A One-Piece Lunar Regolith-Bag Garage Prototype

 

Excerpt from Final Report

 

 

Gweneth A. Smithers, Mary K. Nehls, Mary A. Hovater, and Steven W. Evans

NASA/Marshall Space Flight Center, Huntsville, Alabama

 

J. Scott Miller

Qualis Corporation/Marshall Space Flight Center, Huntsville, Alabama

 

Roy M. Broughton, Jr., David Beale, and Fatma Kilinc-Balci

Auburn University, Auburn, Alabama

 

3.3 Full-Scale Prototype, and Erecting at MSFC

 

The full-scale prototype arch was sewn at Kappler Inc. in Guntersville, AL according to the design supplied (Figures 3.17 and 3.18). The final structure was constructed from a coated TwaronTM  fabric . The bag abrasion tests suggested that a tight weave was needed to contain the fine particle sized filler material, and that perhaps a coating would also be helpful. The fabric specification selected was a 70X70 (220 Dtex) plain weave supplied by Lincoln Textiles (style 3041.062.02.000). It was coated with a layer of ethylene/vinyl acetate copolymer. This copolymer is often used as a hot melt adhesive for laminating. The uncoated fabric weight was 3.6 oz/yd2 and the coating added about 3 oz/yd2.  The fabric proved to be impervious to the fine vermiculite, and strong enough for this prototype. The seams were double stitched with a black polyester sewing thread (T 70 Anafil, 16 oz., bonded, from American and Efird, Mt. Holly, NC ). The zippers were a standard urethane coated coil construction used for water repellent clothing and supplied by Kappler.  During the construction and erection of the structure, there were no seam or zipper failures. Occasionally, loose threads became entangled in the zippers, but on removal of the loose threads the zippers continued to function. The structure was erected at MSFC over several days, requiring the assistance of as many as 5 personnel at any one time. The purpose of this effort was two-fold, 1) to investigate methods of fabric construction and erection of a bag arch that may be able to serve as a standing lunar garage, and 2) to provide a standing structure at MSFC that could serve as a proof-of-concept and platform for test and observation. The prototype with filled bags was designed to be able to stand on its own without external support (i.e. be stable), which it did once the support system used for erecting was removed. A foundation support consisting of two 2-inch by 4-inch boards was left in place to guarantee support at the base and prevent the legs from spreading. The prototype included 60 pockets in a top-connected configuration, that was envisioned, once erected, to look like Figures 3.17 and 3.18, which include:

 

  • 20 pockets at the bottom measuring 6” x 2’ in cross-section.
  • 20 pockets above the bottom pockets, measuring 6” by 1.5’ in cross-section.
  • 20 pockets that form the crest of the arch, measuring 6” by 1’ in cross-section.

 

Before discussing the actual erecting at MSFC using 46 bags, prototype arch designs of a 60 bag catenary shape (e.g. Figure 3.18) were analyzed using the funicular polygon technique. Under its own weight, the arch of Figure 3.18 revealed that it could hinge (Figure 3.19).  However, taller configurations were found to be stable (Figure 3.20), and are expected to be able to support more weight.

 

Figure 3.17. Concept Drawing of Arch

Figure 3.18. Front Dimensioned View, Dimensions in Feet.

 

 

Stability analysis using the funicular polygon (Figure 3.19) shows that the structure of Figure 3.18 can possibly form hinges. The analysis was done assuming no applied forces were acting on the system except vermiculite weight. One location was identified as lying between the 15th and 16th bag starting from the left on the figure. Considering that this structure is symmetrical, the analysis indicates that hinges will form on the other side as well. Figure 3.19 below shows that hinges appear to form around the 15th and the 16th bag.

Figure 3.19. Funicular Polygon Showing Possible Hinging in Arch Design

 

As can be seen in the image above, the hinges will tend to open towards the outside of the bags, but since these bags are top-connected by a stiff fabric layer, the fabric layer will prevent any opening and therefore not allow a hinge to form. Hence this shows that this configuration may be made stable using the top-connected bag construction. Another analysis for a wider base below indicates that hinges can form near the 20th and 21st bag. This formation of hinges is shown in the Figure 3.20. These hinges will open on the inside of the bags, where there is not a connecting fabric layer.

 

Figure 3.20. Another Set of Possible Hinge Locations

 

Another configuration was analyzed to evaluate its stability. The height was increased and width decreased. The actual dimensions are shown in Figure 3.21.

Figure 3.21. Taller design, dimensions in feet.

 

On performing analysis on this structure it was seen the funicular polygon lies inside the structure, making the structure stable. 3.22 shows the funicular polygon for this configuration. Increasing the arch height usually improves stability.

Figure 3.22. Funicular polygon of 60 bag taller, stable structure.

 

 

 

Erecting the MSFC Top-Connected Bag Arch

 

Only 46 of the 60 bags were needed when building at MSFC. This size fitted within the construction space available, achieved a sufficiently large structure for presentation, while shortening the time to erect, and also demonstrating how the extra 14 bags could potentially be used to serve as supplemental supports. Figure 3.23 shows a structure, designed to be stable. Only 3 large bags were present as bottom bags. A wooden frame in Figure 3.24 was constructed that served to guide erecting toward the approximate catenary shape at 5 points, where pipes were placed on the frame for the fabric to hang. Actual pipe locations can be compared to design dimensions in the following the table. (Note that in the top-connected bag arch, the pipes provide minimal support because of the bending stiffness that this construction offers). 2x4’s attached to the bottom and sides of the frame (not visible) served as the foundation, preventing the bottom bags from slipping to the left or right.

Figure 3.23. CAD Model Template to Guide Erecting

 

Table 3.1 Construction details

POINT

X

Y

Actual Pipe Locations

A

0

0

 

1

1.45

18

 

2

2.151

23.746

 

3

2.985

29.475

 

4

3.9

35.18

 

5

5.1

40.857

 

6

6.4

46.49

 

7

7.9

52.08

 

8

9.6

57.61

 

9

11.59

63.06

 

10

13.806

68.411

Lower Level Pipes: x=14.5”, y=70”

11

16.31

73.63

 

12

19.129

78.687

 

13

22.29

83.536

 

14

25.817

88.127

2nd Level Pipes: x=25.7”, y=88”

15

29.724

92.399

 

16

33.56

95.88

 

17

38.66

99.73

 

18

43.65

102.65

 

19

48.913

105.087

 

20

54.39

106.955

 

21

60

108.287

Top  Pipe: x=60, y=108”

 

 

Figure 3.24. Air-Filled 46 Bag Structure, 5 pipes guiding bag filling.

 

After airfilling the bags, filling of bags with vermiculite proceeded from bottom bags up. Bags were filled using a Flexible Screw Conveyor System (Hapman, Kalamazoo, MI), which can be seen in Figure 3.24. The white pipe contains a helicoid screw (without a center core tube), rotated by a motor that feeds and forces vermiculite into the bags. The white pipe was inserted into the bag to within ~1 foot of the bag end, and the motor turned on to rotate the screw. As vermiculite flowed out, the tube was slowly and incrementally pulled from the bag. This operation was labor-intensive, required human assistance to distribute the vermiculite as it came out of the tube into the bag (Figure 3.25).

Figure 3.25: Bag Filling Process

Lower bags were filled and formed to a near rectangular shape (Figure 3.26), while trying to provide bag angle (notice the black zippers) as the structure grew.

 

Figure 3.26. Rectangular Packed Bags

 

Filling of the top 20 bags required a different technique.  Unfilled bags now hang down from the top fabric, and cannot be filled with a rectangular shape and maintain soil strength because of the looseness of the bag. Therefore the top 20 bags must be filled to capacity with vermiculite, which causes them to round. With the top three bags unfilled, the topmost filled bags were nearly touching, which made it difficult to fill the top three bags. The maximum amount of material that could be placed in a bag was restricted by the Helicoid Screw System, which was limited to a relatively low compaction pressures because of the stiffness and strength limitations of the relatively flexible and shaftless helicoid. Low compaction pressure contributed to the top three bags not filling to the desired pressurs and fullness. It was impossible to reach into the space and pack the bags by hand. The final erected prototype is shown in Figure 3.27 (front view) and Figure 3.28(rear view). Note that in both views the pipes have been removed from their 2x4 supports, so the structure is standing without external support. We noticed that the structure did settle ~2” once the top 3 pipe supports were removed. 

 

Figure 3.27. Front View

 

Figure 3.28. Rear View. Note the zippers.

 

Upon review of the standing structure and the process of erecting, we noted the following:

 

  1. The left side of the structure in Figure 3.27 (right side in Figure 3.28) is the “good side”. It was built and maintained a catenary shape very near design specifications, except for the topmost bags.

 

  1. The right side of the structure in Figure 3.27 (left side in Figure 3.28) is the “bad side”. Here we saw several flaws that were a result of bags that were not filled to capacity.

 

  1. Figure 3.29a show that several bags on the bad side with a flattened profile that lost the catenary-shape curvature. Bags slipped downward despite our best attempts to erect the structure with a catenary shape; bags slippage was visible and occurred over a several second interval.  Slippage is attributed to a shortage of vermiculite due to incomplete packing, which did not occur on the good side; the vermiculite grains may slide (shear) with respect to each other inside the bags.  This situation is correctable (but difficult given that zippers are only on one end) by hand loading more vermiculite through an open zipper and forcing material into the bag with a plunger. A simpler fix would have been to have used an auger-type system which deposits vermiculite under higher pressure than the helicoid.

 

Figure 3.29. a. Upper left photograph shows bags in a straight line, deviating from a caternary shape. b. Lower left photograph show a bag that is tightly packed and bulging. c. The upper right photograph show tightly packed 1’ bags on the good side; it is difficult to insert a finger between the bags.  d. The lower right figure shows loosely packed bags on the bad side; the finger is easily inserted between the bags. 

 

4.  Figure 3.29b shows a tightly filled bag; it bulges and exhibits a hardness which can be felt by applying finger pressure. Tightly filled bags are necessary to create bags with sufficient vermiculite strength. Part of the internal bag pressure is a result of loading from the bags above.

 

5. Figure 3.29c shows how another characteristic of a well-built structure with tight bags. Here it is difficult to insert a finger between the bags, implying that the bags are tightly packed with respect to each other. Compressive and shearing loads are transmitted without failure across the fabric boundary, from bag-to-bag.

 

6. Figure 3.29d shows a characteristic of inadequately packed bags. Here it is easy to insert a finger between the bags – this may imply the beginnings of hinge formation. The bags themselves are loosely packed - by applying finger pressure the bags easily indented. The dark patch in the figure represents glue that was placed in-between the bags as an experiment. The glue did not appear to affect the structure. If the glue had affected the structure, the glued fabric would have been in tension, and this was not the case.

 

In summary, the top-connected bag structure is stable if erected correctly - by filling the bags with sufficient pressure and designing using masonry arch structural design principles. Several recommendations and comments for future work are as follows:

 

  • The auger system used in our work was unable to fill the bags to sufficient pressure. An auger with a central shaft would have provided higher pressure and a more stable structure. It would have been convenient to have had a method for easily “topping off” bags with vermiculite once the auger was removed and filling was thought to have been complete. The top 3 bags were not filled to our satisfaction.
  • We can envision that other methods of erecting the structure are possible, and simpler and more appropriate for erecting on the moon. Sensors could be used to measure bag pressure and deviations of the shape from the desired while erecting, so adjustments can be made. Shaping of bags during and after filling was a challenging manual task.
  • Analysis can be greatly improved, by incorporating both fabric and regolith materials and their characteristics in the analysis, perhaps using nonlinear finite element analysis. Improved analytical methods could be used to optimize the design. Regolith and vermiculite soil strength parameters (determined by standard soil mechanics tests or the Lunar Sourcebook) need to be incorporated into the modeling, along with fabric strength parameters.