Hm, this isn’t intuitive to me. How could a rocket that was designed to withstand the pressures and conditions of space not be able to take some salt water? And what about just adding a layer of coating that would protect it?
When it’s closed and pressurized you have a very hard time crushing it. The internal pressure is converted to a force of tension that resists deformation. Once it’s been opened, you can crush it with one hand from the side. But it’s much stronger along the axis of the cylinder, since the force is directed through all the material rather than deforming it inwards.
A rocket if scaled down to the size of a coke can has walls much thinner than a coke can, and is much longer relative to its width. You can create great torques by hitting the sides to bend it, or crush it inwards. Imagine the force of tens of tons of water suddenly slapping onto the side of this tank as waves lap around, unevenly across multiple parts of the tank.
Consider a rocket.
It must, with the least possible amount of mass, generate a high acceleration along its direction of motion while subtending a very small surface area in that direction of motion. This dictates that it is long and thin, and able to withstand high forces along that long axis. But every kilogram you add to its mass is one kilogram you can’t get to orbit, or a couple more kilograms of fuel. You make it withstand the forces of its operational environment well. Other forces not so much. As far as I know most rockets get a good deal of their shear-strength perpendicular to the long axis from pressure within their tanks—some cannot even be held sideways without being pressurized. I’ve seen figures to the effect that more than half of the structural strength of a falcon 9 tank comes from it being pressurized by inert gases during operation.
Optimizing for its operational environment plus mass puts limits on how much you can optimize for other things. There is no one abstract ‘durability’ factor.
Salt water does not play nice with metals, especially fine tubes and components that will be channeling ridiculous energies and god knows how many RPM and very fine precision. It corrodes, it dries up and gums joints, etc. Shuttle SRBs were a different story—they were big, dumb, thick, wide metal tubes filled with fuel that burned like a sparkler, not fine actuators of active control. If someone could find a simple coating that could protect complex powerful machines from the damage of salt water without tradeoffs, I’m sure the navy would LOVE to know about it.
So I (now) understand that rockets are designed to be thin and light to reduce drag and gravitational forces. As far as the cost-benefit of adding a protective layer, the cost is the added drag/gravitational forces, but the benefit seems to be huge (able to land it in water). Is it really that hard to generate the necessary propulsion forces?
Obviously the answer is “yes”, but that goes against what makes intuitive sense to me. Can you explain?
I think part of the problem is a fundamental misunderstanding of what parachuting into the ocean does to a rocket motor. The motors are the expensive part of the first stage; I don’t know exact numbers, but they are the complicated, intricate, extremely-high-precision parts that must be exactly right or everything goes boom. The tank, by comparison, is an aluminum can.
The last landing attempt failed because a rocket motor’s throttle valve had a bit more static friction than it should, and stuck open a moment too long. SpaceX’s third launch attempt—the last failed launch they’ve had, many years ago with the Falcon 1 - was because the motor didn’t shut off instantly before stage separation, like it should have. As far as I know, people still don’t know why Orbital ATK (FKA Orbital Science)’s last launch attempt failed, except that it was obviously an engine failure. We talk about rocket science, but honestly the theoretical aspects of rocketry aren’t that complicated. Rocket engineering, though, that’s a bloody nightmare. You get everything as close to perfect as you can, and sometimes it still fails catastrophically and blows away more value than most of the people reading this thread will earn in their lifetimes, leaving virtually nothing to tell the tale of what happened.
What does all that have to do with parachute recovery of booster stages? Well, once you’ve dunked those motors in saltwater, they’re a write-off. They can’t be trusted to ever again operate perfectly without fairly literally rebuilding them, which defeats most of the purpose of recovering the booster.
There’s nothing you could coat a rocket motor with that would both survive that motor operating and make it economical to re-use the motor after plunging into the ocean. The closest thing I can think of would be some kind of protective bubble that expands to protect the motors from the ocean once their job is done. It would need to be watertight, impact-resistant (the rocket still hits the water pretty hard, even with parachutes), able to deploy around the motors reliably, avoid causing a bending moment that collapses the tank (which has minimal pressure, because its fuel is depleted and any excess pressurizing agent you carry is wasted weight to orbit), and able to operate after being exposed to the environment in close proximity to a medium-lift rocket’s primary launch motors. Maybe it’s possible, but I can’t think of how to do reliably enough to be worth the added cost on launch.
Each additional kilogram of the rocket is probably extremely expensive. I don’t how much weight would that extra protective layer add, but I can imagine it could more than double the weight of the rocket, and the weight of the fuel it needs, etc.
Look at the Tsiolkovsky rocket equation—a rocket’s delta-v (velocity change potential) is proportional to the log of its mass ratio (its mass with fuel divided by its mass without fuel). For modern rockets, that means about twenty kilos of fuel for every kilo of anything else (the rocket included). You really don’t want to add structural mass if there’s any way to avoid it.
Just guessing: In space the pressure comes from inside, in water from outside; it is easier to protect in one direction than in both. In space the pressure difference is at most one atmosphere, in water it depends on how deep you submerge. The salt water also attacks the metal chemically.
Hm, this isn’t intuitive to me. How could a rocket that was designed to withstand the pressures and conditions of space not be able to take some salt water? And what about just adding a layer of coating that would protect it?
Consider a Coke can.
When it’s closed and pressurized you have a very hard time crushing it. The internal pressure is converted to a force of tension that resists deformation. Once it’s been opened, you can crush it with one hand from the side. But it’s much stronger along the axis of the cylinder, since the force is directed through all the material rather than deforming it inwards.
A rocket if scaled down to the size of a coke can has walls much thinner than a coke can, and is much longer relative to its width. You can create great torques by hitting the sides to bend it, or crush it inwards. Imagine the force of tens of tons of water suddenly slapping onto the side of this tank as waves lap around, unevenly across multiple parts of the tank.
Consider a rocket.
It must, with the least possible amount of mass, generate a high acceleration along its direction of motion while subtending a very small surface area in that direction of motion. This dictates that it is long and thin, and able to withstand high forces along that long axis. But every kilogram you add to its mass is one kilogram you can’t get to orbit, or a couple more kilograms of fuel. You make it withstand the forces of its operational environment well. Other forces not so much. As far as I know most rockets get a good deal of their shear-strength perpendicular to the long axis from pressure within their tanks—some cannot even be held sideways without being pressurized. I’ve seen figures to the effect that more than half of the structural strength of a falcon 9 tank comes from it being pressurized by inert gases during operation.
Optimizing for its operational environment plus mass puts limits on how much you can optimize for other things. There is no one abstract ‘durability’ factor.
Salt water does not play nice with metals, especially fine tubes and components that will be channeling ridiculous energies and god knows how many RPM and very fine precision. It corrodes, it dries up and gums joints, etc. Shuttle SRBs were a different story—they were big, dumb, thick, wide metal tubes filled with fuel that burned like a sparkler, not fine actuators of active control. If someone could find a simple coating that could protect complex powerful machines from the damage of salt water without tradeoffs, I’m sure the navy would LOVE to know about it.
Great answer!
I see, that makes much more sense now. Thank you!
So I (now) understand that rockets are designed to be thin and light to reduce drag and gravitational forces. As far as the cost-benefit of adding a protective layer, the cost is the added drag/gravitational forces, but the benefit seems to be huge (able to land it in water). Is it really that hard to generate the necessary propulsion forces?
Obviously the answer is “yes”, but that goes against what makes intuitive sense to me. Can you explain?
I think part of the problem is a fundamental misunderstanding of what parachuting into the ocean does to a rocket motor. The motors are the expensive part of the first stage; I don’t know exact numbers, but they are the complicated, intricate, extremely-high-precision parts that must be exactly right or everything goes boom. The tank, by comparison, is an aluminum can.
The last landing attempt failed because a rocket motor’s throttle valve had a bit more static friction than it should, and stuck open a moment too long. SpaceX’s third launch attempt—the last failed launch they’ve had, many years ago with the Falcon 1 - was because the motor didn’t shut off instantly before stage separation, like it should have. As far as I know, people still don’t know why Orbital ATK (FKA Orbital Science)’s last launch attempt failed, except that it was obviously an engine failure. We talk about rocket science, but honestly the theoretical aspects of rocketry aren’t that complicated. Rocket engineering, though, that’s a bloody nightmare. You get everything as close to perfect as you can, and sometimes it still fails catastrophically and blows away more value than most of the people reading this thread will earn in their lifetimes, leaving virtually nothing to tell the tale of what happened.
What does all that have to do with parachute recovery of booster stages? Well, once you’ve dunked those motors in saltwater, they’re a write-off. They can’t be trusted to ever again operate perfectly without fairly literally rebuilding them, which defeats most of the purpose of recovering the booster.
There’s nothing you could coat a rocket motor with that would both survive that motor operating and make it economical to re-use the motor after plunging into the ocean. The closest thing I can think of would be some kind of protective bubble that expands to protect the motors from the ocean once their job is done. It would need to be watertight, impact-resistant (the rocket still hits the water pretty hard, even with parachutes), able to deploy around the motors reliably, avoid causing a bending moment that collapses the tank (which has minimal pressure, because its fuel is depleted and any excess pressurizing agent you carry is wasted weight to orbit), and able to operate after being exposed to the environment in close proximity to a medium-lift rocket’s primary launch motors. Maybe it’s possible, but I can’t think of how to do reliably enough to be worth the added cost on launch.
Each additional kilogram of the rocket is probably extremely expensive. I don’t how much weight would that extra protective layer add, but I can imagine it could more than double the weight of the rocket, and the weight of the fuel it needs, etc.
Look at the Tsiolkovsky rocket equation—a rocket’s delta-v (velocity change potential) is proportional to the log of its mass ratio (its mass with fuel divided by its mass without fuel). For modern rockets, that means about twenty kilos of fuel for every kilo of anything else (the rocket included). You really don’t want to add structural mass if there’s any way to avoid it.
Just guessing: In space the pressure comes from inside, in water from outside; it is easier to protect in one direction than in both. In space the pressure difference is at most one atmosphere, in water it depends on how deep you submerge. The salt water also attacks the metal chemically.