Monthly Archives: February 2023

This past two weeks has seen a number of disparate tasks completed in parallel.

The two decades old shipping container serving as our Crew Quarters was designed to be air tight but not to hold pressure over time. This is compounded by skirmishes with forklifts that resulted in a few patches, bowed panels, and disconnected interfaces. As we have this past six months repaired the major leaks, smaller holes have been exposed. Our most recent pressure test discovered a leak at the overlapping seam between the stainless steel frame and aluminum ceiling roll at the back of the 40′ shipping container. Kai removed the failed rivet and drew open the enveloped just wide enough to inject 795 silicone. A new rivet drew the the interface tight again. Bindhu then scraped and cleaned the entire interface that spans the width of the container, followed by an application of aluminum tape with assistance from John Z. The entire perimeter of the wall-to-ceiling interface is now complete.

Kai and Sean worked to prepare for the installation of the electrical sockets in the workshop (20 foot container) with fabricated aluminum plates to give strength to the in-house designed, modular wall mount system. Subsequently, three 20 amp electrical circuits were installed. The original, exterior steel door to the Test Module was removed last fall, sanded, primed, and recently top-coated with Rust-Oleum white enamel paint, the same that now protects the entire exterior of the refurbished shipping containers and airlock. With new, steel bearing hinges, the door was replaced and is incredibly smooth in its operation, ready for three more decades of service.

Following several applications of PB Blaster over the prior week, Kai and John Z attempted (in vain) to remove an old “J” pipe fitting on the exterior of the B2 airlock installed at SAM last year. The 4 foot, 30 lbs wrench was apparently not large enough, and in the end the fitting was cut free, the remnant filled with foam, to be capped and sealed soon.

Arizona State University undergraduate student and new volunteer at SAM Tasha Coelho applied diligent attention to the final effort in restoration of the original Test Module lung. She scraped free all loose paint and then hand-sanded the underbelly of the pan in preparation for a coat of Rust-Oleum primer. Luna then applied two coats of a beautiful, dark blue-grey semi-gloss enamel.

Sean and our new volunteer Matthias Beach tackled the installation of the new variable frequency drive (VFD) blower to inflate the lung and pressurize SAM. Matthias is a Former Air Force Computer and Switching Systems Specialist and citizen telecommunications field service specialist who brings skills in construction, electrical wiring, and equipment operation to this endeavor at a critical time. The VFD motor is 1 HP (as compared to the former blower of 1/3 HP) with a significantly larger blower which will result in our ability to inflate SAM in far less time than the former 15 minutes, and with the variable speed control, maintain a constant, minimal influx of air for those teams preferring to run SAM in either of the two pass-through modes.

In the midst of our third snow storm this winter, Matthias, Sean, Tasha, and Luna completed the wiring of the Test Module that was begun last year, with eight 20 amp circuits ready to handle the most arduous of hydroponic racks.

Some would say the bathroom is the single most important room in a Mars habitat. We agree. And as we work, adolescent bobcats watch our every move. Not very Mars-like, but very much a part of the southern Arizona experience.

In building SAM there are three primary objectives with each material chosen:

1. Will it last 15-30 years?
2. Does it provide the required strength and seal?
3. Does it off-gas once cured?

The SAM team has gained a deep understanding of working with metal, from cutting to cleaning, welding to grinding, primer, and paint, and with foam insulation panels and plastics too. The fiberglass reinforced panels (FRP) were installed in the 20 foot shipping container (SAM Workshop) over preformed Insofast panels. Given that the walls of the shipping containers cannot be penetrated for the potential of forming a leak, nor can the 40′ be welded due to the potential of melting the insulation between the stainless (interior) and aluminum (exterior) walls, adhesives must be employed. With a selection of dozens of brands and types, each of which carries its own marketing and promises, the SAM team has learned (sometimes the hard way) what works and what does not.

Silcone-base adhesives will adhere to most materials, but the final product is not a structure bond, the kind that can carry a load. While the Dow Corning 795 is foundational to the success of Biosphere 2, the Test Module, and SAM, it is not applied where load-bearing elements come to play (e.g. the shipping container wall, aluminum-based foam panels, and furring strips).

Loktite PL3 is the low-VOC adhesive recommended by Insofast and used in their instructional videos. While the final, cured product is stable and strong, it was learned that this product loses integrity when pressed too thin between the adhere layers. When pressed to a few thousandths of an inch, as is desired with wood glue, the product fails and the two layers can be separated by hand. But with the toothed ribs of the Insofast panels, the PL3 bead remains 1/8-1/4″ and provides substantial integrity, as marketed. However, this presents a challenge if the surface of bonded layer must remain parallel to the underlayment or even across multiple panels.

Sika Construction Adhesive is not found in the adhesives section of the local vendor, rather with concrete block and building materials. It is much thicker than PL3, gray in color, and does not appear to harden completely, rather it remains relatively flexible yet holds to a greater diversity of surfaces, including FRP. It is very difficult to work with, but holds to polystyrene where PL3 specifically states it cannot adhere to this medium.

Roberts 7200 Base Bond is a very different kind of product from the PL3 or Sika adhesives. It is much easier to work with, spreads with a toothed trowel and covers like oat paste. It is “zero” VOC (1g per liter according to the package) and has no odor when being applied. It provides quite a bit of working time in low temperatures, and can be cleaned up with water before it begins to harden. However, it must have at least one porous surface in order to form a bond, with the other surface designed for vinyl baseboard trim. The instructions make it clear that even a coat of paint on the wall must be removed in order for the paste to adhere to the wall board.

Tightbond III is the standard among woodworkers, the strongest of this family of wood glue products, and a staple in any wood shop. As the amount of wood in SAM is limited, it was used only to build the reinforced door frame and overhead beam that gives tremendous strength and stability to the nearly free-standing SAM bathroom.

Nashua Aluminum HVAC tape is what Mark Watney should have brought to Mars. It is one of the most versatile, capable products on the market, and the modern replacement for duct tape. While 795 and rivets have been used to seal more than 200 holes in the refurbished 40′ container (SAM Crew Quarters), it is aluminum tape that has properly secured the entire interface between the stainless steel walls and aluminum roof. We found it to be highly adhesive, malleable (to a limited degree), water resistant, and air tight.

This following test matrix was developed by SAM team member Luna Powell. All samples were tested at 24 hours, 3 days, and 5 days. The following summary is at the close of 5 days.

Adhesive	Wood	Backer Board	Aluminum
PL3	Cured to the point of not moving	Strong, almost impossible to peel up	Did not even adhere (at all)
Sika	Remains soft to the touch after ~5 days curing	Remains malleable, not impossible to peel up	Peeled up easily
Base Bond	Won’t shift side to side, but easy to peel up	Readily peels up; did not stick to backer board	Peeled up easily; wet in the middle

With bonded interfaces between dissimilar metals, metal to foam, foam to wood, wood to wood, and wood to polystyrene, the types of applied adhesives varies. This past month has seen a good bit of experimentation, some failure, and success with what is hoped to be a multi-decade solution. While there are certainly many more to explore, from off-the-shelf two-part epoxies to marine grade epoxies; from UV activated cements to fire resistant silicones, SAM has been built principally from what is immediately available from the consumer market.