totm nov 2023 SpaceX Discussion Thread

[Ultimate Steve]

6 minutes ago, darthgently said:

I can’t get the image of using RCS on the ends of the catch arms to damp oscillations in the final part of the closing motion.  Crazy, I know.  Someone tell me why it wouldn’t work so I can stop thinking about it.  Honest request

I don't see why it wouldn't work. There's probably better and less complex ways to achieve the same effect, though.

[darthgently]

6 minutes ago, Ultimate Steve said:

I don't see why it wouldn't work. There's probably better and less complex ways to achieve the same effect, though.

The oscillations really seem too much to me.  What ways are easier?  My line of thought was to apply dampening force where it would be most responsive and effective, which would be at the ends of the arms.  But more mass out on the ends or anything that cluttered the avenue the booster would travel is not good either.  All I could think of was RCS.

I guess I’m just getting mentally ratholed in how to move massive stuff with so much material flex quickly with surgical precision.  It’s a very intriguing problem and I’m all ears

[mikegarrison]

Why would you use RCS? Mechanical connections to the tower would probably be a more reliable, cheaper, and more effective option.

Besides, large structures having some compliance is a good thing. What evidence do you have that they need some more damping?

Edited September 7 by mikegarrison

[Vanamonde]

Some posts removed. Stick to the topic, please. 

[tater]

 

[darthgently]

3 hours ago, mikegarrison said:

Why would you use RCS? Mechanical connections to the tower would probably be a more reliable, cheaper, and more effective option.

Besides, large structures having some compliance is a good thing. What evidence do you have that they need some more damping?

It isn't just the catch tower per se.  It is the problem of controlling flex in massive structures in general that I'm finding interesting. 

But the ends of the arms do appear to be oscillating a quarter a meter or so at least.  Not having finer control just annoys me as it could make corner cases more slippery in a catch scenario.   

The same class of problems will emerge in large rotating habitats for example.  So the idea of using RCS to dampen structural oscillations in massive structures occurred to me, that's all.  If you don't see it as interesting that's fine, I get that.

Linear electrical motors on tendons in opposition on the inside and outside of the arm seems like it would work also, but they'd need to be able to exert a lot of force

Edited September 8 by darthgently

[cubinator]

30 minutes ago, darthgently said:

It isn't just the catch tower per se.  It is the problem of controlling flex in massive structures in general that I'm finding interesting. 

But the ends of the arms do appear to be oscillating a quarter a meter or so at least.  Not having finer control just annoys me as it could make corner cases more slippery in a catch scenario.   

The same class of problems will emerge in large rotating habitats for example.  So the idea of using RCS to dampen structural oscillations in massive structures occurred to me, that's all.  If you don't see it as interesting that's fine, I get that.

Linear electrical motors on tendons in opposition on the inside and outside of the arm seems like it would work also, but they'd need to be able to exert a lot of force

RCS would probably work for dampening oscillations, but so would the stuff they use in buildings that regularly experience earthquakes.

[mikegarrison]

36 minutes ago, cubinator said:

RCS would probably work for dampening oscillations, but so would the stuff they use in buildings that regularly experience earthquakes.

I'm not sure now what you are talking about. Are you saying the whole launch tower is swaying unacceptably?

To control sway in large towers (usually from wind, not from earthquakes) they often use tuned mass dampers. Very large and heavy blocks on very large springs (and usually with some kind of damping). https://en.wikipedia.org/wiki/Tuned_mass_damper By their nature, they would not work if installed out on the arms, because the arms have to be able to swing. But any sway of the tower would likely be more visible out at the tip of the arms, I suppose.

But structures move. They have to move and flex for strength. The amount of movement has to be controlled, but trying to eliminate it is often not a good choice. Who says the tower is moving too much? This is just something someone thinks "looks wrong"?

[cubinator]

24 minutes ago, mikegarrison said:

I'm not sure now what you are talking about. Are you saying the whole launch tower is swaying unacceptably?

To control sway in large towers (usually from wind, not from earthquakes) they often use tuned mass dampers. Very large and heavy blocks on very large springs (and usually with some kind of damping). https://en.wikipedia.org/wiki/Tuned_mass_damper By their nature, they would not work if installed out on the arms, because the arms have to be able to swing. But any sway of the tower would likely be more visible out at the tip of the arms, I suppose.

But structures move. They have to move and flex for strength. The amount of movement has to be controlled, but trying to eliminate it is often not a good choice. Who says the tower is moving too much? This is just something someone thinks "looks wrong"?

No, I'm not saying there's anything wrong with the tower flexing. Just that you can build dampening into the structure itself. 

[zolotiyeruki]

Also, because the arms are actuated (and not just a rigid structure),  the PID control loop for their movement can be tuned to prevent and/or mitigate that oscillation.

[darthgently]

5 hours ago, zolotiyeruki said:

Also, because the arms are actuated (and not just a rigid structure),  the PID control loop for their movement can be tuned to prevent and/or mitigate that oscillation.

A PID loop at the "shoulder" end will involve limiting the acceleration and max rates of motion to what the hydraulics, design, and material can handle without oscillation and overshoot, which may not be acceptable for fast and accurate motion with zero overshoot.   

The arms motion got my attention because the range of oscillation seems to be close to the edge of the distance the catch hooks protrude on the booster.  Ideally there would be zero overshoot (no impacting the booster) with oscillation range at the hooks well below the protrusion of the hooks.  All achieved at reaction speeds that could ideally handle sudden wind gusts, engine failures, unplanned collisions, etc.

 

The catch arms are what got me thinking about it but a separate thread will be fine as huge orbital structures, short of being built of unobtainium, will need active dampening.

And if you've read this far, you already know that to make giant battle mechs we need to study this.

[Pthigrivi]

23 hours ago, tater said:

 

That sounds remarkably reasonable. I still think it'll be 6 before humans go given all of the other logistic challenges but I like the energy. 

[JoeSchmuckatelli]

1 hour ago, Pthigrivi said:

That sounds remarkably reasonable. I still think it'll be 6 before humans go given all of the other logistic challenges but I like the energy. 

I'm just looking forward to some huge ship making the trip.  If they even hit planet it's a win. 

I mean, even a high speed catering event would provide some data, if only for the Observer. 

But truth is I'm hoping they orbit, drop Stsrlinks and then attempt the landing 

[StrandedonEarth]

4 hours ago, Pthigrivi said:

That sounds remarkably reasonable. I still think it'll be 6 before humans go given all of the other logistic challenges but I like the energy. 

“If those [first] landings go well…” Yeah, good luck with that. And then two years between iteration attempts, unless they send a trio of different designs to see what works best.  
It seems to suggest that ISRU equipment would have to be on those first flights, to fuel later flights for the trip home. Unless the plan is for one-way trips from the start, lol

[tater]

1 hour ago, StrandedonEarth said:

“If those [first] landings go well…” Yeah, good luck with that. And then two years between iteration attempts, unless they send a trio of different designs to see what works best.  
It seems to suggest that ISRU equipment would have to be on those first flights, to fuel later flights for the trip home. Unless the plan is for one-way trips from the start, lol

Initial, uncrewed landings would be 1-way.

He said landings, plural, so I assume they might either test slightly different designs (of legs?), or different EDL profiles, that or stagger arrival times, with a goal of learning from the first, and tweaking profile as required.

I would imagine they should get some comms there, honestly they should have done so this synod. But they could at least deploy some starlinks with high-gains at the same time.

[darthgently]

6 hours ago, tater said:

I would imagine they should get some comms there, honestly they should have done so this synod. But they could at least deploy some starlinks with high-gains at the same time.

Same with Moon comms.  It’s amazing we still have blackouts on the far side.  In KSP comms are my first priority; it seems obvious but that is in just a game I suppose.  China has a relay to the dark side iirc.

Mars content: I wonder how high an altitude a stock Starlink and terminal can operate at the sacrifice of latency?  An extremely sparse net could work really well up higher as the fewer sats would still have laser LOS with their neighbors.  Maybe as few as three per orbit but anything in the single digits would be cool.  And higher up would mean fewer orbits.  The return path to earth would need some hardware change or addition either to each sat or dedicated relays 

 

[tater]

Starlink is at 550 km—which is around twice the orbital altitude of MRO.

They should make an expended stage that includes multiple Starlink 2 busses with high-gains, and additional solar panel area.

[Nuke]

10 hours ago, StrandedonEarth said:

“If those [first] landings go well…” Yeah, good luck with that. And then two years between iteration attempts, unless they send a trio of different designs to see what works best.  
It seems to suggest that ISRU equipment would have to be on those first flights, to fuel later flights for the trip home. Unless the plan is for one-way trips from the start, lol

i cant imagine it being a bad idea to ship a bunch of life support consumables and equipment there. spare parts, machine tools, construction supplies, etc.  the cost of loading up the rocket with that stuff will be minuscule in the mission budget and you need ballast for a legit test anyway. even if it ends up getting scattered in the martain soil some of it might be useful later.

9 hours ago, tater said:

Initial, uncrewed landings would be 1-way.

He said landings, plural, so I assume they might either test slightly different designs (of legs?), or different EDL profiles, that or stagger arrival times, with a goal of learning from the first, and tweaking profile as required.

I would imagine they should get some comms there, honestly they should have done so this synod. But they could at least deploy some starlinks with high-gains at the same time.

it seems like it would be better to send several so you can get more data for the next window. you cant move fast and break things if your iteration interval is 18 months. in that case its move a lot break a lot in parallel.

2 hours ago, darthgently said:

Same with Moon comms.  It’s amazing we still have blackouts on the far side.  In KSP comms are my first priority; it seems obvious but that is in just a game I suppose.  China has a relay to the dark side iirc.

Mars content: I wonder how high an altitude a stock Starlink and terminal can operate at the sacrifice of latency?  An extremely sparse net could work really well up higher as the fewer sats would still have laser LOS with their neighbors.  Maybe as few as three per orbit but anything in the single digits would be cool.  And higher up would mean fewer orbits.  The return path to earth would need some hardware change or addition either to each sat or dedicated relays 

 

fill one of the starships up with starlink: mars edition. long range comms and gps capability, and mostly for telemetry to anylize the results of landing. have it show up ahead of the rest of the swarm and deploy before it gets there. it dont need to land so use the surplus delta-v to move the eta forward.

Edited September 9 by Nuke

[Terwin]

2 hours ago, darthgently said:

Same with Moon comms.  It’s amazing we still have blackouts on the far side.  In KSP comms are my first priority; it seems obvious but that is in just a game I suppose.  China has a relay to the dark side iirc.

I thought the moon has wonky gravity that causes problems for long-term orbits, and that is why we have mars satellites before we have moon satellites.

[darthgently]

1 minute ago, Terwin said:

I thought the moon has wonky gravity that causes problems for long-term orbits, and that is why we have mars satellites before we have moon satellites.

Probably.  The lunar mascons are an annoyance but I think between not being too low and newer ion thrusters combined with using solar pressure and such it isn’t hard to maintain a stable orbit for decades.  Not sure though

China is doing it

[Meecrob]

On 9/7/2024 at 5:53 PM, Vanamonde said:

 

Hows it going, my Homie?

Bang up job, keeping us on task and topic!

-CMB out

 

How many likes do you get a day? I gotta know how to distribute them wisely.

Edited September 9 by Meecrob

[Cracktacular]

Hello everyone, I'm new here, but I can tell this is a cool place!

[tater]

Aiming for 3:38am Eastern time for launch.

[Ultimate Steve]

Polaris Dawn launch in about 4 and a half minutes.

https://www.spacex.com/launches/mission/?missionId=polarisdawn

 

[CBase]

Congrats. I miss the g-meter indication from my kerbals for human missions. Would be cool addition to the overlay.