Pneumatic Lunar Drill for Emplacement of Lunar Corner Cube Reflector

Development and Testing of the Pneumatic Lunar Drill for the Emplacement of the Corner Cube Reflector on the Moon

K. Zacny1*, D. Currie2, G. Paulsen1, T. Szwarc3, P. Chu4

1 *Kris Zacny, Honeybee Robotics, 398 W Washington Blvd, Suite 200, Pasadena, CA 91103, (p) 510-207-4555, zacny@honeybeerobotics.com2 University of Maryland, College Park, MD 207423 Stanford University, Department of Aeronautics and Astronautics, Stanford, CA, 943054 Honeybee Robotics, 1110 NASA Parkway, Suite 440, Houston, TX 77058

1.0 AbstractLunar Laser Ranging provides a highly accurate measurement of the distance between ground stations on Earth and reflectors on the surface of the Moon. Since retroreflectors were initially placed during the Apollo missions, the ground stations improved the ranging accuracy by a factor of 200 and now the Apollo-era arrays on the Moon pose a significant limitation to the ranging accuracy. The new Lunar Laser Ranging Retroreflector (i.e., the Lunar Laser Ranging Retroreflector for the 21st Century or LLRRA-21) would provide extensive new information on the lunar interior, General Relativity, and cosmology.

During the day/night lunar cycle, when the thermal variation of the surface is approximately 300 °C, the regolith will rise and fall by almost 500 µm. Yet, it is estimated that the thermal variation 0.5 m to- 1 m below the surface is less than much 1 °C. Thus for the lunar emplacement to support 10s of microns ranging accuracy to achieve 10 µm ranging performance, the Reflectors must be anchored to that thermally stable mass at 0.5 m or greater depth.

In this paper, we present a novel method of deploying LLRRA-21 with a Corner Cube Reflector (CCR) on the Moon. The emplacement approach uses a gas-powered drill consisting of a >50 cm long, slim, hollow rod with a perforated anchor-cone at its lower end and the CCR mounted to the top. Gas supplied from a small tank is directed into and down the rod and out through the cone, lofting the soil out of the hole and allowing the rod to sink under it’s own weight to a depth of 0.5 m. depth.

To determine the system performance, we conducted several of tests in compacted JSC-1a lunar soil simulant and inside a vacuum chamber. In several tests, the rod successfully sunk under its own weight of 16 N to a depth of 50 cm in 4-6 minutes.

The pneumatic system is the game-changer for subsurface access. The extremely low mass and volume required to reach 50 cm, along with very simple penetration method allow the CCR to remain in a variety of payload architectures.

Keywords or phrases. Lunar Laser Ranging, Corner Cube Reflector, Drilling, Gas Drill, Pneumatic Drill, Moon, llrra-21

2.0 Introduction While the Apollo retroreflector arrays are still in operation and continue to produce new science results, the combination of the lunar librations and the design of the arrays currently limit the range accuracy obtained for each single photo-electron return to ~20 mm (Currie et al., 2011). A next generation lunar retroreflector such as the Lunar Laser Ranging Retroreflector for the 21st Century (LLRRA-21) holds promise for great improvements in the existing values on the interior properties and the expectation of addressing new discoveries (e.g., the solid inner core). The magnitude of these improvements for this next generation science will depend critically on the method of deployment of the LLRRA-21.

During the day/night lunar cycle (equatorial surface temperatures are 390 K by day and 100 K by night), the regolith will rise and fall by almost 500 µm. Yet, it is estimated that the thermal variation 50 cm below the surface is much less than 1 °C throughout the month (Nagihara et al., 2010). Thus for the lunar emplacement to support 10s of microns ranging accuracy to achieve 10 µm ranging performance, the Corner Cube Reflectors (CCR) must be anchored to that thermally stable mass at 50 cm or greater depth.

This paper describes a method of deploying and anchoring the next generation Corner Cube Reflector system (LLRRA-21) to ~50 cm depth. It presents test results from gas-assisted drilling inside a 3.5 meter tall vacuum chamber to approximately 50 cm depth in compacted lunar soil simulant JSC-1a.

2.1 Challenges Posed by the Lunar Environment To design the deployment system, we first need to understand the lunar environment. The Moon is different from Earth in three important aspects with respect to the LLRRA-21CCR deployment:

• Low Gravity: The gravity is 1/6th g. This has a direct effect on how much drill force can be applied.

• Hard Vacuum: Surfaces will have no water films or oxides. Soil strength increases (Johnson et al., 1973)

• Lunar regolith: Lunar regolith is dry, granular, and interlocking – it has apparent cohesion (Heiken et al., 1991). To excavate it, we need to either remove soil to the surface or to pulverize it in the hole and press it into the borehole (Carrier, 2005). This necessitates a drill with an auger, a pile driver, or a pneumatic system (for blowing the soil out of the way beneath the penetrating rod).

2.2 Methods of Penetrating Lunar RegolithThere are various methods of getting to the subsurface. Trades given in Table 1 and follow-on sections describe each in detail. Note that Table 2 also gives Go/NoGo for systems under consideration. The items in bold indicate parameters which disqualified the system from a ‘GO’ classification

Lunar regolith below the first 10-20 cm is highly compacted and highly cohesive. The relative density reaches 100%, i.e. regolith can not be compacted any more. Thus, the only way a probe can move forward is if it can remove the regolith from underneath or if it has enough energy to crush/pulverize the regolith, re-compact it and push it to the side (exceed bearing pressure of the overburden regolith). There is no other way.

No method exists that has been shown to successfully penetrate highly compacted lunar regolith to a depth of 50 cm - 1 meter within the mass, volume and power constraints of the small lunar landers. However, some approaches are more likely than others to reach 1m depth.

2

2.2.1 Apollo Lunar Surface DrillThe Apollo astronaut deployed drill used 1 inch augers with a bit at the end. It was driven by a 450 Watt drill and penetrated up to 3m depth. Because of mass, volume and power constraints this approach will not be possible for small robotic missions.

2.2.2 Percussive Approach Apollo 15 and 16 used a so called Soil Recording Penetrometer (SRP), to determine soil strength. The SRP had a rod with 12.8 or 20.3 mm cones attached to the end and was pushed to a depth of 30 cm-60 cm (deeper in softer soils). The forces required to push that deep were in the range of 44 lbs (111 N), which translate to 264 lbs (1200 N) on Earth (a mass of 44 lbs on the Moon provides as much force as a mass of 264 lbs on Earth, because of Earths 6x larger gravity). Pushing a rod to a depth of 1 m from a lander weighing even 1 ton will be challenging.

The alternative approach is to use a percussive system (high-frequency and low-energy impacts). We performed tests in which a rod was hammered into columns of compacted (1.9 g/cc) lunar regolith simulant (JSC-1A) to a depth of ~1 m. This compaction was consistent with densities of lunar soil on the Moon. All penetrometer designs reached ~1m depth, though with different speeds (the fastest reached the bottom in a few tens of seconds, the slowest took 3 min). However, this method is more like drilling (except slightly lighter) and in turn requires a powerful vacuum rated hammering system. For this reason it is not suitable for light-weight landers.

2.2.3 Pneumatic Penetration The pneumatic penetration approach relies on excavating the soil by displacing it (moving it out of the hole with gas). The gas (at pressures as low as a few psi absolute or psia) is injected through a cone shaped nozzle into the soil (Figure 1). The injected gas, as it expands and escapes, exchanges momentum with the soil particles and carries them out of the hole.

The gas could be provided from the lander propulsion system, which uses compressed helium to pressurize tanks. This helium is vented to vacuum upon touchdown. But in our case, it could be used to penetrate to 1m depth.

3.0 Gas Assisted Drilling on the Moon

3.1 Why Gas Drilling? Lunar regolith below the uppermost 10-20 cm layer is highly compacted and cohesive (Carrier et al., 1991). The relative density (Dr) at such depths reaches 100%; one cannot compact the regolith further without altering the size and the shape of the individual soil grains.

There are only two ways for a probe to penetrate subsurface in such conditions. The first method is to loosen up the compacted soil and remove soil particles from the hole and in turn make space for the probe to move deeper. Removing the soil can be done by rotating auger (as in drilling) or blowing the soil out of the hole using compressed gas though in vacuum; almost any gas pressure would do the job (Mumm et al., 2009, Zacny et al., 2011). The second method is for the probe to crush and pulverize individual soil grains, recompact the pulverized soil, and push it into the borehole (Mumm et al., 2009). Crushing and compacting the soil can be achieved by deploying a percussive (i.e. hammering) probe that can deliver sufficient impact energies to crush the soil and re-compact it while pushing it into the borehole walls (Zacny et al., 2010).

The LLRRA-21Corner Cube Reflector (CCR) anchor system described here relies on excavating the soil by loosening the soil and removing it from the hole (Zacny et al., 2011). The soil removal is

3

accomplished by gas (with pressures as low as a few kPa absolute) jetting out through a cone-shaped nozzle. The injected gas, as it expands and escapes, exchanges momentum with the soil particles and carries them out of the hole. Deflection plates on the surface can be used to divert this dusty gas away from the lander or a rover (Figure 1).

The deployment system can be designed so that if the anchor runs into a rock, the deployment arm can lift it out of the hole and move it to another spot. However, probability of the anchor running into a rock with a diameter twice or greater than that of the cone is <0.5% (Carrier, 2005). The Apollo astronauts drilled a total of eight ~2-3m holes (~20 m total) and did not hit a rock.

Unlike conventional drills, the gas-assisted drilling does not require electric motors and batteries to penetrate into the subsurface. Hence it is inherently lighter, more compact, and easier to deploy. In its simplest deployment, a gas tank can just be opened once and never closed until emptied of gas.

3.2 Applications of Gas to Lunar Excavation - Previous WorkThe pneumatic excavation approach was proposed before by others, but never tested in a vacuum chamber. For example, Sullivan et al. (1993) evaluated the feasibility of pneumatic transfer for the movement of regolith (lunar soil) at lunar gravity conditions on NASA's KC-135 reduced gravity aircraft. They found that the choking velocity (in the vertical transfer) and the saltation velocity (in the horizontal transfer) at lunar gravity were reduced to 1/2-1/3 of the velocity required at 1 g (choking and saltation velocities are minimum gas velocities that keep particles aloft). This means that the excavation efficiencies are expected to double or triple under the lunar gravity environment.

Schaefer et al., (1994) used a slim tube to penetrate up to 2 m into sand. They varied gas pressures (1379 kPa and 1724 kPa), type of gas (N2 and CO2), as well as pulse durations (30 and 40 pulses per minute). Depending on the pressure and pulse duration, the depth of 2 m was reached within 2-8 minutes. The mass ratio of gas used to sand removed was of the order of 1:5. Komle et al., (2008) presented a design for a suction drill and performed hydrodynamic analysis of a dusty flow.

Pneumatic mining was also proposed by Zacny et al. (2004). The proposed idea was to inject gas into soil to break apart and entrain the lunar soil particles, and then loft them out of the hole in the escaping gas stream. This dusty gas was then to be directed to a cyclone separator to separate solids from gas. Several tests were conducted inside a vacuum chamber at 600 Pa pressure (5 torr) to establish feasibility of using this method of mining. The most interesting finding was that gas pressures required to loft soil particles were much lower than gas pressures used by Sullivan et al., (1993) and Schaefer et al., (1994). In particular, 1 liter of gas (Air) at 3.3 kPa absolute was sufficient to loft 10 g of soil at velocities in excess of 5 m/s. The mass of air used was 37 mg. Since the pressure inside the chamber was on the order of 600 Pa, the pressure ratio was in the range of 1:5. This pressure ratio was similar to that used in tests conducted by Schaefer et al., (1994), however, the mass lifting efficiency was much higher, of the order of 1:200; that is 1 g of gas at 3.3 kPa absolute was sufficient to lift >200 g of soil. The mass lifting efficiencies obtained by Schaefer et al., (1994), during atmospheric tests, were in the range of 1:5.

3.3 Applications of Gas to Lunar Excavation - New Experimental WorkMore detailed tests were also conducted by Zacny et al., (2008, 2009, 2010). During these tests, a continuous stream and a short puff of gas were used to loft soil particles out of the soil bin. All tests were conducted in 1-5 torr vacuum. Results indicated that 1 g of gas (air) at 101 kPa absolute (i.e. at atmospheric pressure) was sufficient to lift 6000 g of soil particles at high velocity (>10 m/s). Tests conducted at various chamber pressures suggested that gas lofting efficiencies will increase if the pressure were to be further reduced to hard vacuum. In particular, it is expected that gas lofting efficiencies in lunar vacuum will exceed 1: 10 000 (i.e. 1 g of gas at ~100 kPa absolute would be sufficient to loft over

4

10,000 g of soil). Tests conducted at 1/6th g and 3 torr vacuum (inside a Zero-G Corp aircraft) confirmed 1:6000 lifting efficiencies at 3 torr vacuum (Zacny et al., 2010). It was also determined that gravity effect was relatively small when short pulses of gas were used.

Several tests were also performed to determine feasibility of using gas-spear technique for probing below the lunar surface (Figure 1). The experimental set up included a rod with a 2.5 cm diameter, 30° cone having gas injection holes. The rod was connected to a Z-stage via a load cell. The Z-stage allowed lowering the rod into the soil bin at predetermined velocity, while the load cell was used to measure the forces required to push the rod into the soil. All tests were performed under atmospheric conditions in a 10 cm diameter and 100 cm deep bin filled with JSC-1a lunar soil simulant vibratory compacted to 1.9 g/cc. Please refer to Gouache et al. (2010) for comparison of three regolith simulant preparation methods: rain, pour, and vibrate. Alshibli and Hasan (2009) measured the minimum and the maximum index density of JSC-1A by following the Method A of ASTM-D4253 standard and found them to be 1.556 g/cc and 2.016 g/cc, respectively. Hence the density of 1.9 g/cc corresponds to the Relative Density (Dr) of 75%.

During the tests, the cone was initially lowered into a highly compacted JSC-1a soil as shown in Figure 2. A set force of 270 N (60 lbs) was reached within the first few centimeters of penetration. Once gas was injected (via injector holes in the cone), soil was lofted out of the hole and vertical thrust dropped to zero. These tests therefore demonstrated that while the gas jets were removing the soil, practically no push force was required to advance the rod into the soil.

Table 3 shows data from four penetrometer experiments. In the first test, no gas was used and the cone breached a 450 N force limit after penetrating only 5 cm into compacted JSC-1a. In the subsequent tests, the type of gas and pressure was varied. In these tests, injecting gas into the soil at moderate pressures was sufficient to reduce the penetrator force to zero and reach the target depth of 60 cm in just 2 minutes. There was no measurable difference in performance between the air at 138 kPa and 35 kPa nor between the Air at 35 kPa and Helium at 35 kPa. This is probably due to the fact that no lower bound pressure and flow rate were reached during these tests (one would have noticed lower bound if the penetration force had increased). The best result was achieved during the Test 4; only 11 g of He gas was used to reach 60 cm depth in just 2 minutes.

We also performed preliminary tests in a lunar analog site on the slopes of Mauna Kea volcano on the Big Island of Hawaii (Figure 3). The experiment included a hollow rod with a cone at the bottom and the CCR on top. A pneumatic line was connected to the side of the rod. We used compressed air from a compressor but in the flight system, pressurant helium from a lander propulsion system will be used instead. Prior to testing, we measured soil strength using geotechnical tool – a rod with a load cell on top and the same diameter cone as in the CCR at the bottom (Figure 3). We determined it takes >130 lbs to manually push the cone into the soil to a 50 cm depth.

However, when the pneumatic line was opened the rod almost fell into the subsurface with little push (the entire test took maybe a second). We estimate the push force must have been <3 kg or 30 N, but it was difficult to measure it accurately since the rod sunk in very fast.

The experiment successfully demonstrated that when gas is injected into the soil through the cone, it reduces deployment forces by more than an order of magnitude.

3.4 Selection and Preparation of the Analog Lunar SoilSoil preparation for vacuum chamber testing is an extremely difficult and time consuming process. The soil has to be prepared to reflect the properties of the lunar regolith. If the soil properties are much

5

different than that of the lunar regolith, the test data will be compromised and in turn, all the equipment designed and built based on the acquired test data will not work properly once deployed on the Moon.

Deployment of the Heat Flow Probe (HFP) experiment on the Apollo 15 mission is a good example of a failed experiment due to tests being done on Earth in soil that was not a good lunar soil simulant. In particular, the hollow drill stems for the HFP experiment were tested in soil that was at a looser state than the actual lunar soil. The drill worked fine on Earth, but when deployed on the Moon, the drill stems got stuck at a depth of ~1.4 m and never reached the target depth of 2.4 m (Heiken et al., 1991). The depth of 1.4 m unfortunately was not sufficient to acquire high fidelity thermal data. After the Apollo 15 experiment, the Apollo 16 and 17 drill stems for the HFP experiment were redesigned to work in a more compacted soil and performed flawlessly when deployed on the Moon.

Over the past decades several US agencies and countries developed various lunar soil simulants (McLemore, 2010). These include JSC-1a first developed by NASA Johnson Space Center and now manufactured by Orbitec, NU-LHT-2M produced by NASA Marshall Space Flight Center (MSFC) and United States Geophysical Survey (USGS), GRC-1 and GRC-3 from NASA Glenn Research Center (GRC), FJS-1 from Japan, and OB-1 from Canada.

The main attributes that the soil should have to correctly simulate lunar soil properties for geotechnical purposes (excavation, drilling etc.) are particle size distribution and particle shape. The soil then has to be compacted to reach its maximum density (approximately 2 g/cc). Most simulants have the same particle size distribution but their particle shapes may differ. Some simulants, for example, have agglutinates, which are highly abrasive particles made of fused glass and rock and naturally occurring only on the Moon. Though, simulants with agglutinates would be the most appropriate for testing, the final choice of the soil simulant is actually driven by its availability and cost, with the latter being anywhere between $10,000 and $60,000 per ton.

It is relatively easy to prepare soil for atmospheric testing. Soil can be first placed in a bin and then either compacted by vibrating the bin or using plate compactors (also known as soil compactors) which are used everyday in the construction industry. Either system can compact the soil to the desired density; however the choice of the method will eventually be made based on the ease of applying it to the actual experiment. Vibrating large bins can be ineffective, while a plate compactor may not fit into a small soil bin.

Preparing soil in vacuum is very difficult. During the vacuum pump down process, atmospheric gas trapped between individual soil grains initially escapes and as it does so, loosens up the soil. Once the pressure close to the triple point of water is reached (611 Pa), trapped moisture vigorously boils off creating large volumes of high pressure vapor. These vapors act as small explosives since the end result is the soil being blown out of the soil bin.

In order to prevent these vapor explosions the soil has to be in a very loose state (i.e. have high permeability) during the pump down process to allow the vapors to escape between the soil grains. In addition, the soil should be heated to above 100° C to drive off volatiles. Therefore, soil can only be compacted once the target vacuum has been reached. This requires utilization of some kind of densification mechanisms (e.g. vibrator) and attaching it to the soil chamber in such a way as to be able to turn it on or off from outside of the chamber.

It should be noted that soils with different particle size distribution will behave differently during the pump down process. This is mainly attributed to the fact that these soils will have different permeability. Figure 4, for example, shows three different powders: relatively coarse Basalt Cuttings, relatively fine Santa Barbara Sandstone Cuttings, and mid-sized Ohio Sandstone Cuttings. When these three rock

6

powders were places inside individual test tubes and subjected to decreasing pressures inside of a vacuum chamber, they all behaved differently. Basalt cuttings were immediately lofted out of the tube by escaping gas and moisture, Ohio Sandstone Cuttings formed several plugs which were lifted up the tube by expanding gas and vapors, and gas escaping through the Santa Barbara Sandstone cuttings would make it “bubble”.

4.0 Testing of the Gas Assisted Drill for the Deployment of CCR

4.1 Experimental Set Up Figure 6 shows the 3.5 meter tall, 1 meter deep and 1 meter wide vacuum chamber with a soil bin inside it. The chamber door has large windows for easy viewing of the experiment. The chamber has been rated for 10-2 torr vacuum, however with its current pump the lowest pressure that can be achieved is approximately 1 torr.

The acrylic soil bin stands 1.25 m tall and is 53 cm in diameter, with JSC-1a filled 85 cm deep. The bin has a lid on top to reduce soil spillage during pump down and during the actual tests. In particular, during the pump down period, the trapped gas and moisture escape between the soil particles and loft some of the soil particles up. The volume of escaping gas and water vapor can be sufficiently high to create micro-explosions. These micro-explosions force the soil upwards and out of the soil bin. If the soil is very wet and pumping speed not controlled very well, a substantial volume of soil can be lost. Normally, the soil is placed in a bin and fluffed up to increase permeability and make it easier for trapped gas and moisture to escape. Once the vacuum is achieved, the soil is vibratory compacted using a vibrator mounted at the bottom of the soil bin.

In each test, the CCR was mounted on top of the CCR hollow rod (an anchor) and placed on the surface of the compacted JSC-1a lunar soil simulant. The CCR in Figure 6 is shown prior to deployment to a 50 cm depth.

Figure 7, Figure 8, Figure 10, Figure 11 show details of the retro reflector (i.e. Corner Cube Reflector or CCR), the CCR housing, CCR deployment rod, and the cone, respectively. The CCR rod and the cone form the CCR anchor.

The CCR consists of a solid prism composed of fused silica. The CCR housing was made of aluminum. Currie et al., (2011) describe in detail the design and the thermal/vacuum tests of the housing and the CCR.

In this particular set up, the diameter of the rod was 13 mm while the diameter of the cone was 14.5 mm allowing approximately 0.75 mm of clearance on each side of the rod for the soil particles to move up the hole.

The rod was made of aluminum; however, in the next iteration Invar will be used. Invar, known generically as 64FeNi in the US, is a nickel steel alloy having very low coefficient of thermal expansion. Common grades of Invar have a coefficient of thermal expansion of about 1.2 × 10-6 K-1 (1.2 ppm/°C). Having a material with a low coefficient of thermal expansion is of paramount importance since the CCR should maintain its vertical position in order to avoid errors associated with measurement of the distance between the earth and the Moon.

Figure 9 shows details of the CCR deployment set up inside the chamber prior to the test. The CCR system consists of a housing with the CCR inside it, and the CCR anchor: a hollow rod with a cone at

7

its end. Gas enters through a feed line near the top of the rod, and exits through jets in the cone at the bottom of the rod.

The tests were run without placing the CCR in the housing in order to reduce the weight of the deployment system. The mass of the housing and the anchor was 1.622 kg. In the first test additional mass of 2 kg was placed inside the housing to increase the mass of the system and in turn increase the excavation force. This test is indicated in Table 4, column 2 (“Additional mass”).

To prepare the column of JSC-1a soil simulant for testing, the soil was placed in a large bin shown in Figure 9 and the chamber pressure was reduced to approximately 1 Torr. A heater placed 12 cm from the bottom of the bin was used to heat the soil and drive off moisture and volatiles. The heater was left on during and between all tests to aid in removing any remaining volatiles. Soil at the bottom of the bin and in close contact with the heater reached a temperature of 82 ºC while the temperature approximately 20 cm above the heater was measured to be 25 ºC. Because of the soil’s low thermal conductivity it was very difficult to heat the entire column of the soil.

Once the vacuum was established a vibratory actuator mounted to the bottom of the bin was used to compact the soil in vacuum. The density of the compacted soil was approximately 1.9 g/cc, which corresponds to the Relative Density, Dr of 75%. Between the tests, this vibrator was also used to level the surface of the soil and recompact the soil disturbed by tests.

During each test, a device called a dust shield was placed on the surface of the soil (see Figure 9). The shield was made from an inverted bowl with large sections cut out along its sides. During operation of the deployment system, the dust shield routed dust expelled from the borehole radially outwards to prevent it from traveling upward toward the CCR housing.

The tests began by allowing atmospheric air (or in Test 4, 768.0 kPa absolute compressed air) to flow through the feed line, through the rod, and out of the four jets in the cone. The flow rate could either be held constant or varied to achieve a desired rate of penetration. Mass flow rate was measured with an Alicat Scientific MC-10SLPM flow controller. Tests were stopped either when penetration stopped despite a large flow rate or at 50 cm depth when the rod descended to the point that the feed line prevented further penetration.

4.2 Experimental ResultsTable 4 shows the sequence of tests run in the vacuum chamber. In all tests, the rod was placed on the top of the soil – no external means were used to push the rod into the soil. The penetration was achieved by blowing the soil underneath the cone while the vertical force was provided by the weight of the system only. The mass of the CCR housing and the anchor was 1.622 kg. Only in the first test additional mass of 2 kg was added to the CCR to increase the system mass. This was done to determine if higher vertical force would aid in excavation rate.

During the six tests several variables were tested to determine their effect on penetration depth, penetration rate, and mass of gas used. The test was considered a success if a depth of 50 cm was reached. This was accomplished in Tests 1, 2A, and 3. In Tests 2 and 4, it was observed that the deflection shield impeded the progress by rubbing against the side of the rod.

In all tests but Test 3, the gas flow rate was manually increased to maintain the penetration rate and to conserve the mass of gas. This is because initially lower flow rate is required to remove the soil than later on, when the cone is at greater depths. However, the purpose of Test 3 was to determine how much gas will be required if a gas cylinder was opened once and never closed. Opening the cylinder once rather than slowly increasing the gas flow makes the deployment operation easier. Comparing results from Test

8

2A and Test 3 (in Test 2A the gas flow was slowly increased), it can be seen that in Test 2A the anchor reached the target depth of 50 cm in 6 minutes rather than in 4 minutes in Test 3, however, in Test 2A only 10 g of gas was used, while in Test 3 almost 18 g of gas was required.

The difference between Test 1 and Test 2A is that in Test 1 additional mass of 2 kg was placed inside the CCR housing bringing the total weight of the system from 16 N to 36 N. In addition, a cone diameter in Test 1 was 21 mm while in Test 2A it was reduced to 14.5 mm. The results indicate that the anchor reached the required depth of 50 cm in 4 minutes with 9 g of gas in Test 1, and in 6 minutes with 10 g of gas in Test 2A. Thus, it can be concluded that adding additional mass to the system (Test 1) has a similar effect as decreasing cone diameter (Test 2). Since it costs more to place larger load on the Moon, it is preferable to decrease the size of the cone to minimum.

During the first few tests, the shield had to be modified several times to allow uninterrupted diversion ofdusty gas. It was determined that the ideal shield would look similar to a Frisbee.

Visual observation of the penetration progress showed that penetration does not occur continuously but rather in discrete steps. This was probably caused by the fact that gas pressure was insufficient to continuously keep lofting the soil out of the hole. Only once the gas pressure built up sufficiently high at the cone, the soil was explosively lofted up and out of the hole allowing the cone to penetrate deeper into the soil.

This sequence of tests confirmed the ability of gas drilling to insert a CCR system into lunar soil. In addition, the tests showed that penetration efficiency is mostly insensitive to gas pressure (although this may be due to the gas jets choking the flow in this particular test setup), and only slightly sensitive to system mass. However, further testing will be necessary to test new gasses such as helium and to optimize the design of hardware components such as the cone diameter, geometry and jet-holes sizes and location. Test 3 is the configuration that should become the baseline for any future testing, because simply opening a tank valve and allowing it to vent is the simplest method to implement on a spacecraft mission.

In Test 5, a copper mesh instead of a pointed cone was used. It was believed that ejecting the gas from larger surface area would enhance the penetration rate. However, the bottom of the mesh got quickly clogged with soil particles and the penetration stopped. The gas was therefore not removing the soil underneath the penetrator tip, but instead blowing out to the side along the path of least resistance.

The important question about the pneumatic system is whether the ‘airborne’ soil would cover the lander and its solar panels, for example. Our observations as well as viewing of Apollo Lunar Module descent and ascent movies showed that lunar dust flows in straight paths (unless deflected) and settles as soon as the gas is turned off. This is because there is no air on the moon, and hence no gas molecules to keep the dust suspended or deviate the flow path. Our observation of the shield effectiveness showed that soil particles flow in straight lines and often towards the ground (because of lack of air).

Figure 13 shows particle settling speed for different particle sizes at Mars (5 torr) and Earth (760 torr) pressure (Fuerstenau, 2006). It can be seen that particles fall down orders of magnitude faster under Mars pressure conditions than at 760 torr pressure. Hence it can be concluded that atmospheric pressure, rather than gravitational acceleration, is the predominant parameter in controlling settling speed of particles here. Since the Moon has no atmosphere (and hence the pressure is of the order of 10 -14 torr), the settling speed will be faster than at 5 torr pressure.

Analysis of the Apollo Lunar Module descent footage shows that the soil particles excavated by the rocket thruster, flow in a straight line; they do not move anywhere else as is observed on Earth where a

9

dust cloud is formed very quickly (Metzger et al. 2009; Immer et al., 2006). In addition, as was observed during the Surveyor 5 descent, firing of the main engine for 0.55 second produced no crater or significant dust cloud1.

5.0 Deployment Concept for Lunar Laser Ranging In order to reduce deployment complexity, it is required to reduce the number of actuators and replace those by springs and pin pullers. Figure 14 illustrates an example of a deployment system that takes advantage of the energy stored in springs as well as in gas. In this example, the CCR system is mounted on the Astrobotic Inc. lunar lander. Astrobotic is one of the Google Lunar XPrize teams and plans to land on the Moon in 2014 (Gump, 2012). One of the instruments that may fly on its first mission is CCRLLRRA-21.

Note that this is a one-shot deployment and once triggered the CCR deployment cannot be stopped, and the mechanism pulled out, and moved to another location, if for example a small rock stops the penetration (though, as explained in the section “Why Gas Drilling”, the probability of encountering a rock is < 0.5%). This “stop-retract-move” capability is possible but with a more complex and in turn heavier and more expensive mechanism.

With reference to Figure 14, the LLRRA-21CCR system is integrated with its deployment system on the lander deck. Upon touch down, the first pin puller is released allowing the deployment system linear stage move the CCR to the edge of the lander deck. Once the linear stage reaches end of travel, the preloaded, redundant torsion springs cause the CCR rod to pivot upwards. Once the Pivot Assembly reaches hard stop, the limit switch is triggered and the Pin Puller #2 is actuated, releasing the CCR Rod. The rod then drops to the ground, under its own weight. Upon the impact, the Puncture Rod connected to the cone is pushed upwards and punctures sealed gas cylinder. As soon as the gas is released from gas cylinder, it flows down the rod and exits out the nose cone jets allowing the anchor to penetrate into the subsurface. Once the rod starts penetrating into the subsurface, a non-contact position sensor triggers Pin Puller #3 releasing spring loaded arms. This allows the CCR rod to plunge into soil unhindered. The rod is designed to reach > 50 cm depth.

We would rely on the experimental test data from tests conducted in a vacuum chamber and at 1/6 th g (during the reduced gravity flights) to determine how much gas is required to stop the CCR rod penetration at approximately 50 cm depth. However, if mass and budget permits, a more expensive system could be designed employing a range finder and a gas valve.

In this design, gas is stored in a tank placed between the rod and the CCR. The tank is opened and allowed to vent, providing 18 g of gas at or near 4.0 liters per minute, much like in Test 3 (in Test 3, the CCR rod was able to reach a depth of 50 cm in approximately 4 minutes using 18 g of air). As seen in Figure 15, a titanium tank capable of holding this mass of nitrogen, based on tank design done for the Mars Science Laboratory's SAM instrument suite, would have a mass of only 467 g.

In Test 3, the volume of excavated soil was approximately 80 cm3 (hole surface area was ~14.5 mm diameter while the whole depth was 50 cm). This corresponds to a mass of soil of ~160 g (at the density of 2 g/cc). Since the mass of gas used was 18 g, the ratio of soil mass to gas mass is 1:9.

In Test 3, the diameter of the rod was 13 mm while the diameter of the cone was 14.5 mm allowing approximately 0.75 mm of clearance on each side of the rod for the soil particles to move up the hole. Since the rod does not see large forces (it supports a weight of the CCR only), the diameter of the rod and in turn the diameter of the cone could be substantially reduced. For example, decreasing the

1 http://solarsystem.nasa.gov/missions/profile.cfm?Sort=Alpha&Letter=S&Alias=Surveyor%2005

10

diameter of the cone from 14.5 mm to 6 mm would reduce the excavated area from 160 mm 2 to 30 mm2

(i.e. by a factor of 5) and the excavated volume from 80 cm3 to 15 cm3. In turn, the mass of the required gas would be reduced to approximately 2 g and the mass of the gas tank will be reduced from 467 g to 100 g.

6.0 Conclusions In this paper, we present a novel method of deploying LLRRA-21 using a Solid Corner Cube Reflector on the Moon. The proposed emplacement approach uses a gas-powered drill consisting of a >50 cm long, slim hollow rod with perforated anchor-cone at its lower end and the CCR mounted to the top. Gas supplied from a small tank is directed into and down the rod and out through the cone lofting the soil out of the hole and allowing the rod to sink to 50 cm depth.

To determine the system performance, we conducted several tests in compacted JSC-1a lunar soil simulant and inside a vacuum chamber. In all tests, the rod with a cone at the end was initially placed on the soil surface. Once vacuum was reached, the soil bin was temporarily vibrated to compact the soil.

The test involved opening and monitoring gas flow and recording the rate of the rod sinkage into the soil. The rod and the CCR housing weighed 1.6 kg. This mass provided the only vertical force (16 N) to the cone. In each case, the gas-ejected soil particles traveled up the hole (between the hole and the rod), and were effectively deflected sideways by the surface shield.

In several tests, the rod successfully sunk under its own weight to a depth of 50 cm and in 4-6 minutes. The gas pressure used to excavate the soil underneath the cone was 101 kPa (absolute) and the mass of gas used was 10-20 g. Since the rod does not carry substantial weight, the rod diameter and in turn the cone diameter can be reduced from 14.5 mm to 6 mm. Reducing the cone diameter to 6 mm will reduce the mass of gas to 2 g and in turn the mass of gas tank from 467 g to 100 g.

These tests successfully demonstrated the gas assisted drilling approach. Thus, the LLRRA-21 will require no power and could be deployed from light-weight lunar platforms.

As a final note, releasing of gas into lunar regolith may affect in-situ exobiology or chemistry; however, since the goal of the system is to mechanically anchor CCR to the sub-surface, the gas contamination will not be an issue.

AcknowledgementsThis project was funded by the National Aeronautics and Space Administration I (NASA) Lunar Science Institute (NLSI) Lunar University Network for Astrophysics Research (LUNAR) project (Principal Investigator Dr. Jack Burns).

ReferencesAlshibli K., and A. Hasan, Strength Properties of JSC-1A Lunar Regolith Simulant, J. Geotech.

Geoenviron. Eng. 135, 673 (2009); doi:10.1061/(ASCE)GT.1943-5606.0000068Carrier, W. D., Olhoeft, G. R., and Mendell, W. (1991) “Physical Properties of the Lunar Surface”, Lunar

Sourcebook, G. Heiken, D. Vaniman, and B. M. French, eds., Cambridge University Press, Cambridge, pp. 475-594.

Carrier, D., The four things you need to know about the geotechnical properties of lunar soil, Lunar Geotechnical Institute, September 2005.

Currie D., Dell'Agnello S., Delle Monache G. A Lunar laser ranging retroreflector array for the 21st century. (2011) Acta Astronautica, 68 (7-8), pp. 667-680.

11

Fuerstenau, S. D. (2006), Solar heating of suspended particles and the dynamics of Martian dust devils, Geophys. Res. Lett., 33, L19S03, doi:10.1029/2006GL026798.

Gouache, Thibault and Brunskill, Christopher and Scott, Gregory and Gao, Yang and Coste, Pierre and Gourinat, Yves Regolith simulant preparation methods for hardware testing. (2011) Planetary and Space Science, vol. 58 (no 14-15). pp. 1977-1984. ISSN 0032-0633

Gump, D., Astrobotic Inc., personal communication, May 2011Heiken, G., D. Vaniman, and B. M. French, eds., (1991) Lunar Sourcebook, Cambridge University Press,

CambridgeImmer, C., J. Lane, P. Metzger, and S. Clements, “Apollo Video Photogrammetry Estimation of Plume

Impingement Effects,” Earth and Space 2008, 11th Biennial ASCE Aerospace Division International Conference on Engineering, Construction and Operations in Challenging Environments, ASCE, Reston, VA, 2008.

Johnson, B. V., W. W. Roepke, and K. Strebig, (1973), Shear Testing of Simulated Lunar Soil in Ultrahigh Vacuum, US Bureau of Mines Report of Investigations 7814

Kömle, N.I., P. Weiss, K.L. Yung: Considerations on a suction drill for lunar surface drilling and sampling: I. feasibility study, Acta Geotechnica, 3, 201-214, 2008.

Mclemore, C., (2010), Simulant Listings, http://isru.msfc.nasa.gov/lib/Documents/Simulant-listing.pdf, Accessed 12 July 2011

Metzger, P., C. Immer, C. Donahue, B. Vu, R. Latta, III, M. Deyo-Svendsen, “Jet-Induced Cratering of a Granular Surface With Application to Lunar Spaceports,” Journal of Aerospace Engineering, Vo. 22, No. 1, 2009, pp. 24-32.

Mumm, E., K. Zacny, N. Kumar, Percussive and Proboscis Based Lunar Heat Flow Probes, AGU Fall Meeting, San Francisco, 15-19 December, 2009.

Nagihara, S.; Taylor, P. T.; Williams, D. R.; Saito, Y., Long-Term Warming of Surface and Subsurface Temperatures Observed at Apollo 15 and 17 Sites: Implications for Future Lunar Geophysical Missions Ground-Based Geophysics on the Moon, January 21-22, 2010, held in Tempe, Arizona. LPI Contribution No. 1530, p.3008

Schaefer, J., J. Neathery, and J. Stencel, (1994) Evaluation of a Pneumatic Marian Soil Sampler Concept, NASA-CR-197539

Sullivan T., E. Koenig, C. Knudsen, and M. Gibson, Pneumatic conveying of materials at partial gravity, Journal of Aerospace Engineering, Vol. 7, No. 2, April, 1994.

Zacny, K., K. Huang, M. McGehee, A. Neugebauer, S. Park, M. Quayle, R. Sichel, G. Cooper, “Lunar Soil Extraction Using Flow of Gas” Proceedings of Revolutionary Aerospace Systems Concepts - Academic Linkage (RASC-AL) Conference. April 28-May 1, 2004. Cocoa Beach, Florida.

Zacny, K., M. Quayle and G. Cooper, Enhancing Cuttings Removal with Gas Blasts While Drilling on Mars; J. Geophys. Res., 110, E04002, 2005

Zacny, K., J. Wilson, J. Craft, V. Asnani, H. Oravec, C. Creager, J. Johnson, and T. Fong, Robotic Lunar Geotechnical Tool, ASCE Earth and Space 2010, 15-17 March 2010, Honolulu HI.

Zacny, K., G. Mungas, C. Mungas, D. Fisher, and M. Hedlund, Pneumatic Excavator and Regolith Transport System for Lunar ISRU and Construction, Paper No: AIAA-2008-7824 and Presentation, AIAA SPACE 2008 Conference & Exposition, 9 - 11 Sep 2008, San Diego Convention Center, San Diego, California

Zacny, K., R. Mueller, G. Galloway, J. Craft, G. Mungas, M. Hedlund, and P. Fink, Novel Approaches to Drilling and Excavation on the Moon, AIAA-2009-6431, Space 2009.

Zacny K., J. Craft; M. Hedlund; P. Chu; G. Galloway; R. Mueller, Investigating the Efficiency of Pneumatic Transfer of JSC-1a Lunar Regolith Simulant in Vacuum and Lunar Gravity During Parabolic Flights. AIAA Space 2010, AIAA-2010-8702, Aug 31-Sep 2, 2010, Anaheim, CA

Zacny, K., D. Currie, G. Paulsen, Development and Testing of Gas Assisted Drill for the Emplacement of the Corner Cube Reflector System on the Moon, NLSI Forum, 19-22 July 2011, Moffett Field, CA

12

Table 1. Trade Matrix for Subsurface Penetration

Table 2. Trade Matrix for Subsurface Systems Applied to Small Lunar Landers

Table 3. Data from gas-spear tests under atmospheric conditions.

Table 4. Results of CCR deployment system vacuum chamber testing. Test 6 is a proposed test which would prove the concept of carrying and venting an onboard gas tank.

Figure 1. Pneumatic spear approach for excavating holes in soil. Note the deflector on top to divert the ‘dusty-gas’ away from the lander.

Figure 2. Data showing vertical push force applied to a penetrator. The push force dropped to zero when regolith was being lofted out of the hole with compressed gas.

Figure 3. CCR deployed pneumatically in Mauna Kea, HI tephra (lunar analog) in February 2010. Left: the system before being deployed. Right: system after being deployed. Note that no dust shield was used since not much dust was generated. The metallic sheets are aluminum foils.

Figure 4. Particle size distribution of the three different rock powders (Zacny et al., 2005) Figure 5. Three different powders inside individual test tubes in a 5 torr vacuum. From left to right: Ohio

Sandstone powder (OS), Santa Barbara Sandstone powder (SB), Basalt powder (B).

Figure 6. Open vacuum chamber following test 1. The CCR system was successfully deployed 50 cm into the soil. The chamber is 1 meter wide, 1 meter deep, and 3.5 meters tall.

Figure 7. The CCR consists of fused silica glass with mutually orthogonal mirrors on three of its faces and is designed to reflect light directly back toward its source. The mass of the CCR is 0.852 kg.

Figure 8. The CCR housing sits atop the anchor, and is held in place with a bolt. The top of the anchor can be seen in the lower part of the image. The housing has a mass of 1.256 kg.

Figure 11. Chamber interior prior to experimental tests. Figure 9. The rod anchors the CCR system into the Lunar soil. The rod is hollow which allows gas to

flow from a source (in experimental tests, from a feed line) to the jets in the cone below. The image shows the upper part of the anchor. The combined mass of the rod, a threaded insert to connect to the cone, and a block to connect the rod to the CCR housing is 332 g.

Figure 10. The cone is attached to the bottom of the rod with a threaded insert. The rod and the cone

forms the CCR anchor. Four jets surround the cone's tip and are spaced apart 90º. The jets expel gas downward and slightly outward. The cone has a mass of 34 g.

13

Figure 12. In Test 5 a copper mesh tip was used. The soil was sprayed in all directions while the bottom of the mesh got clogged very quickly. The penetration rate stopped at a depth of 10 cm. The exterior of the dust shield is seen at the bottom and right side of the image.

Figure 13. Settling speed of particles under Mars (5 torr) and Earth (760 torr) pressures (Fuerstenau,

2006). Figure 14a: Stowed position of the CCR (see arrow)Figure 14b: CCR Deployment Mechanism in stowed positionFigure 14c: Pin Puller #1 is actuated, releasing the spring loaded linear stage. Linear stage is on rollers for

smooth deploymentFigure 14d: Linear stage reaches end of travel. Roller (1) constraining CCR is free, due to slot in roller

track (2). Preloaded, redundant torsion springs (3) cause CCR rod to pivot upwardsFigure 14e: Pivot Assembly reaches hard stop. Limit switch is triggered. Pin Puller #2 is actuated,

releasing the CCR RodFigure 14f: Nose Cone (3) contacts ground. Puncture Rod (2) is pushed upwards. Tip punctures sealed

gas cylinder or turns on a valve (1). Gas is released from gas cylinder, exiting out the nose cone jets (4). CCR Rod plunges into the soil

Figure 14g: (Left) Non-contact position sensor is triggered at end of travel. (Center) Pin Puller #3 is actuated, releasing spring loaded arms. (Right) CCR Rod continues to plunge into soil unhindered.

Figure 14h: Final position of the CCR with respect to landerFigure 14. A prototype design for a pneumatically actuated CCR deployment system on the Moon. Figure 15. Graph showing the cubic dependence of air mass and tank mass with tank diameter. Note the

y-axis for air is scaled by a factor of 10. The tank needed to hold 18 g of air is ~10 cm. When the cone diameter is reduced from 14.5 mm to 6 mm, only 2 g of gas will be required.

14

Table 1. Trade Matrix for Subsurface PenetrationHow it works Advantages Disadvantages

Drill

Apollo approach:

Manually drill a rod 1m depth and leave it in place.

Proven during Apollo 15-17

Complex Automation

High Mass & Power (450 Watt)

Hammer/

Percussive

Hammer the rod to 1m Simple deployment

Lighter than drill

Need vacuum rated percussive hammer

PneumaticExcavate using gas Very low mass and power

Can use Helium from propulsion system

Potential dust if no cover employed

Table 2. Trade Matrix for Subsurface Systems Applied to Small Lunar LandersCan get to 0.5-

1m? Mass Power Volume Complexity Go / NoGo

Drill Y H H H H NG

Percussive Y M M M M NG

Pneumatic Y L L L L G

Table 3. Data from gas-spear tests under atmospheric conditions.

    Regulator Measured Gas Mass Gas Actual Depth

    Pressure Vol. Flow Rate Density (test duration ~2 min) WOBPenetrate

dTest# Gas [kPa] [psi] [SCFM] [LPM] [kg/m^3] [g] [N] [cm]

1 n/a n/a   n/a n/a n/a n/a 450 <52 Air 138 20 1.5 42 3.0 240 0 603 Air 35 5 0.8 23 1.7 78 0 604 He 35 5 0.8 23 0.2 11 0 60

15

Table 4. Results of CCR deployment system vacuum chamber testing. Test 6 is a proposed test which would prove the concept of carrying and venting an onboard gas tank.

Test #

Add’l.

Mass

Cone Diam.

Inlet Pressur

e

Flow Rate Approx. Duratio

n

Gas Volum

e

Gas Mas

s

Penetration

Notes

[kg] [mm] [kPa abs.]

[liters/min]

[min] [m^3] [g] [cm]

1 2 21 101.3 0.2 ramped to 4.0

4:00 0.008 9 50 Shield vents too small and became clogged.

2 0 21 101.3 0.2 ramped to 4.0

7:00 0.012 14 38 Shield vents enlarged beginning with this test. Shield began to slide during the last 10 cm of penetration and may have impeded progress.

2A 0 14.5 101.3 0.2 ramped to 3.0

6:00 0.009 10 50 Repeat of test 2, but cone was thinned (resulting in final design) and max flow rate was 3.0 LPM. Shield began to slide during last 10 cm of penetration and may have impeded progress.

3 0 14.5 101.3 4.0 4:00 0.016 18 50 Penetration is always in steps of ~1 cm, even at the beginning of the test. Never a smooth, constant penetration rate even when flow is constant.

4 0 14.5 768.0 0.0 ramped to 0.65 (4.0 Std LPM)

4:00 0.008 9 38 Shield became very crooked at 38 cm depth, test stopped. It seems flow is choked at the cone - hence no significant improvement in penetration rate.

5 0 14.5 101.3 0.2 ramped to 5.0

3:00 0.008 9 10 Copper mesh tip used, entire 10 cm penetration due to the crater formed by the airflow. No penetration deeper than the bottom of this crater. Very little dust thrown upward.

16

Figure 1. Pneumatic spear approach for excavating holes in soil. Note the deflector on top to divert the ‘dusty-gas’ away from the lander.

Figure 2. Data showing vertical push force applied to a penetrator. The push force dropped to zero when regolith was being lofted out of the hole with compressed gas.

17

Figure 3. CCR deployed pneumatically in Mauna Kea, HI tephra (lunar analog) in February 2010. Left: the system before being deployed. Right: system after being deployed. Note that no dust shield

was used since not much dust was generated. The metallic sheets are aluminum foils.

Figure 4. Particle size distribution of the three different rock powders (Zacny et al., 2005)

18

Figure 5. Three different powders inside individual test tubes in a 5 torr vacuum. From left to right: Ohio Sandstone powder (OS), Santa Barbara Sandstone powder (SB), Basalt powder (B).

Figure 6. Open vacuum chamber following test 1. The CCR system was successfully deployed 50 cm into the soil. The chamber is 1 meter wide, 1 meter deep, and 3.5 meters tall.

19

Figure 7. The CCR consists of fused silica glass with mutually orthogonal mirrors on three of its faces and is designed to reflect light directly back toward its source. The mass of the CCR is

0.852 kg.

Figure 8. The CCR housing sits atop the anchor, and is held in place with a bolt. The top of the anchor can be seen in the lower part of the image. The housing has a mass of 1.256 kg.

20

Figure 9. Chamber interior prior to experimental tests.

Figure 10. The rod anchors the CCR system into the Lunar soil. The rod is hollow which allows gas to flow from a source (in experimental tests, from a feed line) to the jets in the cone below. The image shows the upper part of the anchor. The combined mass of the rod, a threaded insert to

connect to the cone, and a block to connect the rod to the CCR housing is 332 g.

21

Figure 11. The cone is attached to the bottom of the rod with a threaded insert. The rod and the cone forms the CCR anchor. Four jets surround the cone's tip and are spaced apart 90º. The

jets expel gas downward and slightly outward. The cone has a mass of 34 g.

Figure 12. In Test 5 a copper mesh tip was used. The soil was sprayed in all directions while the bottom of the mesh got clogged very quickly. The penetration rate stopped at a depth of 10 cm. The

exterior of the dust shield is seen at the bottom and right side of the image.

22

Figure 13. Settling speed of particles under Mars (5 torr) and Earth (760 torr) pressures (Fuerstenau, 2006).

Figure 14a: Stowed position of the CCR (see arrow)

Figure 14b: CCR Deployment Mechanism in stowed position

23

Figure 14c: Pin Puller #1 is actuated, releasing the spring loaded linear stage. Linear stage is on rollers for smooth deployment

Figure 14d: Linear stage reaches end of travel. Roller (1) constraining CCR is free, due to slot in roller track (2). Preloaded, redundant torsion springs (3) cause CCR rod to pivot upwards

Figure 14e: Pivot Assembly reaches hard stop. Limit switch is triggered. Pin Puller #2 is actuated, releasing the CCR Rod

Figure 14f: Nose Cone (3) contacts ground. Puncture Rod (2) is pushed upwards. Tip punctures sealed gas cylinder or turns on a valve (1). Gas is released from gas cylinder, exiting out the nose cone jets (4). CCR

Rod plunges into the soil

24

Figure 14g: (Left) Non-contact position sensor is triggered at end of travel. (Center) Pin Puller #3 is actuated, releasing spring loaded arms. (Right) CCR Rod continues to plunge into soil unhindered.

Figure 14h: Final position of the CCR with respect to lander

Figure 14. A prototype design for a pneumatically actuated CCR deployment system on the Moon.

25

Figure 15. Graph showing the cubic dependence of air mass and tank mass with tank diameter. Note the y-axis for air is scaled by a factor of 10. The tank needed to hold 18 g of air is ~10 cm.

When the cone diameter is reduced from 14.5 mm to 6 mm, only 2 g of gas will be required.

26