SpaceX Starship/Super Heavy Engineering General Thread 4

[livingjw]

.....

- Stringers and frames contribute little to nothing in resisting hoop loads.

....
John

Nevertheless, frames are able to perceive some part of ring loads.

Yes, near the frame, but tank wall will fail first away from the frame.

John

[InterestedEngineer]

.....

- Stringers and frames contribute little to nothing in resisting hoop loads.

....
John

Nevertheless, frames are able to perceive some part of ring loads.

Yes, near the frame, but tank wall will fail first away from the frame.

John

To a non MechE (me), it seems unintuitive that hoops don't help with hoop stress.

Is it because the welds of the hoops are in tension instead of compression?

[edzieba]

.....

- Stringers and frames contribute little to nothing in resisting hoop loads.

....
John

Nevertheless, frames are able to perceive some part of ring loads.

Yes, near the frame, but tank wall will fail first away from the frame.

John

To a non MechE (me), it seems unintuitive that hoops don't help with hoop stress.

Is it because the welds of the hoops are in tension instead of compression?

It's because hoops and stringers help with buckling loads, but internal pressurisation is not a buckling load.

[PreferToLurk]

.....

- Stringers and frames contribute little to nothing in resisting hoop loads.

....
John

Nevertheless, frames are able to perceive some part of ring loads.

Yes, near the frame, but tank wall will fail first away from the frame.

John

To a non MechE (me), it seems unintuitive that hoops don't help with hoop stress.

Is it because the welds of the hoops are in tension instead of compression?

It's because hoops and stringers help with buckling loads, but internal pressurisation is not a buckling load.

Yes, we understand that's what they are designed to do.  The question is why don't they ALSO help with hoop stress, since they are all welded together? 

[InterestedEngineer]

.....

- Stringers and frames contribute little to nothing in resisting hoop loads.

....
John

Nevertheless, frames are able to perceive some part of ring loads.

Yes, near the frame, but tank wall will fail first away from the frame.

John

I think maybe I can answer this for myself, but would love some confirmation other than "that's stuff is for compression load".

First definition per Wikipedia:

The hoop stress is the force over area exerted circumferentially (perpendicular to the axis and the radius of the object) in both directions on every particle in the cylinder wall. It can be described as:

Adding internal hoop rings (like hoops on a barrel, except internally) doesn't help, just moves the failure point, because there will be areas with no hoops, and it is "every particle"

Additionally the welds of the hoops to the walls likely weaken those areas, so if there is any help from the hoop it is offset by the reduction in strength of the weld.

To cross check this:  Would external hoops help with hoop stress, or not?  Assume kept in place by compression, not by welds.

I think the answer is "no", but that's where I'd like some confirmation

[sevenperforce]

First definition per Wikipedia:

The hoop stress is the force over area exerted circumferentially (perpendicular to the axis and the radius of the object) in both directions on every particle in the cylinder wall. It can be described as:

Adding internal hoop rings (like hoops on a barrel, except internally) doesn't help, just moves the failure point, because there will be areas with no hoops, and it is "every particle"

Additionally the welds of the hoops to the walls likely weaken those areas, so if there is any help from the hoop it is offset by the reduction in strength of the weld.

To cross check this:  Would external hoops help with hoop stress, or not?  Assume kept in place by compression, not by welds.

I think the answer is "no", but that's where I'd like some confirmation

Assuming that Wikipedia article is correct, then I would say yes, external hoops would help with hoop stress. Even though the stress is being exerted from every point perpendicular to the radius and axis of the cylinder, such stress would tend to produce an increase in the circumference of the cylinder, and an external hoop would constrain that.

[livingjw]

- Under internal pressure the stresses near the frame are lower due to the frame taking up some of the load.

- Move away from the frame and all you have is the wall it self to resist the pressure. Wall material will fail in tension here long before it fails near the frames.

John

[PreferToLurk]

- Under internal pressure the stresses near the frame are lower due to the frame taking up some of the load.

- Move away from the frame and all you have is the wall it self to resist the pressure. Wall material will fail in tension here long before it fails near the frames.

John

That makes sense. I was incorrectly thinking about it as if the supports would be able to help divide the load. But the load is being applied to every point simultaneously, making it impossible to divide it across extra load paths.

Kind of a basic engineering concept but helpful for enthusiasts like me, thanks.

Edit: well, some forces are applied everywhere, but the head pressures will obviously have a gradient.  Similar enough principle though - no splitting of the load allowed.

[InterestedEngineer]

- Under internal pressure the stresses near the frame are lower due to the frame taking up some of the load.

- Move away from the frame and all you have is the wall it self to resist the pressure. Wall material will fail in tension here long before it fails near the frames.

John

That makes sense. I was incorrectly thinking about it as if the supports would be able to help divide the load. But the load is being applied to every point simultaneously, making it impossible to divide it across extra load paths.

Kind of a basic engineering concept but helpful for enthusiasts like me, thanks.

Edit: well, some forces are applied everywhere, but the head pressures will obviously have a gradient.  Similar enough principle though - no splitting of the load allowed.

Doing some research on coopery the hoops are not there for hoop strength, they are for structural integrity when the wood swells up from getting wet.  This forces the joints between the wood pieces to close.  Their function for wood barrels is unique to the nature of wood.

Calling a force 'hoop stress' for metal tubes creates a bit of confusion in the amateur.

[livingjw]

- Under internal pressure the stresses near the frame are lower due to the frame taking up some of the load.

- Move away from the frame and all you have is the wall it self to resist the pressure. Wall material will fail in tension here long before it fails near the frames.

John

That makes sense. I was incorrectly thinking about it as if the supports would be able to help divide the load. But the load is being applied to every point simultaneously, making it impossible to divide it across extra load paths.

Kind of a basic engineering concept but helpful for enthusiasts like me, thanks.

Edit: well, some forces are applied everywhere, but the head pressures will obviously have a gradient.  Similar enough principle though - no splitting of the load allowed.

Doing some research on coopery the hoops are not there for hoop strength, they are for structural integrity when the wood swells up from getting wet.  This forces the joints between the wood pieces to close.  Their function for wood barrels is unique to the nature of wood.

Calling a force 'hoop stress' for metal tubes creates a bit of confusion in the amateur.

- I have to disagree.

- Barrel hoops do counter internal pressure. They are stressed in tension by the transfer of pressure forces applied to the staves by the internal pressure.

- The wood is not even joined together between the staves. The hoops are the only thing holding the staves together. Even if you glued the staves together, they still would be extremely weak since the internal pressure is loading them up perpendicular to their grain.

- Internal pressure is first applied to the staves.
- Staves have thickness and can support the pressure load through bending between the hoops.
- All of the pressure loads are transferred to the hoops.

- A thin metal skin cannot support bending loads. (think of a balloon, especially a long thin balloon i.e. a rocket tank)
- A thin metal skin must first deform to create curvature in order to support a pressure load.

John

[RoboGoofers]

The adage "a chain is only as strong as the weakest link" applies here.

or another example: If you put a rubber band around a balloon, it would still burst at the same pressure.

[Tangilinear Interjar]

Here's a great picture of where hoops are used to reinforce, note the change in spacing.

I wonder how much of this is to also manage the water hammer effect during shutdown?

Picture from Boca Chica Gal.

[livingjw]

Here's a great picture of where hoops are used to reinforce, note the change in spacing.

I wonder how much of this is to also manage the water hammer effect during shutdown?

Picture from Boca Chica Gal.

- Hoops are there to stabilize the down-comer when the surrounding LOx pressure is higher than the LCH4 pressure in the tube. I believe SpaceX had one crushed.

- LOx pressure at the bottom of the tank is higher than near the top, so more stabilizing hoops are needed to guard against buckling.

- No effect on hammer wave inside the tube. The hoops are on the outside.

John

[Tangilinear Interjar]

Here's a great picture of where hoops are used to reinforce, note the change in spacing.

I wonder how much of this is to also manage the water hammer effect during shutdown?

Picture from Boca Chica Gal.

- Hoops are there to stabilize the down-comer when the surrounding LOx pressure is higher than the LCH4 pressure in the tube. I believe SpaceX had one crushed.

- LOx pressure at the bottom of the tank is higher than near the top, so more stabilizing hoops are needed to guard against buckling.

- No effect on hammer wave inside the tube. The hoops are on the outside.

John

I'm pretty sure that squished downcommer was caused by pulling a vacuum inside it, not over pressure of the LOx tank.

When would the pressure inside the downcommer be lower than the LOx tank? I would expect that except for special testing cases that downcommer should always be filled higher than the LOx level.

Regarding water hammer, I'm not taking about interfering with a pressure wave, I'm talking about dealing with the pressure spike created by the momentum of the fuel suddenly being restricted due to a rapid change of the volume of flow.

Ever seen a burst copper pipe caused by water hammer? It happens (usually) closest to the valve that caused it and I bet a series of small reinforcing rings near the valve would stop the burst.

I've built a number of booster pump stations for municipal water systems and they use large surge tanks basically identical to what is being shipped to Boca Chica right now. Just shutting down one pump in the pump house and reducing flow by 20 percent creates a huge surge. Shutting down 33 engines at once will be a pretty significant change in momentum!

[DanClemmensen]

I've built a number of booster pump stations for municipal water systems and they use large surge tanks basically identical to what is being shipped to Boca Chica right now. Just shutting down one pump in the pump house and reducing flow by 20 percent creates a huge surge. Shutting down 33 engines at once will be a pretty significant change in momentum!

That sounds like a great argument in favor of a sequenced shutdown. For 33 engines, how long would such a sequence need be to mitigate the hammer effect?

[TrueBlueWitt]

Quick thought for Manned Starship/HLS

In space the Ship LOX tank should still have a large quantity of gaseous oxygen.

Since you're carrying this along, could this be tapped off for use in Space Suits?

Second part, Could the residual GOX in that tank be repurposed for Life Support Use? Swap back in C02 that's scrubbed/cooled to maintain pressure. Or just use the boiloff.

[DanClemmensen]

Quick thought for Manned Starship/HLS

In space the Ship LOX tank should still have a large quantity of gaseous oxygen.

Since you're carrying this along, could this be tapped off for use in Space Suits?

Second part, Could the residual GOX in that tank be repurposed for Life Support Use? Swap back in C02 that's scrubbed/cooled to maintain pressure. Or just use the boiloff.

A human needs about 0.84 Kg of oxygen a day.  A Raptor 2 engine needs about 536 kg/s at full thrust, so one second of Raptor 2 will supply about 638 person-days of oxygen. I don't think recycling will be needed. although you might be able to use the CO2 for evaporative cooling in the EVA suits.

[tbellman]

Quick thought for Manned Starship/HLS

In space the Ship LOX tank should still have a large quantity of gaseous oxygen.

Since you're carrying this along, could this be tapped off for use in Space Suits?

Kathy Lueders' source selection statement for HLS Option A does mention "Additionally, SpaceX's design allows for the sourcing of excess propellant, which will provide crew with a large reserve supply of life support consumables in the event of a contingency event".  So it appears someone already thought of that very idea. 

Second part, Could the residual GOX in that tank be repurposed for Life Support Use? Swap back in C02 that's scrubbed/cooled to maintain pressure. Or just use the boiloff.

Backfilling with carbon dioxide doesn't sound like a good idea.  There's a good chance that it will just freeze out and form ice (freezing point approximately 100°C higher than oxygen's boiling point) in the tank.  And the amount of pressure lost by using up oxygen for life support will be quite small, as Dan already mentioned.

[livingjw]

Here's a great picture of where hoops are used to reinforce, note the change in spacing.

I wonder how much of this is to also manage the water hammer effect during shutdown?

Picture from Boca Chica Gal.

- Hoops are there to stabilize the down-comer when the surrounding LOx pressure is higher than the LCH4 pressure in the tube. I believe SpaceX had one crushed.

- LOx pressure at the bottom of the tank is higher than near the top, so more stabilizing hoops are needed to guard against buckling.

- No effect on hammer wave inside the tube. The hoops are on the outside.

John

I'm pretty sure that squished downcommer was caused by pulling a vacuum inside it, not over pressure of the LOx tank.

When would the pressure inside the downcommer be lower than the LOx tank? I would expect that except for special testing cases that downcommer should always be filled higher than the LOx level.

Regarding water hammer, I'm not taking about interfering with a pressure wave, I'm talking about dealing with the pressure spike created by the momentum of the fuel suddenly being restricted due to a rapid change of the volume of flow.

Ever seen a burst copper pipe caused by water hammer? It happens (usually) closest to the valve that caused it and I bet a series of small reinforcing rings near the valve would stop the burst.

I've built a number of booster pump stations for municipal water systems and they use large surge tanks basically identical to what is being shipped to Boca Chica right now. Just shutting down one pump in the pump house and reducing flow by 20 percent creates a huge surge. Shutting down 33 engines at once will be a pretty significant change in momentum!

- To collapse the downcomer, all that is needed is a sufficiently large pressure difference. Doesn't matter how it came about.

- In this situation, the pressure difference is caused by the combined effects of ullage pressure and head pressure. Don't need a vacuum. Don't know why you think that.

- The LCH4 column height can be higher than the LOx height and still have a lower total pressure, due to lower ullage and lower density.

- If they filled the LOx tank and reduced the ullage pressure of the LCH4 tank and or emptied the LCH4 tank, this would set up the conditions for collapse of the downcomer.

- A snafu during the test could have caused the downcomer collapse, or they might have been testing the limiting case for collapse of the downcomer. They added more external reinforcement rings to the downcomer after the collapse, which implies that they expect this condition to occur occasionally during operation.

- How does reinforcement frames/rings outside of the downcomer effect the flow of LCH4 inside? I don't understand the connection you are trying to get at.

John

[Nomadd]

- A snafu during the test could have caused the downcomer collapse, or they might have been testing the limiting case for collapse of the downcomer.
John

The word "miscommunication" was used.