or chat before

anyways - creating a trench that forms a cone 26 m across and 21 m deep doesn't seem so bad

Superscript numbers are unicode, so anything that supports that should work.

hm. well, i'll remember that.

If the ^ shortcut does not work on your keyboard, ctrl+shift+u+00B3 should always work.

yeah, but the chances i'll remember that are much lower :D

m^3

00B2 for ², etc.

I think of those as the "B" numbers.

or - use firefox? is that why yours doesn't need that extra step?

It is not program specific, so I guess it is part of the Norwegian keyboard layout. Let me test...

superior norwegian keyboard layouts...

I tested the English layout, and nope, the ^ is located on the 6 key, and it only shows as ^3

You guys only have two characters per key? Pffft...

huh. well, i'll try to remember the ctrl+shift+u+00B(x) solution, but i imagine that i'll just expect people to know what i mean by m3

@Hohmannfan necessity is the mother of invention :P

wow :D i'd like to have that arrangement and stick in a bunch of different things

Seriously though, we only have three additional letters. But for some reason they added a ton of other characters as AltGr and AltGr+shift alternatives.

But really, ⅝ ?

heheheh - greek letters is awfully nice

and if you work with imperial measurements, 5/8 comes up a lot

especially in construction.

Ħ : "It is used in Maltese for a voiceless pharyngeal fricative consonant". yeah, we totally need that on our layout.

anyways... there is a lot of excavation considered in this project before this solar thermal system is done

What has that to do with keyboard layouts? :)

i would star that except i don't want to encourage you ;)

i'm not sure why i didn't think of that before. Maybe i did when i was making the model, but why would we not simply excavate a volume shaped however we want, set up the supports and the insulation, and fill it back in again with the material we excavated?

If you have a large block of regolith, 300K at one end and 1700K on the other, in flow equilibrium, what is the heat flow?

ouch - that sounds hard

Even that is a not totally realistic, the cross section area increases with distance.

Perhaps I should figure out how a sphere would behave? That number is going to be useful anyway.

if this is done by excavation, a cylinder is easy, and a sphere is not too hard either

and one might as well right? less surface area.

A spherical cow^H^H^H lava pool is most easy to calculate.

lol

The equations are not too complicated, but a simulation would be less brain work to create.

so, this would end giving us a distance that needs to exist between the two temperatures?

More or less. But more importantly the W/m²

i tend to think in terms of the models i need to build :P

Hey, look, it is a designer over here!

:D look at my pleasant smile and fashionable clothing

(actually i am an exception in that regard. no fashionable clothing.)

pubs.usgs.gov/of/1988/0441/report.pdf

oh wow! i have looked for a reference like that before with no luck. awesome.

hm. i think we want mafic igneous rocks, air in pores

without phenocrysts

The "Rock conductivity change with temperature" part is also very important

actually, it would be best to fill the outer areas of the tank with fine grain regolith so it doesn't conduct well

yes. so the chart on page 12 gives us a measure at 300 K, and i'd say it would be fair to take the bottom one, without the phenocrysts, and with no porosity

so, about 1.5 W/mK?

Approximately doubled over the interval, yeah. The simulation only has to be written once, so we can try many different rock structures as long as we can find data.

'the pressure applied to rocks in figures 14 to 19 was about 5 MPa'

Not a lot.

at the top of the tank there should be almost none at all

50x air pressure is nothing to think about.

temperatureGradient.cpp

is that a file?

Soon.

See? We want to hear about questions like that.

oh yes we do. i'm about to answer.

target.lroc.asu.edu/unavailable.html

AHHHHH

what's happening???

AHHHHHH!!!!!!!!!!!

sigh - ok, let me see how this lunaserv thing works...

moontrek.jpl.nasa.gov/#

well, from reassessing my moon map options, i went again to LMMP, which i hadn't visited in ages because it was inferior to Quickmap

it has some nice features now... but doesn't seem to have the NAC images, the beautiful ones that went down to 50cm per pixel

Some data.

Warning, this is an untested simulation.

Varying heat conductivity not implemented yet, assumed to be 1.5W/mK all the way.

Varying heat capacity not implemented either, assumed to be 2.28e6 J/m³ all the way.

This is the temperature drop I get over two meters from a 5 meter radius spherical pool:

Not reliable in any way, but it may indicate that the hot part is only a couple of meters thick.

The walls leak about 1-2kW/m²

that seems to jibe pretty well with a result in that paper i listed up near the beginning of this conversation

But I am in no way trusting this simulation yet, though it gives reasonable numbers at the moment.

trusting...

Garden path typo

as in, you are on a garden path?

Like in en.wikipedia.org/wiki/Garden_path_sentence

Increased the conductivity to 3 W/kM, this time it is 5 meters instead of 2:

Though a radial diffusion behaves differently than my spherical diffusion (quadratic vs cubic)

With no better insulation than this about ~100MWh is going to be lost during a lunar night. This is promising.

There is also the consideration that the turbines are going to be running, pumping heat out of that store continuously

and i wondered if it would be the best option to place the heat exchange pipes for that around the perimeter of the tank, where they absorb heat that is approaching the exterior of the tank anyhow

To keep the lava liquid the storage has to be much larger than the amount of power we want to be able to drain.

it is okay if it solidifies overnight, i think

Turning heat into electricity is not very efficient as well. This is going to be a huuuge lava pool

Reflective sail hoovering behind the Moon...

consider the latent heat of the lava solidifying

and perhaps the turbines can be designed to run on good old steam

It feels like a couple of reactors are a lot easier to build than this though.

well, if they are, then they are.

we still need lava for other things, so the data remains useful.

what would we be talking about? a lava pool hundreds of meters across?

i mean, if we actually tried to use it for power

Heeey, Anatoly Zak was on the space show 22-feb!

@kimholder 20x20x20, maybe

but, that doesn't seem bad...

you mean just the liquid part, right?

Yes,

so maybe 25 x 25 x 25

or 30 x 30 x 30

the excavation is no problem...

the insulation isn't a big deal...

supporting the mass? well, not the easiest thing, and heat will leak there, but not tragic...

gotta love that low gravity at moments like this

compared to building nuclear reactors? i don't know - there must be hard things about that too, no?

Anatoly Zak is on the Hotel Mars portion quite often, actually

i believe there are one or two fulls shows just with him too

Zak had a lot more recognizable accent than I expected.

oh yes, very sexy russian spy sounding :)

Russianspaceweb.com and the "Russia in space" book are very good resources about the Russian space program.

@Hohmannfan that's the figure for 1 MW?

The value for 0.5 MW - 10MW

ah - the estimate is that rough, eh?

Yep, eh.

hey - i hardly say 'eh' ever anymore. i'm glad i still have it in me. :)

let me work on the model considering everything you've said.

there is an engineer who contacted me because i'd asked about a good physics-based answer as to why an airship to orbit doesn't work on the space show

Airship, as in hydrogen?

sorry, i got distracted for a moment... yes, there is this guy, his company is JP Aerospace

jpaerospace.com

Is not that kind of obvious? If the air is dense enough to lift you, you are not in space.

David has had him on a few times because he offers some legit and interesting services, but he persists in saying it is possible to use lift to get to the Karman line

Also: orbit? [insert xkcdsidewaysmotioncomic here]

he says you can use a giant inflated wing to slowly accelerate as the vehicle rises so its velocity always allows it to have enough lift to remain aloft

and there was a show where someone called in and challenged that, but i felt the argument wasn't all that clear - not wrong, just not that clear

and we have a question about it that was never answered well, in fact the accepted answer claims it is possible.

Accelerate? With what? All air breathing engines are pretty much unusable above 2-3 km/s. That is far from orbit.

he says with ion drives

sort of...

at any rate, because i asked for a more thorough explanation, i received two. one is on the space show blog.

Enough air to generate lift, but at the same time so little that the drag can be counteracted with ion engines? I am very sceptical to this.

disq.us/p/1ht1m4f

and a bit above that was the second reply that explained why not, from Mike Murphy, who originally wrote me personally with the explanation, through moonwards.

and he and i got into quite a conversation. i was quite satisfied that the airship doesn't work, but i was happy to have a much better explanation as to why not.

If you can make a wing with a lift-to-drag ratio of 100 (almost impossible), you still need ion engines with a thrust-to-weight ratio of at least 1/100. (assuming the wing does not weight anything).

(i thought about writing it up as an answer to that old question, but i don't know that i can use those references really.)

yes - other interesting points: the sound barrier

and heating

anyhow... Mike was chatty enough, and had written me at length enough, that i asked him if he'd be willing to advise me on aspects of Moonwards

NSTAR is two orders of magnitude short, but the lighter-than-air lift helps a little, and so does scaling. I can not say this launch system is completely impossible, but it is far fetched.

John Jurist said the drag increases with the square of velocity, which seemed to make it impossible

his answer is quite good

But the lift increases proportionally with drag, so that does not matter.

but if the drag is slowing you down, how do you maintain the lift?

The required lift is going to be constant (well, decreasing as you approach orbital velocity), and as the lift-to-drag ratio is a ratio, the drag is also slowly shrinking as you always choose an altitude only giving you the lift you need.

The physics behind the concept is fine, the materials and drives probably not.

but you need to constantly overcome that drag to get the lift, and the energy you need to do that, or the wing size to compensate, become so huge due to the squaring, it becomes unmanageable

i'd be interested in your opinion of the responses offered on the blog

No, lift = drag * LTDratio

More drag is good.

Hi, @Algohi

especially since Mike may be advising us some

@Algohi, do you have an opinion on the matter?

As the acceleration happens in air unusable for a jet engine, the only benefit from this method of launch is the Isp of the ion engines. That benefit is somewhat less as most of the work is for the structure to support itself.

As I see it, this is less realistic than for instance a liquid core NTR.

> The remaining lift force must come from aerodynamic lift providing 49,000 Newtons. Ignoring the vehicle configuration needed to provide such lift, the drag is the lift reduced by a factor of 70 or 700 Newtons at sea level. Using the same drag coefficient times area value as in the first scenario, the 700 Newtons of drag at sea level corresponds to an airspeed of about 6 meters/second, so liftoff would occur at that forward airspeed.

> At an altitude of 40 km, the first scenario required thrusts of 757 Newtons at 100 meters/second and 2,639 Newtons at 200 meters per second to overcome drag and provide the 1.32 milliGee net forward acceleration.

Yes. Exactly.

Almost a kN from ion engines is unrealistic.

but that is just for 200 m/s too

For larger speeds, you pick a higher altitude.

Double speed means four times as much drag, but also four times more lift. That is a lot more than needed, so you need four times less air density.

John Powell, the guy trying to do this, says it is a kind of 'dirty plasma' engine with more thrust

it only has an Isp of 1200

:D He is inventing exotic engines for this. Did he mention EM drives too? :P

no, but i did :D

:(

actually, i mentioned Mach drives

hey, there is enough documentation that it may actually work

and i thought it would be an interesting exercise to say if it could work if you didn't have to take all the propellant with you

Isp of 1000 is about what a solid core H2/NTR has.

the propellant mass needs run away in any scenario that either respondent named

it's like the rocket equation, but worse

Stop, no Mach drives, please.

Jurist said L/D of 70 was 'wildly optimistic' too

Yes it is. I said "almost impossible" to a value of 100.

Mike raised the issue of heating due to drag, and John pointed out the small matter of breaking the sound barrier

"minor issues"

anyhow... let me try to finish redoing the notes on the solar thermal station

i guess i should try to make it as good as possible, and then we can decide if we are going to simply discard it :P

Ok then. I am going to research stuff about operating reactors in space.

ntrs.nasa.gov/…

starting on page 36, there are various drawings that may be almost enough for me to model it fairly well

That is a small reactor though.

does the design change a lot as it is scaled up?

I am not a nuclear engineer.

Maybe.

well, if i model that, it might be enough for me to get a good opinion from someone at NASA as to that matter

see, somehow i have to do enough to get a serious opinion from them

Zak has a chapter in the book describing various systems, but I remember where my copy is.

or i just push ahead and fix it later

oh, that sounds good

i do need reference that are adequate for modelling though

"NASA also said that production of Pu-238 was being ramped up to 1.5 kg/yr by the mid-2020s, and by the end of 2015 over $200 million had been spent on this"

This substance is expensive.

what we'd really like, though, is a thorium reactor. a lot to ask, but hey, we just say the money is there.

i am now scrolling up to get the key figures you provided to improve the TESS notes

Why thorium? That is just a complicated way to make uranium for your reactor.

because it can be had locally

there are reactors that run on thorium

Uranium is on the Moon too.

btw - someone said regenerative fuel cells area good power storage option

And reactors do not run on thorium.

oh, i see - it's used to make uranium 233

it would be nice if mining and refining enough fuel for that from local material was easy.

of course, the fuel doesn't weigh so much that it wouldn't be a lot easier to get it from Earth, unless you are getting it as a by-product of other mining activities

But honestly, reactor fuel is probably THE most mass efficient thing to send to the Moon.

beat you :)

what about the shielding and all that?

Hm? Fresh reactor fuel is barely radioactive.

Spent fuel on the other hand...

don't you need neutron reflectors to make it work?

Are we talking about transporting fuel now?

As for nuclear engines, the reflectors are for making them much smaller.

no, i mean the mass of the reactors that need to be imported to make working units

Yeah, maybe a beryllium/uranium combination has less mass.

i probably shouldn't be multi-tasking on this. going over the molten regolith heat storage power system paper is hard enough. i think there are figures buried in there that would be useful...

Figuratively buried in the regolith.

i should open this on my tablet, i'm going to need to highlight and make notes

Pu-239 has a lower critical mass than both U-235 and U-233. Perhaps the reactors can be smaller?

actually, i was mistaken about the whole role beryllium plays. i thought it had to be there for a reactor to work at all. We're building an entire system of tethers in order to make importing and exporting high volumes of cargo cheap

the small amount of mass saved by using plutonium has got to be insignificant

Yeah, probably. The plumbing and moderator blocks are maybe the largest components

A reactor can not be that big. The rd-0410 is a complete engine at 2500kg

oh look - that's what we wanted before, right?

Exactly what we want.

this is what happens when you read something properly instead of skimming (like i usually do)

We should have a good way to have a shared archive with all these pdfs

well, i have been wanting to redo the wiki section of the repo in a major way for that very reason

The melting point is easy to spot, but what is the drop?

an effect on conductivity due to the energy absorbed by melting?

No, in the specific heat diagram.

oh - but it's still due to melting, no - it hauls in energy due to the state change?

and then once it's done, the curve returns to normal?

But there is no data point after the spike though...

um...

extrapolation i guess

I can also extrapolate like that. [deleted so nobody get confused with the wrong diagram]

yes, but i imagine the extrapolation they did fits with what substances usually do

there is also this

so, the data is actually from the previous plot. which extends beyond the dip.

would you like the derivative term approximations for energy transfer from a cylindrical source?

Very yes.

That is real data in contrast to me guessing rock parameters.

it was wrong of me to post the link to the paper and expect you to pore over it. i knew it held these terms, but i guess i didn't say so.

They used cylindrical shells, I used spherical. Numerical simulation FTW.

it has it for a sphere too. hang on.

Yeah, I have not read it yet.

(i read this a long time ago, and i really don't know how to say 'yeah, but the derivative terms of the approximations are in that paper)

As the surface is a real thing, a cylindrical model is probably better.

yeah, a cylindrical tank also makes a lot more sense

Interesting shape of that graph. I have to implement the varying specific heat.

over a series of day/night cycles, the behaviour changes because the assumption here is that this is just a pipe that goes into the ground. No tank.

and at one point i thought, 'adding some more mirrors just to make up for the losses to the ground is simpler than making a tank'

but now i suspect i'm wrong

The author (who, by the way, is Anthony Colozza) figures that 1800 K is the top temperature the system could be expected to withstand with constant use.

:D i am happy

Using the models for conductivity and specific heat made it look very close to the diagrams they provided.

10 m is rather a large buffer zone to put in a tank

how hot can we let the walls get and still control heat loss well?

does it make sense to put the heat exchange pipes draining off heat to run turbines around the perimeter so they cool the outer edges?

Forgot to add that to the diagram: The lava pool is in a cylinder 5m in radius.

ah. and however deep.

Infinite.