You are not logged in.
So let's assume Mission One successfully gets humans to Mars and propellant production is sucessfully under way...
What about rocket maintenance. What do you think needs to be done in terms of being able to get the pioneers back to Earth? Here are some thoughts/guesses:
1. A visual inspection of the heat resistant tiles to check for damage, and you need to replace them.
2. You need to carry out a visual inspection of the rest of the rocket's exterior to check for stone damage or anything else untoward.
3. Possibly...I don't know...do you check evey weld individually...if so do they need portable X ray equipment.
4. You need to undertake a visual inspection of the engines and possibly some sort of structural analysis.
5. You need to check all the electrical systems and life support. Presumably this needs to be done periodically throughout the sojourn on the planet, so that when you do finally come near the launch date, you can be assured that everything is, as much as possible, in good working order.
6. Testing the fin actuation would be desirable but presumably that will be impossible.
7. The tanks need to be checked for any signs of weakness before refuelling takes place and the refuelling/boil-off process needs to be monitored.
8. Boulders and rocks need to be removed from the vicinity of the Starship, so as to ensure there is no blast damage to any habs, or potentially to the rocket itself.
9. You need an equivalent of the ground control role to take place as the launch approaches. Presumably there will be great reliance on automated systems but also, presumably, every crew member will have responsibility for monitoring particular aspects of the launch countdown process. Space X will have to think about how that is organised. Ground control on Earth can help presumably but the 15 communications delay would mean, I guess, that for the last 15 minutes the crew is essentially on its own. Slight problem there - they have to be suited and booted, ready for launch but also need to monitor the launch build. I am guessing they would have to be in seat cradles but also have monitor screens in front of them.
Thoughts on other aspects of maintenance?
Looking at the above list, I think we can say:
A. There is a need to access the rocket's exterior up to a height of about 50 metres (160 feet). There are boom lifts which can reach that height and beyond:
https://www.youtube.com/watch?v=FBvWkQtzJ2Y
The sort of thing we see working on Starship construction at Boca Chica.
The one featured in the video weighs in at under 28 tons. So, disassemble perhaps, it could be taken to Mars. Obviously it would need to be Mars-rated - a costly and lengthy process I imagine.
Are there alternatives? The craft will be made of steel, so I guess some version of a "spiderman" magnetic suit might work but would be hair-raising.
B. The crew need to master Ground Control procedures.
C. How much can the crew achieve in terms of repair? Presumably electronics, piping and so on won't be a major problem. Major structural damage would be a different matter. Replacing tiles should be OK. They might be able to re-weld if necessary. But they do potentially have the choice of a couple of Starships to return. Of course they will undertaking this as EVAs which is always demanding.
I suppose all the above argues for possibly a somewhat larger crew than most of us imagine...maybe nearer 10 than 6.
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
300-series stainless steel is NOT magnetic.
GW
GW Johnson
McGregor, Texas
"There is nothing as expensive as a dead crew, especially one dead from a bad management decision"
Offline
Like button can go here
Whoops! Thanks for the clarification...pop goes another clever idea!
300-series stainless steel is NOT magnetic.
GW
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
Louis,
I think you covered most of it.
* Heat shield - visual inspection for cracks or delaminations (I would try to develop a robot to do this because it would be much lighter and easier to hoist than a human and potentially much faster to use; might simply make top-to-bottom passes around the circumference of the Starship until the entire outside of the ship has been inspected- recall that NASA eventually devised a system for Space Shuttle tile inspection and replacement that would work in a hard vacuum)
* Propellant tank cracks - eddy current and ultrasonic inspection is used on austenitic stainless steel (we need to do this every time we have an impact event, otherwise known as landing, or propellant refill event; you kinda "need" a robot for this work and some kind of apparatus to allow that robot to weld cracks from inside the tank; this robot would be as close to "Spiderman" as we're going to get and some other "shop vac" robot needs to clean up the mess after welding has been completed)
* Engine components - borescope the pump blades / pump housings / propellant feed lines, main injector plate, and ultrasonic test the channel wall nozzle (you have 3 onboard spares for everything but the nozzles if you decide to stop in LEO instead of reentering and landing on Earth and if you had to use any of them you could get rid of tons of extra mass and just swap the "good" engine assemblies onto the vacuum nozzles, since the sea level engines would not have been fired up prior to your mission and therefore should be in pristine condition)
* Electronics - SpaceX really needs to use rad-hard electronics in Mars-bound Starships, even if that increases cost (this vehicle will be in deep space far too long to rely on commodity electronics or absurdly heavy radiation shields)
* Software - a test suite needs to be created to ensure that software programs weren't scrambled by radiation events (software updates get reloaded from mission control on a routine basis)
* Electrical - there's no point in time in the mission when the power management subsystem can be completely "down", so any servicing will be done to one part of a redundant system (high quality wiring and connectors should minimize maintenance requirements; mostly, this just has to work without much servicing)
* Life support - this system would also be subject to ongoing operation and maintenance throughout the entire flight, which implies redundancy (once again, if this isn't stone cold reliable and durable we can't go to Mars to begin with)
* Hydraulics - this is not strictly part of the engine's gimbal control mechanism since the finds and landing gear would presumably be hydraulically powered as well, but if this kind of thing isn't already so highly reliable that it's not a question mark in my mind, I'd definitely have second thoughts about going at all
Electronics and software are always major considerations when those things are controlling every aspect of how the vehicle functions, or doesn't. No wand-waving is allowed here. A comprehensive test suite is required and it won't come at a significant mass penalty, even if it's somewhat costly to develop, so there's no reason not to have one.
We do need simple lightweight tarps to keep the dust off the Starship for inspection and repair. We also need simple aramid fiber ropes and Aluminum pulleys with a winch for servicing the heat shield and propellant tanks. This is how all real world vertical structures are serviced, meaning ropes and pulleys and winches.
We could certainly call that multi-purpose vehicle structures servicing robot "Spiderman" or "Spiderbot", if it makes it seem cool and high-tech to people. I think the robot needs a chassis, perhaps the size of a space suit life support backpack, with the ability to attach multiple tools (inspection cameras, welding tools, welding debris collection tools, tile replacement arm, etc).
Offline
Like button can go here
Fuel tank needs to be designed like the centaur with a bulkhead so as to be able enter the tank and that is not what musk is building even with the thoughts of the boil off landing tank internal to the main unit used to get the ship from orbit of earth once refueled to mars.
Another critical part of the tank exam is they need to be clear of fuel and oxydizer for man to even begin to think about visually observing the continued use of the ship for a trip home. The harness will be guided of course for the special exam to be performed. If tanks are not sound then examine a cargo ship and engines to convert it into a crewed vessel for trip home. Move life support and other things to it as needed.
Tiles are a problem if cracks as cargo ship most likely will not be as they will not come home initially and that means if we find any bad tiles of the crewed ship we are going to have a problem converting a cargo ship for crewed use if the tiles are different which transpiring will cause. Only way around that is to make all ships transpiring just without the use of it for the mars staying options. That would mean reserve fuel could be used in the refueling process from the cargo ships for a crewed ship refueling.
The electrical should work just fine so long as no static discharges have occured to them and the batteries have not gone bad from being in a stored state.
hydralics are a simple fluid level and leak checking.
life support systems of the ship should be is a sleeping mode so long as the cargo ships are the main system for surface use. There will be limited supplies on the return trip and the fuel reload may lessen the amount if you can not get the tanks full.
There is an xray machine that can walk the diameter of a submarine examining the hull but thats not going to so easy on mars with a near non magnetic material so something else will be needed to allow for the unit to perform the function of hull exam. Some series 300 do have some magnetic but not much as they are typically with the 200 which is also considered stainless and are used for anti corrosion for salt water applications.
Not sure if static engine tests are such a good thing but before they are a complete the exam of the chamber will be needed but its got to have a clean bill of health for the tanks other wise its not coming home even if we have good engines. If tanks are good but a bad engine is found then its got to have a unit from a cargo ship.
Offline
Like button can go here
Re exterior examination - I have seen suction pad systems for people to examine vertical surfaces. But of course this is not a very thick exterior surface, so not sure how that would work.
I think the boom lift is probably the best system.
Fuel tank needs to be designed like the centaur with a bulkhead so as to be able enter the tank and that is not what musk is building even with the thoughts of the boil off landing tank internal to the main unit used to get the ship from orbit of earth once refueled to mars.
Another critical part of the tank exam is they need to be clear of fuel and oxydizer for man to even begin to think about visually observing the continued use of the ship for a trip home. The harness will be guided of course for the special exam to be performed. If tanks are not sound then examine a cargo ship and engines to convert it into a crewed vessel for trip home. Move life support and other things to it as needed.Tiles are a problem if cracks as cargo ship most likely will not be as they will not come home initially and that means if we find any bad tiles of the crewed ship we are going to have a problem converting a cargo ship for crewed use if the tiles are different which transpiring will cause. Only way around that is to make all ships transpiring just without the use of it for the mars staying options. That would mean reserve fuel could be used in the refueling process from the cargo ships for a crewed ship refueling.
The electrical should work just fine so long as no static discharges have occured to them and the batteries have not gone bad from being in a stored state.
hydralics are a simple fluid level and leak checking.
life support systems of the ship should be is a sleeping mode so long as the cargo ships are the main system for surface use. There will be limited supplies on the return trip and the fuel reload may lessen the amount if you can not get the tanks full.
There is an xray machine that can walk the diameter of a submarine examining the hull but thats not going to so easy on mars with a near non magnetic material so something else will be needed to allow for the unit to perform the function of hull exam. Some series 300 do have some magnetic but not much as they are typically with the 200 which is also considered stainless and are used for anti corrosion for salt water applications.
Not sure if static engine tests are such a good thing but before they are a complete the exam of the chamber will be needed but its got to have a clean bill of health for the tanks other wise its not coming home even if we have good engines. If tanks are good but a bad engine is found then its got to have a unit from a cargo ship.
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
"A comprehensive test suite" - is that the sort of thing they use to diagnose engine faults in automobiles?
Louis,
I think you covered most of it.
* Heat shield - visual inspection for cracks or delaminations (I would try to develop a robot to do this because it would be much lighter and easier to hoist than a human and potentially much faster to use; might simply make top-to-bottom passes around the circumference of the Starship until the entire outside of the ship has been inspected- recall that NASA eventually devised a system for Space Shuttle tile inspection and replacement that would work in a hard vacuum)
* Propellant tank cracks - eddy current and ultrasonic inspection is used on austenitic stainless steel (we need to do this every time we have an impact event, otherwise known as landing, or propellant refill event; you kinda "need" a robot for this work and some kind of apparatus to allow that robot to weld cracks from inside the tank; this robot would be as close to "Spiderman" as we're going to get and some other "shop vac" robot needs to clean up the mess after welding has been completed)
* Engine components - borescope the pump blades / pump housings / propellant feed lines, main injector plate, and ultrasonic test the channel wall nozzle (you have 3 onboard spares for everything but the nozzles if you decide to stop in LEO instead of reentering and landing on Earth and if you had to use any of them you could get rid of tons of extra mass and just swap the "good" engine assemblies onto the vacuum nozzles, since the sea level engines would not have been fired up prior to your mission and therefore should be in pristine condition)
* Electronics - SpaceX really needs to use rad-hard electronics in Mars-bound Starships, even if that increases cost (this vehicle will be in deep space far too long to rely on commodity electronics or absurdly heavy radiation shields)
* Software - a test suite needs to be created to ensure that software programs weren't scrambled by radiation events (software updates get reloaded from mission control on a routine basis)
* Electrical - there's no point in time in the mission when the power management subsystem can be completely "down", so any servicing will be done to one part of a redundant system (high quality wiring and connectors should minimize maintenance requirements; mostly, this just has to work without much servicing)
* Life support - this system would also be subject to ongoing operation and maintenance throughout the entire flight, which implies redundancy (once again, if this isn't stone cold reliable and durable we can't go to Mars to begin with)
* Hydraulics - this is not strictly part of the engine's gimbal control mechanism since the finds and landing gear would presumably be hydraulically powered as well, but if this kind of thing isn't already so highly reliable that it's not a question mark in my mind, I'd definitely have second thoughts about going at all
Electronics and software are always major considerations when those things are controlling every aspect of how the vehicle functions, or doesn't. No wand-waving is allowed here. A comprehensive test suite is required and it won't come at a significant mass penalty, even if it's somewhat costly to develop, so there's no reason not to have one.
We do need simple lightweight tarps to keep the dust off the Starship for inspection and repair. We also need simple aramid fiber ropes and Aluminum pulleys with a winch for servicing the heat shield and propellant tanks. This is how all real world vertical structures are serviced, meaning ropes and pulleys and winches.
We could certainly call that multi-purpose vehicle structures servicing robot "Spiderman" or "Spiderbot", if it makes it seem cool and high-tech to people. I think the robot needs a chassis, perhaps the size of a space suit life support backpack, with the ability to attach multiple tools (inspection cameras, welding tools, welding debris collection tools, tile replacement arm, etc).
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
Louis,
At present, only part of the software and hardware used to diagnose engine sensor and software faults in modern passenger vehicles is included with the engine control unit. Most of what the onboard computer can "determine" is limited to circuit continuity or signal input issues and readings outside of whatever was pre-programmed as "acceptable", as it relates to some pre-established normative operating scenario. However, the test suite I made reference to comprises a set of hardware and software used to determine the health of the engine by making a more comprehensive evaluation of what the various sensor inputs actually mean. The engine's computer can and typically does record the data if a sensor's input is outside of some pre-established input range, but it can't tell you why it's out of range.
Is an electrical connection merely loose from engine vibration, is the sensor itself faulty, is some contaminant causing erroneous readings to be returned by the sensor, or is something about to fail?
That's the kind of information that an engine control unit can't tell you because there's an entire software and hardware test suite not included with the vehicle's onboard computer, along with simple human observation, that's used to determine what a sensor readout means in the grand scheme of things. Some problems are far more obvious than others and sometimes multiple issues can and do occur at the same time.
If a sensor connection was loose and the sensor itself was contaminated with conductive particulate matter, what would its readout actually mean?
I'll put it this way. As simple as mechanical temperature sensors or valves are, the springs still need to be replaced from time to time. After you add some electronics and a computer control program to the mix, have you increased or decreased the number of potential failure modes?
Offline
Like button can go here
I think I read somewhere that it's typical for a rocket - maybe it was the Saturn 5 - to launch with as many as 4,000 fault indicators. So much is being monitored and not everything will work right at launch, as in "near perfect". Some faults just affect overall performance, meaning the flight is less efficient than it should be or perhaps puts added stress on other components like gyro correction, but some faults obviously are critical and should lead to an abort.
I do see this as one of the most challenging aspects of a Mars Mission - the launch back to Earth. Imagine that small crew have to stand in for the whole of Ground Control in those last 15 or whatever minutes. It's a huge responsibility on top of the psychological pressure of their own precarious situation, perched on top of the rocket, not knowing if this first return launch will be successful.
Louis,
At present, only part of the software and hardware used to diagnose engine sensor and software faults in modern passenger vehicles is included with the engine control unit. Most of what the onboard computer can "determine" is limited to circuit continuity or signal input issues and readings outside of whatever was pre-programmed as "acceptable", as it relates to some pre-established normative operating scenario. However, the test suite I made reference to comprises a set of hardware and software used to determine the health of the engine by making a more comprehensive evaluation of what the various sensor inputs actually mean. The engine's computer can and typically does record the data if a sensor's input is outside of some pre-established input range, but it can't tell you why it's out of range.
Is an electrical connection merely loose from engine vibration, is the sensor itself faulty, is some contaminant causing erroneous readings to be returned by the sensor, or is something about to fail?
That's the kind of information that an engine control unit can't tell you because there's an entire software and hardware test suite not included with the vehicle's onboard computer, along with simple human observation, that's used to determine what a sensor readout means in the grand scheme of things. Some problems are far more obvious than others and sometimes multiple issues can and do occur at the same time.
If a sensor connection was loose and the sensor itself was contaminated with conductive particulate matter, what would its readout actually mean?
I'll put it this way. As simple as mechanical temperature sensors or valves are, the springs still need to be replaced from time to time. After you add some electronics and a computer control program to the mix, have you increased or decreased the number of potential failure modes?
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
Louis,
In practice, there's too much data for the ground control station operators to monitor. That's why, if you listen to the ground control callouts on the launch countdown sequence for the Space Shuttle that near to the end of that sequence they'll say "go for auto sequence start". That means the orbiter's onboard computer control system has control over the operation of the vehicle. The humans onboard can interface with the computer to tell the vehicle what they want to do, but then the computer decides what needs to be done to execute those instructions. The only thing the operators can do is monitor the status of critical sensor outputs. This is little different than the way FADEC-equipped engines are controlled by pilots, which are used by virtually all modern military and civilian transport aircraft. Sometimes you can override the way the engine's computer behaves, but that's generally a really bad idea.
Remember that scene from the new Robocop movie where the software engineer says, "He may think he's controlling what he's doing in combat, but in reality he's just along for the ride because there's no way his brain can process and respond to all the input fast enough."? Well, that definitely applies to computer-controlled aerospace vehicles.
There was no physical way for a pilot to keep the F-16 in coordinated flight, so a quadruple redundant digital computer system made hundreds of fine control inputs into the jet's flight control system every second to keep that thing flying nose first through the air. The F-16's inherently unstable design made it very maneuverable, but uncontrollable for a human operator without computer assistance. Although there were lots of earlier computer control systems used in military aviation and rocketry, the genesis for much of the digital computer and engine control technology was the Saturn V rocket, F-16, and Space Shuttle. When everything functions correctly, it looks like magic, but when it doesn't bad things happen so fast no human could ever keep up.
Offline
Like button can go here
All sounds right to me, and the more it's automated, the better as far as the return flight launch goes but I guess at some higher level you'll still need humans to decide whether to launch-abort. There may be automated aborts but I think the human element may still play a part. I don't know. They still do I believe in current rocket launches. But if it could all be automated I would be more than happy!
Louis,
In practice, there's too much data for the ground control station operators to monitor. That's why, if you listen to the ground control callouts on the launch countdown sequence for the Space Shuttle that near to the end of that sequence they'll say "go for auto sequence start". That means the orbiter's onboard computer control system has control over the operation of the vehicle. The humans onboard can interface with the computer to tell the vehicle what they want to do, but then the computer decides what needs to be done to execute those instructions. The only thing the operators can do is monitor the status of critical sensor outputs. This is little different than the way FADEC-equipped engines are controlled by pilots, which are used by virtually all modern military and civilian transport aircraft. Sometimes you can override the way the engine's computer behaves, but that's generally a really bad idea.
Remember that scene from the new Robocop movie where the software engineer says, "He may think he's controlling what he's doing in combat, but in reality he's just along for the ride because there's no way his brain can process and respond to all the input fast enough."? Well, that definitely applies to computer-controlled aerospace vehicles.
There was no physical way for a pilot to keep the F-16 in coordinated flight, so a quadruple redundant digital computer system made hundreds of fine control inputs into the jet's flight control system every second to keep that thing flying nose first through the air. The F-16's inherently unstable design made it very maneuverable, but uncontrollable for a human operator without computer assistance. Although there were lots of earlier computer control systems used in military aviation and rocketry, the genesis for much of the digital computer and engine control technology was the Saturn V rocket, F-16, and Space Shuttle. When everything functions correctly, it looks like magic, but when it doesn't bad things happen so fast no human could ever keep up.
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here