SpaceX Starship : Texas Prototype(s) Thread 3 : Discussion

[StuffOfInterest]

The restrictions in motion are due to the length of the lever arm you have chosen. Add teeth and a worm drive to the circle you are using as an axle and the fin will have over 180 degrees of motion.

I would think 90 degrees would be enough.  Straight out to straight up.  I don't see why they would ever need to pitch the wings down.

[_MECO]

I'm interested to learn how the described "worm drive" which is supposed to eventually replace hydraulics in the production design is going to work. Right now, it looks like the method by which the flaps/fins are actuated is via a lever arm which protrudes inward from the hinge axis. Is this going to eventually turn into a section of gearing which will interface with the worm gear? I've drawn up a crude diagram in MS Paint to show what I mean. It seems like if this approach is taken there could be some limitation in the angular travel of the fin caused by the driven gear butting up against the fin root (or hull, it it's deeper inside the rocket.)

There is no reason that there would be an almost direct drive as shown in your picture. You can see typical pictures of worm gears on wikipedia. https://en.wikipedia.org/wiki/Worm_drive They typically involve the worm being connected to a full gear so there would be no limit on travel.

The worm part is spun directly by the motor, and each spin move the gear by a single tooth, yielding a very high gear ratio. The gear would be connected to a mechanism that moves the lever arm over the same range of travel as for the hydraulic system. The connecting mechanism could be some type of linear gear. https://en.wikipedia.org/wiki/Rack_and_pinion

Overall, the actual actuating part of the mechanism should be basically be identical, just based on gear ratios for direct mechanical coupling instead of hydraulic coupling.

Whether the worm gear engages directly with a part integral to the fin or not has nothing to do with limitations on fin travel, provided some toothed interface is actually how the fins are driven. Even if there is a gear down or other sort of directly rotational force transmission between the worm gear and fin, there will need to be a swept path representing some fraction of a circular gear to allow the fin to have any travel. The only way to avoid this would be to control the fin using a linkage (bottom right of crude diagram) which attaches to something like a toothed bell crank connected to the worm gear. worm gears create extremely high torque rotational motion. Whether it's more efficient to turn that rotational motion into linear motion and then back into rotational motion of the fin or to directly gear train it over is beyond me.

Oh, I correct myself. There totally are worm drives that do directly linear motion. I should know, I own a 3D printer with two of them! Anyway, this is another possible configuration. The geometry of the linkage between the driven threaded slider and lever arm of the fin can be changed to create leverage as needed.

The restrictions in motion are due to the length of the lever arm you have chosen. Add teeth and a worm drive to the circle you are using as an axle and the fin will have over 180 degrees of motion.

That may be possible if the torques involved don't incur too much stress at such a small radius. We do know that if Musk stated during Dearmoon that the forces involved are in the "meganewton range", and if we say that means one meganewton of tangential force acting on the gear interface for a ballpark simplification, and we pretend the gear on the fin is made of a high strength steel which yields at 2 GPa, then in order for the teeth not to shear (without factor of safety) the cross-sectional area of the base of a tooth would need to be 5 square centimeters for every meganewton of tangential force.

 That may not sound like much but it would essentially mean a gear with teeth a centimeter thick at the base and five centimeters tall, for every meganewton of tangential force on the gear interface. This does ignores what the torques and subsequently radii of interfaces would be like and I suppose is thus less relevant to the conversation at hand, but it's still food for thought.

[Confusador]

twitter.com/bluemoondance74/status/1182031512088334336

NOTICE: Cameron County road and beach closures- due to SpaceX activity, scheduled for:

          Wed, Oct. 23- Fri, Oct. 25
            (12pm- 8pm, each day)

co.cameron.tx.us/wp/space-x/

https://twitter.com/bluemoondance74/status/1182031924040273920

https://twitter.com/spacepadreisle/status/1182044032186359813

Definitely not MK1 rollout, likely launchpad construction and or new tank tests.

SPadre seems quite confident that it's not for Mk1, but these dates are 2 weeks out.  Do we have any evidence either way that it isn't planned to be ready to move at that point?

Musk suggested that the 20 km flight was planned for 1-2 months from 28 Sept, and I assume they'll do a tanking test before that so it will need to be at the launch site some time ahead of that.

[Johnnyhinbos]

Fin removed:

https://twitter.com/BocaChicaGal/status/1182032269256675330

That makes so much more sense - I'm relieved. I had been puzzling over how there were going to manage piping / cable runs through that area with the wings in place. It just seemed like going about it backwards. Was hoping they'd be removed after the presentation.

[Asadinator]

Fin removed:

https://twitter.com/BocaChicaGal/status/1182032269256675330

That makes so much more sense - I'm relieved. I had been puzzling over how there were going to manage piping / cable runs through that area with the wings in place. It just seemed like going about it backwards. Was hoping they'd be removed after the presentation.

I guess the fins top supports weren’t permanent either, everyone said they looked flimsy for flight.

[AndyMc]

GPI Coil Building: Building Stainless Steel Tanks on Location - New Coil Building Machine

Our new Coil building Machine can build stainless steel tanks up to 30 meters diameter and 30 meters height on an efficient and safe way. This machine uses a decoil system that unwinds and cuts the stainless steel for one hull part, whereupon it is automatically welded. The machine uses a hydraulic lift system to work from the top down. Because of the compactness of the machine, it is easy to build large, high quality tanks on location.

[mwfair]

I've been puzzling over the operating principle behind the conical shaped things that appear to be cages or slides for the legs.  They are at the aft hinge point for the fin.   Anyone have a notion of how they work, i.e. why they are quarter-cone shaped?

Also, the square tube running partway along the hinge point is puzzling.  (edited: I see now that this square tube is only on the port side) Supporting it seems to be the main purpose for the 3 or 4 triangular mounts inboard of each fin.   It has horizontal slots cut in it that dont seem to have any purpose related to mounting or fabrication.  edit: now I see holes, possibly tapped, perhaps for screwing on the fairing.
Perhaps it is beefy because its part of the substructure of the fin-root fairing, and also  provides strong airflow / heat resistance at the fin root (which is a hot spot at Mach 5 / 90 degree AoA).   But none of that explains the horizontal slot cutouts.  Any guesses?

edited to add more images, comment about tapped holes

[StuffOfInterest]

I've been puzzling over the operating principle behind the conical shaped things that appear to be cages or slides for the legs.  They are at the aft hinge point for the fin.   Anyone have a notion of how they work, i.e. why they are quarter-cone shaped?

I'm not sure about the square tubes but I've always thought that the conical structures are load spreaders for the bottom attach points of the fins.  The hard structure behind the skin may not be right at the point of the hinge so the cones would move the load up and down to where there are hard points behind the skin. 

Another possibility is that the cone shape may provide a passage for the legs once those are mounted in.  The skin in back of the cones would be cut out and the legs would then deploy outward as they slide down.

[_MECO]

I've been puzzling over the operating principle behind the conical shaped things that appear to be cages or slides for the legs.  They are at the aft hinge point for the fin.   Anyone have a notion of how they work, i.e. why they are quarter-cone shaped?

I'm not sure about the square tubes but I've always thought that the conical structures are load spreaders for the bottom attach points of the fins.  The hard structure behind the skin may not be right at the point of the hinge so the cones would move the load up and down to where there are hard points behind the skin. 

Another possibility is that the cone shape may provide a passage for the legs once those are mounted in.  The skin in back of the cones would be cut out and the legs would then deploy outward as they slide down.

If they did house the legs, why would they taper? If it were a channel which would permit a structural member to slide back and forth through it surely it would make more sense if it were of constant width, right?

[livingjw]

GPI Coil Building: Building Stainless Steel Tanks on Location - New Coil Building Machine

Our new Coil building Machine can build stainless steel tanks up to 30 meters diameter and 30 meters height on an efficient and safe way. This machine uses a decoil system that unwinds and cuts the stainless steel for one hull part, whereupon it is automatically welded. The machine uses a hydraulic lift system to work from the top down. Because of the compactness of the machine, it is easy to build large, high quality tanks on location.

Andy,

   Can you put a copy of this in the StarShip manufacturing thread?

John

[livingjw]

I've been puzzling over the operating principle behind the conical shaped things that appear to be cages or slides for the legs.  They are at the aft hinge point for the fin.   Anyone have a notion of how they work, i.e. why they are quarter-cone shaped?

I believe the leg attachment fittings are tapered because they are transferring the load to the skirt via shear.

John

[mwfair]

I've been puzzling over the operating principle behind the conical shaped things that appear to be cages or slides for the legs.  They are at the aft hinge point for the fin.   Anyone have a notion of how they work, i.e. why they are quarter-cone shaped?

I believe the leg attachment fittings are tapered because they are transferring the load to the skirt via shear.
John

Coiuld be that simple.  But the cross section of the 'taper' is circular.  Most load transfer structures (i.e. gussets) are simple plates, i.e. have planar features.   This one has two mirror image quarter cones.  i.e. exact quarter circles with radias about 12" at bottom to 2" at top.
Another oddity is the color.

[sferrin]

Looks more like a temporary tie rod to me. Loose fitting, and the body-side bracket is rather rough. Plenty of unfilled bolholes too. This is definitely where the actuator will go, but this isn't it, yet.

You know, the moment I posted that I had the same thought too. It's incredibly small for a hydraulic cylinder and I don't see any ports for working fluid on the side. Still, seeing the attachment to that mounting point/clevis gives some insight into where the actuator is going to go.

I believe Elon has stated that the fins will be electro hydraulic actuated, at least for the first few MK iterations. This means no major hydraulic plumbing - and powered via the Tesla battery packs mounted to one of the header tanks.


BTW - I am expecting to see more battery packs loaded (or already loaded I suppose) into the lower tank section to actuate the bottom fins. Especially if the actuators are DC instead of AC. Those would have to be some serious gauge wires to be able to run the distance from the bow/nose to the lower fins.

Electro-hydraulic still means the end muscle is a hydraulic actuator, right? If it's a cylinder actuating the flap you'll still need at least a couple of hoses connected to the cylinder.

I'm interested to learn how the described "worm drive" which is supposed to eventually replace hydraulics in the production design is going to work. Right now, it looks like the method by which the flaps/fins are actuated is via a lever arm which protrudes inward from the hinge axis. Is this going to eventually turn into a section of gearing which will interface with the worm gear? I've drawn up a crude diagram in MS Paint to show what I mean. It seems like if this approach is taken there could be some limitation in the angular travel of the fin caused by the driven gear butting up against the fin root (or hull, it it's deeper inside the rocket.)

Maybe we're talking about the same thing but I'd think a ball-screw would work better.   Your config with the ball-screw is more like what was thinking.  Then again if he's talking electromechanical, just swap out the ball-screw for an actuator.

[Johnnyhinbos]

Looks like they've lugged one of the fins down by the road, though not sure what side of the fence it's on...

[RoboGoofers]

I've been puzzling over the operating principle behind the conical shaped things that appear to be cages or slides for the legs.  They are at the aft hinge point for the fin.   Anyone have a notion of how they work, i.e. why they are quarter-cone shaped?

I believe the leg attachment fittings are tapered because they are transferring the load to the skirt via shear.
John

Coiuld be that simple.  But the cross section of the 'taper' is circular.  Most load transfer structures (i.e. gussets) are simple plates, i.e. have planar features.   This one has two mirror image quarter cones.  i.e. exact quarter circles with radias about 12" at bottom to 2" at top.
Another oddity is the color.

to me it just looks like bent sheet steel, not a cone. attached another view from elon's twitter.

[OTV Booster]

Looks more like a temporary tie rod to me. Loose fitting, and the body-side bracket is rather rough. Plenty of unfilled bolholes too. This is definitely where the actuator will go, but this isn't it, yet.

You know, the moment I posted that I had the same thought too. It's incredibly small for a hydraulic cylinder and I don't see any ports for working fluid on the side. Still, seeing the attachment to that mounting point/clevis gives some insight into where the actuator is going to go.

I believe Elon has stated that the fins will be electro hydraulic actuated, at least for the first few MK iterations. This means no major hydraulic plumbing - and powered via the Tesla battery packs mounted to one of the header tanks.


BTW - I am expecting to see more battery packs loaded (or already loaded I suppose) into the lower tank section to actuate the bottom fins. Especially if the actuators are DC instead of AC. Those would have to be some serious gauge wires to be able to run the distance from the bow/nose to the lower fins.

Electro-hydraulic still means the end muscle is a hydraulic actuator, right? If it's a cylinder actuating the flap you'll still need at least a couple of hoses connected to the cylinder.

I'm interested to learn how the described "worm drive" which is supposed to eventually replace hydraulics in the production design is going to work. Right now, it looks like the method by which the flaps/fins are actuated is via a lever arm which protrudes inward from the hinge axis. Is this going to eventually turn into a section of gearing which will interface with the worm gear? I've drawn up a crude diagram in MS Paint to show what I mean. It seems like if this approach is taken there could be some limitation in the angular travel of the fin caused by the driven gear butting up against the fin root (or hull, it it's deeper inside the rocket.)

Maybe we're talking about the same thing but I'd think a ball-screw would work better.   Your config with the ball-screw is more like what was thinking.  Then again if he's talking electromechanical, just swap out the ball-screw for an actuator.

Here’s what I’m seeing.

The hull cutout is way longer than it needs be if the tie rod is a stand in for an actuator of any type.

ISTM that the actuator will be behind the plate and move the plate back and forth resulting in the tie rod moving the fin. Put a piece of stainless on the plate conformal to the hull and the hole is effectively closed off but not air tight.

The actuator for early Marks can be electro/hydraulic and for later Marks straight electric with ball threads. Instead of one motor/thread, size for two and install three for redundancy.

This puts the most temperature sensitive parts in a more benign environment.

Issues:

- The link hookup looks loose. Maybe it’s just a stand in for the presentation.

- the plate looks massive. We don’t know how much it’s been hogged out on the back. The motors may have an open center with ball nut, mount to the plate and work against a fixed thread. I don’t know if this has any serious advantages but it will reduce rotating mass a bit and counter the shift of CG by moving a larger mass counter to the fin.

Phil

[mwfair]

I've been puzzling over the operating principle behind the conical shaped things that appear to be cages or slides for the legs.  They are at the aft hinge point for the fin.   Anyone have a notion of how they work, i.e. why they are quarter-cone shaped?

I believe the leg attachment fittings are tapered because they are transferring the load to the skirt via shear.
John

Coiuld be that simple.  But the cross section of the 'taper' is circular.  Most load transfer structures (i.e. gussets) are simple plates, i.e. have planar features.   This one has two mirror image quarter cones.  i.e. exact quarter circles with radias about 12" at bottom to 2" at top.
Another oddity is the color.

to me it just looks like a bent sheet, not a cone. attached another view from elon's twitter.

Which objects are you looking at?   I am looking at 2 pairs of cones on the far left of your picture, not at the contraption on the floor in the center

[RoboGoofers]

I've been puzzling over the operating principle behind the conical shaped things that appear to be cages or slides for the legs.  They are at the aft hinge point for the fin.   Anyone have a notion of how they work, i.e. why they are quarter-cone shaped?

I believe the leg attachment fittings are tapered because they are transferring the load to the skirt via shear.
John

Coiuld be that simple.  But the cross section of the 'taper' is circular.  Most load transfer structures (i.e. gussets) are simple plates, i.e. have planar features.   This one has two mirror image quarter cones.  i.e. exact quarter circles with radias about 12" at bottom to 2" at top.
Another oddity is the color.

to me it just looks like a bent sheet, not a cone. attached another view from elon's twitter.

Which objects are you looking at?   I am looking at 2 pairs of cones on the far left of your picture, not at the contraption on the floor in the center

They're the same objects, I believe.

[_MECO]

Looks more like a temporary tie rod to me. Loose fitting, and the body-side bracket is rather rough. Plenty of unfilled bolholes too. This is definitely where the actuator will go, but this isn't it, yet.

You know, the moment I posted that I had the same thought too. It's incredibly small for a hydraulic cylinder and I don't see any ports for working fluid on the side. Still, seeing the attachment to that mounting point/clevis gives some insight into where the actuator is going to go.

I believe Elon has stated that the fins will be electro hydraulic actuated, at least for the first few MK iterations. This means no major hydraulic plumbing - and powered via the Tesla battery packs mounted to one of the header tanks.


BTW - I am expecting to see more battery packs loaded (or already loaded I suppose) into the lower tank section to actuate the bottom fins. Especially if the actuators are DC instead of AC. Those would have to be some serious gauge wires to be able to run the distance from the bow/nose to the lower fins.

Electro-hydraulic still means the end muscle is a hydraulic actuator, right? If it's a cylinder actuating the flap you'll still need at least a couple of hoses connected to the cylinder.

I'm interested to learn how the described "worm drive" which is supposed to eventually replace hydraulics in the production design is going to work. Right now, it looks like the method by which the flaps/fins are actuated is via a lever arm which protrudes inward from the hinge axis. Is this going to eventually turn into a section of gearing which will interface with the worm gear? I've drawn up a crude diagram in MS Paint to show what I mean. It seems like if this approach is taken there could be some limitation in the angular travel of the fin caused by the driven gear butting up against the fin root (or hull, it it's deeper inside the rocket.)

Maybe we're talking about the same thing but I'd think a ball-screw would work better.   Your config with the ball-screw is more like what was thinking.  Then again if he's talking electromechanical, just swap out the ball-screw for an actuator.

Here’s what I’m seeing.

The hull cutout is way longer than it needs be if the tie rod is a stand in for an actuator of any type.

ISTM that the actuator will be behind the plate and move the plate back and forth resulting in the tie rod moving the fin. Put a piece of stainless on the plate conformal to the hull and the hole is effectively closed off but not air tight.

The actuator for early Marks can be electro/hydraulic and for later Marks straight electric with ball threads. Instead of one motor/thread, size for two and install three for redundancy.

This puts the most temperature sensitive parts in a more benign environment.

Issues:

- The link hookup looks loose. Maybe it’s just a stand in for the presentation.

- the plate looks massive. We don’t know how much it’s been hogged out on the back. The motors may have an open center with ball nut, mount to the plate and work against a fixed thread. I don’t know if this has any serious advantages but it will reduce rotating mass a bit and counter the shift of CG by moving a larger mass counter to the fin.

Phil

I actually really like that idea! Not sure one way or the other about that base plate including the actuators, but looking closer at BocaChicaGal's images (specifically https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=48895.0;attach=1586474;image) it almost looks like that slot is offset such that it mostly resides on one side of the hinge. I've drawn another crappy MS Paint diagram. Maybe it looks something like this?

[martiantime]

So, this new beautiful bocachicagal's photo reveals lower entry point of the external LOX pipe.

If Stimbergi determined lower bulkhead position correctly, the LOX pipe enters into the LOX tank near the bottom. I didn't expect it. I thought it would be under the LOX tank.