The thread for space cadets!

[red_elk]

Visualization of known asteroids and their discovery in 1980-2012:

 

[hobbsyoyo]

Scott Manley! Thanks for the link, watching it now.

I left some comments on it here.

 

[Berzerker]

the giant impact theory to explain the moon has some competition, multiple smaller objects struck Earth producing moonlets that eventually coalesced.

I'm a bit skeptical, seems we'd have several moons in different orbits or at least surviving debris trails (rings)

this is based on lunar samples

how do researchers know the rocks and dust they gathered represented the Moon when it was newly born?
If the Moon has been battered I'd expect the original surface to be distorted by impacting material

 

[Michkov]

Reading through some articles on the 2 NASA astroid missions for the 20s-30s mentioned the probes using quite lot of instuments similar to those used on other missions. How common is it that you can use off the shelf instruments for these kind off missions? I was under the impression most spacecraft built where more akin to prototypes and not serial production runs(however small).

 

[hobbsyoyo]

the giant impact theory to explain the moon has some competition, multiple smaller objects struck Earth producing moonlets that eventually coalesced.

I'm a bit skeptical, seems we'd have several moons in different orbits or at least surviving debris trails (rings)

this is based on lunar samples

how do researchers know the rocks and dust they gathered represented the Moon when it was newly born?
If the Moon has been battered I'd expect the original surface to be distorted by impacting material

There are a few techniques that allow dating of the lunar material. One method (and it's used on Earth too) is to chemically analyze the soil/regotlith found under rocks or underground. The amount of time the samples spent in the shadows can be backed out of an analysis of certain chemical reactions that happen in the presence of light. From this, they can begin inferring relative ages.

There are geological features that will weather at a certain rate under certain conditions and this can also tell you the relative age of formations of rock. The fact that the moon is such a stable (relative to the Earth) environment really helps with this age-defining process. There are also other chemical and radioactive methods of backing out the age of materials. Even things like reading the He3 content of a regolith sample can tell you something about its age since the Sun streams out a steady wind He3 that gets deposited at a more or less set rate.

Two huge difference between the Moon and the Earth is the lack of an ongoing tectonic and hydrological cycle (they're intertwined) on the Moon. This means the moon simple doesn't change much in absolute terms or rates. As far as I know there is no evidence that the lunar crust underwent the kind of recycling that Earth's does once it had solidified and the late heavy bombardment (which re-melted large chunks of the moon giving us the maria) stopped. Thus, we can be reasonably certain that the rocks formed at that time of solidification and haven't since morphed. Also, they don't show signs of geological morphing either, excluding the maria.

Reading through some articles on the 2 NASA astroid missions for the 20s-30s mentioned the probes using quite lot of instuments similar to those used on other missions. How common is it that you can use off the shelf instruments for these kind off missions? I was under the impression most spacecraft built where more akin to prototypes and not serial production runs(however small).

It's becoming more and more common. Spacecraft are often built around the same 'bus' but many of the specifics vary. The bus can be thought of like the frame of a car. You often couldn't tell two different models of cars apart if you only looked at the frame because they are basically identical. But the end products can be very different.

This is how it's been with satellites - the same basic frames will be re-used but fitted out with very different instruments and systems. This is especially true with electronics and software. A propellant tank is always going to be roughly the same because they all have to do the same job. But the electronic guts of a satellite and the software that run them can be wildly different.

In the early space age, scientists were working with such primitive instruments for data gathering that significant R&D (and overall budgets) went into developing new and better instruments for each new satellite. The same was true (to a lesser extent) of commercial satellites shuttling data around the Earth instead of taking science measurements. Each new model was a big step forward over the last and could deliver more data at faster rates.

Recently, electronics and software have advanced to such a state that for many mission profiles, you don't need (or even want) to re-invent the wheel, so to speak. An off the shelf solution can provide what you want, so you go with that instead of hiring a bunch of PhD's to invent a new one. This has the knock-on effect that as certain components and software become more and more common, they begin to drive ancillary commonalities in other components.

So let's say that all of my satellite's sensors run on different voltages. Say you have a thermal camera running at 3.3V, a laser range finder running at 5V and a radar running at 12V. Each instrument needs separate circuits, harnessing and connectors because of these voltage differences.

Now let's say that you find a radar that runs on only 5V and works as good as the old one. Well now you are strongly incentivized to find a thermal camera that also runs at 5V so you can re-use all the same connectors and harnessing and save yourself a ton of money. Designing the overall system is now much simpler and cheaper because of this.

That's what's going on right now. NASA is saving costs (and increasing overall effectiveness) of their missions by common-sourcing as much as they can. The market is reacting to that by heading together in the same direction vis a vis compatibility and commonality.

Things like the CubeSat platform are accelerating this trend dramatically as well since the whole purpose was commonality from the outset. It's an open-source platform and a lot of companies are trying to grab a chunk of that growing market. This means their components will all have to play nice with each other, so to speak, thereby reinforcing the trend.

I could literally show you how to buy and assemble a satellite using information on the internet and buying components from the internet, right now. That's how far the process has come - high school students design and build satellites for science classes in well-funded school districts now. Obviously there is a huge difference between what you and I could put together in a garage versus the Hubble Space Telescope. But if you only want to get something into space to take a simple set of measurements, that is now relatively achievable (if you have $10 million dollars that is )




Awesome questions!

 

[uppi]

I guess it is also a function of the launch cost. When those are extremely high, it makes sense to spend a fraction of that to overengineer your spacecraft to maximize the return. But if those come down, overengineering might cost much more than the launch so that it is more cost effective to just send another one in 5 years.

I could literally show you how to buy and assemble a satellite using information on the internet and buying components from the internet, right now. That's how far the process has come - high school students design and build satellites for science classes in well-funded school districts now. Obviously there is a huge difference between what you and I could put together in a garage versus the Hubble Space Telescope. But if you only want to get something into space to take a simple set of measurements, that is now relatively achievable (if you have $10 million dollars that is )

That much? I would guess that $1 million or so should be sufficient.

 

[Berzerker]

Thanks... There's quite an age difference between the Moon's original crust and the LHB. While the maria began forming after the LHB what about the debris from all those impacts? I guess what I'm saying is I dont know how we can tell if our samples are the original crust that survived the LHB intact or if we're sampling material that might have arrived or was changed during the LHB.

on planet 9 news, if it exists simulations suggest a %40 chance its a captured rogue, %60 chance it would have escaped rather than be captured

 

[hobbsyoyo]

Thanks... There's quite an age difference between the Moon's original crust and the LHB. While the maria began forming after the LHB what about the debris from all those impacts? I guess what I'm saying is I dont know how we can tell if our samples are the original crust that survived the LHB intact or if we're sampling material that might have arrived or was changed during the LHB.

on planet 9 news, if it exists simulations suggest a %40 chance its a captured rogue, %60 chance it would have escaped rather than be captured

Pretty much anything outside of the maria excluding ejecta (like the rays of Tyco's crater) is pre-LHB. (very roughly speaking)

And you can tell the ejecta is ejecta instead of ancient crust material from changes in the microstructure of the rock due to the immense shock and heating it underwent.

I guess it is also a function of the launch cost. When those are extremely high, it makes sense to spend a fraction of that to overengineer your spacecraft to maximize the return. But if those come down, overengineering might cost much more than the launch so that it is more cost effective to just send another one in 5 years.

Yes this is also very true. I wrote up a long post on it on Reddit but I lost access to that account. :-(

One factor playing against this however is the value (in revenue or science) that the satellites generate. The launch might cost $100 million and the satellite $200 million to build but if it either brings in $1 billion a year or nothing...then you spend $200 million making sure it works correctly. That also accounts for a ton of over-engineering.

But as launch costs drop, supply of satellites will go up, distributing the capabilities of the networks of sensors and transponders while also growing the overall pie.

That much? I would guess that $1 million or so should be sufficient.

Not at current launch prices.

Future vehicles that are coming on the market will come close to $1 million to launch about 15 kg (a typical 6U CubeSat for those that are interested). But they are not yet flying and when they are, they still wouldn't be a million, unfortunately. Rocket Labs out of New Zealand should have their first launch this year if everything continues going well with their rocket development, testing and launch pad construction. They claim the pad on the Southern Island is ready to go and they have really low prices for a 3U CubeSat.

The actual hardware would cost maybe $200,000 to do something really interesting or as low as $10,000-20,000 for something really basic.

You also have to factor in my labor.

Launches that go for less than $10 million (of any mass) right now have government sponsorship of various forms (not always direct monetary contribution but through subsidies and in-kind or service swaps) or are on-off deals due to external factors.

 

[Berzerker]

hate to be a bother but why does the Moon appear to have been slammed by large objects on one side?

I've read a couple theories but neither suggests any link to the LHB

 

[hobbsyoyo]

I would venture to guess (and I really am guessing) that it's because the Moon is tidally locked toward the Earth. One side always faces the Earth which means that the LHB impactors probably moved in the same general direction relative to the Earth even though they were all on different paths.

Because the Moon is always pointed toward the Earth it might then tend to present the same side toward the impactors.


Again that's a gues.

And you're not a bother! I love talking about space and rockets. A good portion of why I come to this forum is to post here.

 

[hobbsyoyo]

It's also worth noting that despite the hundreds of kilos that Apollo brought back (and some grams from the Russian Luna program) we still have a paucity of raw data on lunar rocks, making all theories tentative at best.

 

[uppi]

Not at current launch prices.

Future vehicles that are coming on the market will come close to $1 million to launch about 15 kg (a typical 6U CubeSat for those that are interested). But they are not yet flying and when they are, they still wouldn't be a million, unfortunately. Rocket Labs out of New Zealand should have their first launch this year if everything continues going well with their rocket development, testing and launch pad construction. They claim the pad on the Southern Island is ready to go and they have really low prices for a 3U CubeSat.

The actual hardware would cost maybe $200,000 to do something really interesting or as low as $10,000-20,000 for something really basic.

You also have to factor in my labor.

Launches that go for less than $10 million (of any mass) right now have government sponsorship of various forms (not always direct monetary contribution but through subsidies and in-kind or service swaps) or are on-off deals due to external factors.

I was thinking about 1U CubeSats for which the internet told me, launch costs are about $100k. But I guess that is only if you find a rocket that will take that as secondary payload.

 

[hobbsyoyo]

The internet is wrong. You will not be able to guarantee a launch price at that rate - especially without a sponsor or sweetheart, one-off deal. If you could then about 500 CubeSats would launch each year because there are at least that many universities with big enough endowments and active engineering programs globally that would be doing it. Maybe 50 CubeSats get launched globally on a busy year currently.

 

[uppi]

The internet is wrong. You will not be able to guarantee a launch price at that rate - especially without a sponsor or sweetheart, one-off deal. If you could then about 500 CubeSats would launch each year because there are at least that many universities with big enough endowments and active engineering programs globally that would be doing it. Maybe 50 CubeSats get launched globally on a busy year currently.

When there is a market for 500 for these at $100k, why is your company not exploiting it? Couldn't you put 100 of these on something like a Falcon 1 and still make a profit?

 

[hobbsyoyo]

The falcon 1 was discontinued nearly 10 years ago because the market didn't exist back then. The situation has drastically changed since then. You make it sound trivial to restart production of a retired rocket.

Falcon 1 could not carry 100 CubeSats of any mass due to volume limitations.

I can't really say much more publicly on the subject as I don't speak for the company.

(I can talk about the industry and CubeSats all day long I just can't get super specific about the company and our products, strategy, etc)




The launch today was completely successful. We put 10 SATs into LEO and recovered the booster. We had live streaming of the first stage the entire way down too.

 

[Berzerker]

I would venture to guess (and I really am guessing) that it's because the Moon is tidally locked toward the Earth. One side always faces the Earth which means that the LHB impactors probably moved in the same general direction relative to the Earth even though they were all on different paths.

Because the Moon is always pointed toward the Earth it might then tend to present the same side toward the impactors.


Again that's a gues.

And you're not a bother! I love talking about space and rockets. A good portion of why I come to this forum is to post here.

There's evidence the Moon produced a strong magnetic field at some point during its 1st billion years. That could mean the Moon was spinning too and then became tidally locked, perhaps during the LHB.
Its my understanding that if the Moon had an ocean the near side would be covered by water and the far side would be land. That means the near side lost a bunch of material. But the near side is also denser than the far side, whatever hit the Moon during the LHB was 'heavier'. Thats why the side with the lower elevation faces us.

One theory says the Moon was close and tidally locked resulting in a heating process by which the crust thinned as near side mat'l 'evaporated' and was deposited on the far side. The crust on the dark side is much thicker and that has led to another theory suggesting a 2nd moon joined the Moon and was deposited there in a low energy impact. I think the Moon was hit by material ejected from the Earth during (or starting) the LHB, the Earth was hit removing virtually all evidence of the prior world's surface and one side of the Moon got plastered and then tidally locked.

I guess my point is: if the side of the Moon facing us lost several km (?) of elevation and we're collecting rocks from that area, we're not looking at the original crust. I imagine we gathered rocks from the far side too, we might have better luck finding a semblance of the primordial Moon. But its so heavily cratered. The fact researchers keep changing the dates for the Moon shows its a complex story.

 

[uppi]

The falcon 1 was discontinued nearly 10 years ago because the market didn't exist back then. The situation has drastically changed since then. You make it sound trivial to restart production of a retired rocket.

Well, it is certainly not trivial, but i would hope that you did not destroy the plans, so that you would not have to start from scratch.

Falcon 1 could not carry 100 CubeSats of any mass due to volume limitations.

Huh? If I understand the Falcon 1 User guide correctly (http://www.georing.biz/usefull/Falcon1UsersGuide.pdf), it had a payload volume of almost 4m^3. 100 CubeSats would have a volume of just 0.11m^3. You would have to account for the volume of the deployment system, but surely that would fit?

I understand that you cannot comment on company strategy. I am just saying that if you are not thinking about some kind of dedicated microsatellite launcher you would probably miss an opportunity.

 

[hobbsyoyo]

You are reading the user guide correctly but not accounting for a ton of variables. That volume is at best a "best case" scenario for a single satellite.

The deployers for CubeSats are not terribly large but they are not small either. The bigger problem is the required harnessing and mounting infrastructure for each deployer. Then you have to factor in shielding to prevent cross talk between all the payloads. Then you need thermal conditioning systems for each payload. Then add on the requirement that none of the sats and deployer mechanisms can be allowed to interfere with each other or collide. Suddenly your usable volume has shrunk by an enormous amount. Next add on "special needs" requirements for some of the payloads (if you fly that many, some will be oddballs that fall outside the CubeSat specs in various ways) and the whole scheme spirals to a point where you can't do it.

I'm not saying you couldn't launch more than one CubeSat on a Falcon 1. What I am saying is that strictly pulling the mass and volume limits out of the user guide will lead you to a terribly incorrect first order approximation of what the vehicle could do. And that's coming from someone who LOVES simple, first order approximations. Lots of my blog work is based on this kind of analysis and normally it's a great approach for back of the napkin design work.

Another huge (and not obvious) problem with multiple launches is dealing with all the customers. So a single sat may have an operator, an owner, a builder and an insurer that are all separate entities working on their own schedules. Now multiply that by 100 and suddenly managing a manifest is impossible.

One thing we are doing I can publicly talk about because it's already known is that we work with a single service provider that does all of that leg work. They talk to all of the entities and deal directly with the launch provider so the launch provider doesn't have to do all that legwork. This simplifies the process and will allow us to fly (IIRC) 27 CubeSats on a single device called the Sherpa which will eject them individually on orbit - in addition to another separate, primary payload.

Sherpa (and things like the ESPA ring) handles a lot of the mounting problems mentioned above but it takes an enormous volume relative to the payloads themselves. So you don't wind up with anything close to as many CubeSats as you would think from a first principles approximation.


Finally you have to think about the long term goals of the company and how they may or may not align with the business opportunity you outlined.

 

[hobbsyoyo]

Also CubeSats come in lots of sizes and using a 1U CubeSat as a representative is also not a great start. You can't do a ton of interesting (read valuable $$) stuff with 1U. (Which is the size I think you used to back out total volume, am walking and typing on phone so can't check myself)

Most professional commercial and government CubeSat users start at 3U as this is close the minimum size you need to provide 3 DoF attitude control and/or a basic propulsion system for maneuvering or de-orbiting (government requirement in many cases). There are some schemes that would allow you to do one or the other with just 1U but they leave no room for useful ($$) avionics or experiments.

 

[hobbsyoyo]

Oh and just having blueprints does not a rocket make. Not to be flippant but there is so much infrastructure that goes into making a rocket - even one a company used to build. So much of the cost of rocket production is related to the diameter of a rocket that it's pretty clear setting up parallel production of F1 and F9 is....challenging. To say the least.

@ Bezerker

I'm at the limit of my knowledge of the moon as it pertains to this discussion. I cede all points to you mate.



I will try and put together a better response later as you got me wondering about the Moon now.