SpaceX Mornings Like This

[Anon2]

Elon Musk says we explore interplanetary possibilities so that, "When you think about the future. You can be happy about the future. Not sad or depressed."

Today's launch of Falcon 9 X37B, OTV5 from LC39A had future feels. As in best way to begin the day. Being optimistic over what future holds for humanity and our tiny existence in such vast expanse.

What I've come to realize is it's not just about numbers figures stats drawings and all the STEM. Every functionality in their perspective disciplines.
There's a place for creativity in STEM. One must be, at heart, a maker of stuff that works of practically nothing, to even begin to ascertain what's out there. If THAT were 100% a thing possible to be grasped! Nevertheless we combine strengths in our very best try.
Ingenuity in concert forms most perfect union. PUN.

If you missed this morning's latest lift off and stage one awesome soft landing return. No worries. There'll be more where these and those before have come. Then go in wholly pursuit of happiness. Rock on! This is my tech-less feels-more take. For now

 

[Sparky]

Agree. It's a 20-minute stop at in internet oasis otherwise landscaped with vapidity and cynicism. Rejuvenating.

 

[ccutrer]

I need an option between the two. I like to watch them live, but am not always able. I never miss a replay if I can't watch it live, though.

 

[Anon2]

"Knowledge is only half the task. The other half is love." John Burroughs

 

[bxr140]

I need a different option. I realize I'm biased on this one, but while launches and landings are cool to watch they're basically a footnote in the big picture.

Iffn you're really looking for the feels, I encourage you to look toward the missions. Whether it's exploration or observation or communication, it's the things we huck into space that will really define this moment in the onward progress of civilization. Even the most generic and capitalistic geocomm sattelite is globally unifying people much more than a few spectacular minutes of a falcon 9 launch.

If I'm bummed about one thing, it's about the disproportionate coverage given to launches vs missions. The space industry is terrible at showing people just how affected they are by the things we put into space.

 

[jkn]

I need a different option. I realize I'm biased on this one, but while launches and landings are cool to watch they're basically a footnote in the big picture.

Iffn you're really looking for the feels, I encourage you to look toward the missions. Whether it's exploration or observation or communication, it's the things we huck into space that will really define this moment in the onward progress of civilization. Even the most generic and capitalistic geocomm sattelite is globally unifying people much more than a few spectacular minutes of a falcon 9 launch.

If I'm bummed about one thing, it's about the disproportionate coverage given to launches vs missions. The space industry is terrible at showing people just how affected they are by the things we put into space.

Reason why we have only few small satellites and only one small space station is cost of launch to orbit.

SpaceX is reducing that problem right now. History of space exploration will have two very different periods: Time of expandable rockets and time of reusable rockets. So landing of the rocket is most interesting part!

 

[bxr140]

Reason why we have only few small satellites and only one small space station is cost of launch to orbit.

Legacy launch costs are well under half the capex for a generic space mission like a geocomm, and that fraction goes down as the complexity/uniqueness of a mission goes up. Of course lowering launch costs will make some difference, but when you play out the numbers game, reducing launch cost by half might only reduce your total mission cost by 15% or even less. That's certainly good, but the end result will mostly be a reduction in total annual space expenditure and not so much pumping that savings back into increasing the quantity/frequency/size/complexity of space missions. In fact, the biggest impact from SpaceX lowering launch costs is probably that it is changing the customers' perspective on total cost. The whole industry is being challenged to reduce legacy spacecraft cost, explicitly [in part] because SpaceX proved it can be done with launchers.

That cost reduction is starting to happen in various ways (on orbit servicing, on orbit assembly, megasats, megaconstellations, etc.), but--to take nothing away from spacex--the challenges that stand in the way of those concepts are much more significant than those for landing/reusing rockets. Put another way, it is reduction in the cost of space hardware itself that is the catalyst you're looking for, not expendable vs reusable rockets.

On a positive note, if you're looking for some inflection point that history will really remember decades or centuries from now, it will be on-orbit manufacturing.

 

[jkn]

Put another way, it is reduction in the cost of space hardware itself that is the catalyst you're looking for, not expendable vs reusable rockets.

On a positive note, if you're looking for some inflection point that history will really remember decades or centuries from now, it will be on-orbit manufacturing.

Why space hardware is so expensive? One very important reason is that needs to be extreme light. It needs to be light because of launch costs. SpaceX already offers much lower launch costs than others had. (1/2 or 1/3?) They can reuse 1. stage, but only after long service and check operations. If I remember correctly, they aim to 48 h service time between launches. If they manage to do that, costs will really go down. Perhaps to 1/10 of what it was or even lower. Then extreme weight reduction becomes less important. More standard components will be used. More satellites will be build, so standard parts will become much cheaper.

Cost comparison: Airbus A380 has about same takeoff weight as F9. It can take more fuel than F9 (most of the propellant in F9 is LOX, which is cheap). List price of A380 is 340 M$. So reusable F9 with quick service time should be cheaper to operate than A380. Rocket suffers from much larger thermal tress than airplane, so perhaps it can never last as long. Even so, if everything goes well, costs will really come down.

On orbit manufacturing needs raw materials. So we need Moon base!

 

[ItsNotAboutTheMoney]

Why space hardware is so expensive? One very important reason is that needs to be extreme light. It needs to be light because of launch costs. SpaceX already offers much lower launch costs than others had. (1/2 or 1/3?) They can reuse 1. stage, but only after long service and check operations. If I remember correctly, they aim to 48 h service time between launches. If they manage to do that, costs will really go down. Perhaps to 1/10 of what it was or even lower. Then extreme weight reduction becomes less important. More standard components will be used. More satellites will be build, so standard parts will become much cheaper.

Cost comparison: Airbus A380 has about same takeoff weight as F9. It can take more fuel than F9 (most of the propellant in F9 is LOX, which is cheap). List price of A380 is 340 M$. So reusable F9 with quick service time should be cheaper to operate than A380. Rocket suffers from much larger thermal tress than airplane, so perhaps it can never last as long. Even so, if everything goes well, costs will really come down.

On orbit manufacturing needs raw materials. So we need Moon base!

Not just being light. Launch costs add to cost of satellite failure. Cost of failure is cost of build+launch+downtime. If you reduce the launch cost significantly, and can launch quickly, you've reduced two of the three costs of failure, so the satellite manufacturers would have more opportunity to use redundancy to provide reliability.

 

[bxr140]

First, I'll apologize for initiating the sideways turn this thread has taken from @rachelawilson's original post. As someone in the space biz I completely agree that Its vital for us look beyond financial and technical details of The Now and contemplate the broader significance of what's going down. So thanks for that!

Why space hardware is so expensive? One very important reason is that needs to be extreme light. It needs to be light because of launch costs.

Unfortunately, the analysis is not that straight forward. The cost of making things light--while certainly non-zero--isn't actually a huge line item in the overall budget of a space mission. Light = $$$ just happens to be a fairly intuitive concept and as a result its place on the gut-feel pareto is often higher than it should be...

Jumping to the end result, the hypothesis that significantly lower launch costs will facilitate lower spacecraft costs [ostensibly by allowing mass to solve 'expensive' problems like structural performance, radiation exposure, thermal management, etc] won't pass muster when practical variables are applied. It mostly comes down to the fact that very large investments have correspondingly large expectations. It doesn't really matter how much someone is saving on the launch vehicle cost or even how much they might save one day on the space hardware cost--for a very long time to come its still going to be a massive investment to put hardware into space. And if one is making a massive investment--even if its half the magnitude of an investment they've made before--they'll demand that investment be maximized by a solution that's ultra reliable and that pushes performance to the limit.

That will be the case for as long as we build space hardware terrestrially and put that hardware into orbit on chemical rockets.

On orbit manufacturing needs raw materials.

Exactly! And that is a perfect match for cheap reusable rockets***.

***Until humans develop a solution that doesn't require burning things to put other things into space, not unlike the progress we're making with BEV vs ICE...

 

[808?]

 

[swaltner]

Nicely done on the montage SpaceX!

 

[Nikxice]

The radar glitch segment at 1:21 is pretty wild. It's listed as having occurred in May 2016, corresponding with Thiacom 8. Initially I thought for sure I was watching the explanation for the missing BulgariaSat-1 landing video from this past June. Elon possibly still saving that one for the "Best Of" highlight reel.

 

[Grendal]

It is nail biting to watch the booster slide across the ASDS! That is absolutely crazy.

 

[jkn]

First, I'll apologize for initiating the sideways turn this thread has taken from @rachelawilson's original post. As someone in the space biz I completely agree that Its vital for us look beyond financial and technical details of The Now and contemplate the broader significance of what's going down. So thanks for that!



Unfortunately, the analysis is not that straight forward. The cost of making things light--while certainly non-zero--isn't actually a huge line item in the overall budget of a space mission. Light = $$$ just happens to be a fairly intuitive concept and as a result its place on the gut-feel pareto is often higher than it should be...

Jumping to the end result, the hypothesis that significantly lower launch costs will facilitate lower spacecraft costs [ostensibly by allowing mass to solve 'expensive' problems like structural performance, radiation exposure, thermal management, etc] won't pass muster when practical variables are applied. It mostly comes down to the fact that very large investments have correspondingly large expectations. It doesn't really matter how much someone is saving on the launch vehicle cost or even how much they might save one day on the space hardware cost--for a very long time to come its still going to be a massive investment to put hardware into space. And if one is making a massive investment--even if its half the magnitude of an investment they've made before--they'll demand that investment be maximized by a solution that's ultra reliable and that pushes performance to the limit.

That will be the case for as long as we build space hardware terrestrially and put that hardware into orbit on chemical rockets.



Exactly! And that is a perfect match for cheap reusable rockets***.

***Until humans develop a solution that doesn't require burning things to put other things into space, not unlike the progress we're making with BEV vs ICE...

How large fraction of cost are development and design costs? If we build ten Cassini probes instead of one, total costs will not be 10* higher. How much does it cost to increase mission reliability from 50% to 95%? (or whatever they use). With 10 satellites 50% reliability would not be to bad. We'll send 2 to Jupiter, 2 to Saturn, 2 Uranus, 2 to Neptune and remaining 2 to Pluto. With BFR probes can have more fuel and cut years from travel time, increasing reliability. Some will fail, but we get much more science data back. (Not certain where, but does it matter?)

I don't suggest lowering costs simply by adding mass. For example: That thruster is too large for our mission, but it is tested, we have mass to spare, so we use it and save money. Repeat for all components. When those standard components will be used more often, unit costs will come down.

I don't know anything about manufacturing satellites, but with everything else unit costs will come down with increasing manufacturing rate. Our civilization is based on that!

'In space manufacturing' does not make sense, if raw materials are coming from Earth. Final assembly, check, tuning and perhaps repair on orbit would be useful even with current tech.

 

[bxr140]

I don't know anything about manufacturing satellites, but with everything else unit costs will come down with increasing manufacturing rate.

Of course it will--that's volume production 101; that, as one can intuitively imagine, is one of the main conduits to bring space hardware costs down. See: Starlink and Oneweb.

What's important is that you not confuse the benefit of volume manufacturing with the idea that cheap rockets will drive down space hardware costs.

Make no mistake, the industry is making serious strides toward lowering the cost of spacecraft in a couple different ways, but make no mistake--its still a LOT of money, and even if you magically drop the cost by an order of magnitude, its still a lot of money. The problem is that the short and medium term future of space hardware won't get anywhere near that much cheaper, even when distributing performance and reliability over a volume of units. The cost of space hardware will come down, but it will do so on its own, not as a result of lower launch costs. They are very unrelated. Its all about maximizing performance and reliability of the space hardware.

Not that it matters much on the internet, but I happen to know a thing or three about manufacturing satellites (or at least I like to think so...) and reducing the cost of satellites by implementing innovative and volume production concepts is, among other things, literally my job description.

In space manufacturing' does not make sense, if raw materials are coming from Earth.

On the contrary, it makes a LOT of sense. If you can send up rustbucket falcon 9's full of aluminum/titanium powder (for structure printing), reels of components for CCAs (resistors, diodes, etc.), and stuff like that, you completely change the paradigm of what a spacecraft looks like and how it gets designed. Mass becomes a significantly reduced concern, volume/shape is essentially a non-issue, structural cohesion just needs to stay together in zero-g, etc. Entire design cycles essentially go away when you eliminate the launch environment. Only rudimentary design verification would be required on the ground for a ground-prototype. There's a host of other benefits, from being able to more easily use non-volatile propellants to turning the insurance community on its head to rapid timeline-to-service for a vehicle. Not being grounded by launch delays...all sorts of stuff.

Final assembly, check, tuning and perhaps repair on orbit would be useful even with current tech.

Sort of. We're damn near there with final assembly--the tech exists, it just needs to fly. It helps out a TON with packaging, because you no longer have to figure out how to stow all your gizmos inside the fairing envelope and then also square the geometry so they deploy to wherever you need them.

We're almost there with on orbit repair--it will actually start with some refueling demo missions, then go into service to extend the mission life of any on-orbit asset. Eventually it may turn into repairing/replacing non-op parts or adding capacity/performance upgrades, etc. The reason for this is exactly my point about the cost of space hardware--there's potentially ROI in repairing an on orbit asset because they cost so friggin much money, and prices are NOT going to come down drastically any time soon. Again, regardless of launch cost, if I might beat that horse...

Tuning however is a ways away, and possibly won't ever be a thing. Current tech is very analog--physically adjusting pathway lengths for phasing, mechanically shimming optical surfaces into focus, select-at-test attenuators for pathway gain balancing, of course the black magic art of tuning screws, etc. A lot of that isn't even automated on the ground; automating that on orbit will be a bit of an ordeal. More likely tuning will be OBE because at some point in the near future as more and more components go digital (high power amplifiers are often still vacuum tubes ) all that manual work goes away anyway and 'tuning' is just updating firmware.[/QUOTE]

 

[landis]

The cost of space hardware will come down, but it will do so on its own, not as a result of lower launch costs. They are very unrelated. Its all about maximizing performance and reliability of the space hardware

I would expect SpaceX to apply the same approach SpaceX it used to reduce the cost of building (e.g. testing and using COTS parts rather than space/MIL spec parts) to building the satellites too.

Maximizing performance, reliability (and cost) of the space hardware is a SpaceX 'core competency'.

That sounds related to me...

 

[bxr140]

I would expect SpaceX to apply the same approach SpaceX it used to reduce the cost of building (e.g. testing and using COTS parts rather than space/MIL spec parts) to building the satellites too.

That's absolutely right, and that's what the whole industry is trying to do. What's important is to not conflate methodology and causality.

 

[jkn]

Of course it will--that's volume production 101; that, as one can intuitively imagine, is one of the main conduits to bring space hardware costs down. See: Starlink and Oneweb.

What's important is that you not confuse the benefit of volume manufacturing with the idea that cheap rockets will drive down space hardware costs.

Make no mistake, the industry is making serious strides toward lowering the cost of spacecraft in a couple different ways, but make no mistake--its still a LOT of money, and even if you magically drop the cost by an order of magnitude, its still a lot of money. The problem is that the short and medium term future of space hardware won't get anywhere near that much cheaper, even when distributing performance and reliability over a volume of units. The cost of space hardware will come down, but it will do so on its own, not as a result of lower launch costs. They are very unrelated. Its all about maximizing performance and reliability of the space hardware.

Not that it matters much on the internet, but I happen to know a thing or three about manufacturing satellites (or at least I like to think so...) and reducing the cost of satellites by implementing innovative and volume production concepts is, among other things, literally my job description.



On the contrary, it makes a LOT of sense. If you can send up rustbucket falcon 9's full of aluminum/titanium powder (for structure printing), reels of components for CCAs (resistors, diodes, etc.), and stuff like that, you completely change the paradigm of what a spacecraft looks like and how it gets designed. Mass becomes a significantly reduced concern, volume/shape is essentially a non-issue, structural cohesion just needs to stay together in zero-g, etc. Entire design cycles essentially go away when you eliminate the launch environment. Only rudimentary design verification would be required on the ground for a ground-prototype. There's a host of other benefits, from being able to more easily use non-volatile propellants to turning the insurance community on its head to rapid timeline-to-service for a vehicle. Not being grounded by launch delays...all sorts of stuff.



Sort of. We're damn near there with final assembly--the tech exists, it just needs to fly. It helps out a TON with packaging, because you no longer have to figure out how to stow all your gizmos inside the fairing envelope and then also square the geometry so they deploy to wherever you need them.

We're almost there with on orbit repair--it will actually start with some refueling demo missions, then go into service to extend the mission life of any on-orbit asset. Eventually it may turn into repairing/replacing non-op parts or adding capacity/performance upgrades, etc. The reason for this is exactly my point about the cost of space hardware--there's potentially ROI in repairing an on orbit asset because they cost so friggin much money, and prices are NOT going to come down drastically any time soon. Again, regardless of launch cost, if I might beat that horse...

Tuning however is a ways away, and possibly won't ever be a thing. Current tech is very analog--physically adjusting pathway lengths for phasing, mechanically shimming optical surfaces into focus, select-at-test attenuators for pathway gain balancing, of course the black magic art of tuning screws, etc. A lot of that isn't even automated on the ground; automating that on orbit will be a bit of an ordeal. More likely tuning will be OBE because at some point in the near future as more and more components go digital (high power amplifiers are often still vacuum tubes ) all that manual work goes away anyway and 'tuning' is just updating firmware.

I agree with most you write, but not with this:
"What's important is that you not confuse the benefit of volume manufacturing with the idea that cheap rockets will drive down space hardware costs."

Can you answer this: Assuming only one satellite is build, what is split between design costs and building costs? 50/50%, 90/10% ?

Then we build 10 identical satellites. Design costs will not increase. Total manufacturing costs will increase, but slightly less than 10 fold.

In space it might be possible to make (metal) foams or alloys impossible on Earth. 3D printing flimsy structure could also make sense. As long as electronics is on 2D boards space manufacturing does not give advantage. 2D board is easy to package for transport. Light components survive acceleration easily. Massive ones can be added in orbit. Boards can then be assembled into 3D frame in space. Still mainly final assembly.

On orbit repair has been done for Hubble telescope and space station. We need space station doing satellite service and solar electric space tug to fetch and return satellites.

 

[ccutrer]

On orbit repair has been done for Hubble telescope and space station. We need space station doing satellite service and solar electric space tug to fetch and return satellites.

Those were all done by astronauts, no? Seems like the expense of sending a human up is probably almost as much sending a new satellite up (unless of course we get to the point that we always have a lot of humans up there just jetting around to all the satellites anyway). I'm pretty sure @bxr140 was referring to an automated repair bot of some sort that has some materials and tools and can do everything via remote control.