P08456: Lighting System for an Underwater ROV



P08456: Lighting System for an Underwater ROV

Detailed Design Review Information Packet

Team Members:

Jeremy Schiele: Project Manager

Jonathan Lent: Housing Design

Justin VanSlyke: Mounting System

Ryan Seeber: Control System

Benoit Hennekinne: Circuit Design

Table of Contents

Project Information 3

Engineering Specifications 3

Risk Assessment (FMEA) 5

Mounting Stress Analysis 6

Thermal Management System: 9

Housing Stress Analysis 10

Assembly & Construction: 12

Galvanic Corrosion 14

Bill of Materials 15

Attached: Drawings/Assembly Packet

Light Housing

Lens Cap

Borosilicate Lens

Motor Housing

Hard-mount Extension

Hinge Mounting

Light Unit Assembly

Motor Unit Housing Assembly

Programming Flowchart

Project Information:

The goal of this project is to create a lighting system to be used on an underwater remotely operated vehicle (ROV). A first generation underwater ROV was previously designed by Senior Design project P06066 and used a High Intensity Discharge (HID) lighting system with basic on/off control. That ROV platform is now owned by Hydroacoustics Inc. (HAI) and uses LED lights. This project is focused on the improvement of the HID lighting system by using lower power Light Emitting Diodes (LEDs) and providing greater control capability.

The design’s housing is to be shared with the thruster units being designed concurrently by P08454. The light produced must be able to be dimmed from the control module, this same software package should control the thruster unit as well. LEDs will be used as the light source to minimize power consumption and prolong the unit’s battery life. A study into the implications and benefits of lighting with different spectrum lights is being conducted to determine if a second color of LEDs is required, and if so which color will provide the best light penetration underwater.

Engineering Specifications/Customer Needs:

Shown below is a chart of the specifications set for the LED lighting project, and acceptable and ideal values for each. It should be noted that the final design has the ability to consume more power than the ideal value. However, at a power level substantially below that, the LEDs still emit more than the preferred amount of light. Because of this, the power and luminous flux metrics are those given for a current of 350 mA to the LEDs. The only specification not met is that of the maximum possible heat produced from each light unit. As a solution to mitigate this, planned in the software is a regulator to limit the current to the LEDs so they do not produce the maximum amount of heat possible. When used underwater, the heat dissipation will not be a much of a problem, or a possible source of injury, as possible when the unit is used in air. Also, because the software has yet to be written, its ability to control both thruster and light unit is unknown, but is being planed for. The time for removal will require testing, and will be completed once a prototype is developed.

[pic]

Table 1: Engineering Specifiactions and current design values

A portion of the customer needs generated are able to be addressed at this time, the remaining needs are either standards that the project should be held to (Need no. 5, 10, 12, 13) or have been planned for, and will be addressed at a later time with this design (Need no. 1,7). The light will provide ample light for a camera system, and will be adjustable in intensity and color (white or colored). The power consumed by the LEDs can be increased so that they consume more than the previous design, but will provide more light than the previous design at a lower power level, eg: 300 lumens are produced using 350 mA (3.30 W), but if desired, a potential 675 lumens can be produced with 1 A (12 W).

The adaptability of the system has also been kept in mind. A simple bolt pattern will allow it to mount in many configurations, and make planning to mount this light on future platforms easier. The thruster unit will share a portion of our hosing, and add an extension to allow the motor and magnetic couple to fit inside.

[pic]

Table 2: Customer needs list

Risk Assessment:

The high risk areas and technologies of our project are directly related to the operating conditions the unit will be used under. Submerged to 400ft plus depths has separate risks than operating it out of the water in ambient air. Underwater, the heat is better transmitted via convection into the surrounding water; in air the pressure requirements are not existent, but the heat produces by the LEDs is not able to be dissipated as easily. Because of this, the Failure Mode and Effects Analysis (FMEA) is done under the main operating condition-submerged in water.

The area of highest risk involves the glass lens required to transmit the light out of the housing. It is the most brittle material. The FMEA scenario takes into account the pressure it is under at 500 ft depth, and having some other unknown force to break it (shipwreck, boat hull, etc.) This emergency would be minimized by the fuse connected to each light, contained in the ROV’s main dry storage unit (where electrical controls are housed). The fuse will prevent the batteries from being drained prematurely due to a short caused by sea-water in contact with the PCBs.

[pic]

Mounting System Stress Analysis:

We have decided to go with two different mounting systems for our design. The main reason for this choice is there is a large difference in requirements between the lighting and thruster variants on our design. The driving factors behind this choice are listed below:

• The light housing needs adjustability in several axes (two, in our design).

• The light housing may need to be mounted off the ROV farther

• The thruster housing needs to be as rigid as possible.

• The thruster housing will need to withstand much more stress.

Our light housing mount system is basically a locking hinge. The ROV side has one threaded bolt hole to allow the whole assembly, housing included, to rotate around the bolt before being tightened into position. The second degree of freedom comes from the hinge in the center of the device. Using a wing nut it is possible to change the angle of the light quickly by hand and then secure it down before launching. On the light housing side there is a flange producing the two bolt pattern presented on the housing.

[pic]

To address the concerns presented by the addition of the motor we have created a simple hard mount extension arm. This device consists of two plates connected by a central rod. The two plates have three holes machined in them, one in the center to accept the rod through a loose press fit which will then be welded and two bolt holes for the hardware to secure the extension to the ROV and the housing.

[pic]

Both of these systems have been analyzed with the SolidWorks CFD tool “Cosmos Express” under both their expected and “worst case” loading scenarios. For the hinge assembly an assumption of a solid connection at the hinge was made to facilitate this modeling. The expected loading assumed for the hinge was one (1) pound directly downwards; this represents the weight of the fully assembled light housing. This resulted in a minimum factor of safety of 82.71. The worst case loading was assumed to be 20 pounds normal to the axis of the extension and the previous example; this represents the ROV running into something under its own power or a user accidentally bumping into the light housing. This results in a factor of safety of 3.84.

[pic]

Loading Conditions for Lighting Mount

[pic]

Displacement for Lighting Mount Worst Case

[pic]

Stresses for Lighting Mount Worst Case

For the hard mount extension the forces assumed for the expected condition were three (3) pounds directly downwards; this represents the increased weight of the housing additions and motor assembly. We assumed the motor had a thrust of 30 pounds so represented this by applying a force normal to the axis of protrusion and the weight force. This resulted in a factor of safety of 3.15. The worst case loading was assumed to be when the motor mount was snagged and two motors in two directions were running (60 lb thrust). This makes three forces, the weight and two motors perpendicular to each other. This resulted in a factor of safety of 2.27.

[pic]

Loading conditions for Motor Mount

[pic]

Displacement for Motor Mount Worst Case

[pic]

Stresses for Motor Mount Worst Case

The idea of a break away mounting system was brought up in our previous design review. This would allow for the ROV to free itself without requiring diver intervention. We had not previously considered this idea, but after thinking about it we came to the conclusion that it is not feasible to design a breakable yet sturdy mount for our system. The main reason for this decision was the fact that the thrusters can only put out around 30 pounds of thrust as a best case. Both the thruster mount and the lighting mount have factors of safety of more than two under this loading condition. In order for a break away system to be useful it would need to release at this maximum load, but the resistance of the water and even above water handling could put loads on the mounting system approaching or exceeding the thruster capabilities.

Another option we researched was exploding or self shearing bolts. These would have been the answer to the above problem (with more design considerations for connections to the bolts themselves) however; we were unable to find any that were near the size envelope of our system so these are also not feasible.

Thermal Management System:

This aspect of design is very important because of the very enclosed space in which we must operate heat generating devices. The devices we identified as a major heat source were high power LEDs. Each of our LEDs requires three (3) watts of power to operate. After consulting with the manufacturer we came to the conclusion that at worst, 85% of this power is converted to heat while the rest is emitted as light. This means that at any one time we have at most 7.65 watts of heat to dissipate.

We will be providing “heat traces” from the conductive pads on the LEDs to the mounting holes on the PCB. These heat traces need to be much thicker than a traditional surface trace in order to provide effective conductance between the LED and the housing. As a result we will be required to get have the LED boards made with 6oz copper layers on each of the outer layers, this equates to a trace thickness of 0.0082in (0.00021m) on each side. Beneath the LED will be two small vias used to pass some of the heat through to the trace on the underside of the PCB. The LEDs will be arranged so that the heat pads are 0.4in (0.01m) from the mounting holes. The pattern will be such that only one LED of each “bank” is connected to each hole. The width of the trace between the LED and the mounting hole is 0.2in (0.0051m).

Since the housing has a very large mass of Aluminum and there is cool water flow over the outside of the housing, the majority of the thermal resistance will be caused by the heat traces. Using the dimensions of the trace we are able to determine the ∆T required between the LED pad and mounting hole to dissipate the 2.55 watts of heat from each LED.

[pic]

Where Q is the heat produced by the LED, k is the thermal conductivity coefficient, A is the cross-sectional area of the trace, and x is the distance the heat energy needs to be moved.

[pic]

[pic]

The maximum recommended long term operating temperature of the Luxion Rebel LED is 275°F (135°C). If we keep the assumption that the trace to housing path is the limiting factor in the heat conductance then we can show the temperature of the LED pad to be:

[pic]

Using 41°F (5°C) to represent the water temperature at depth, Tpad is 121.66°F (49.8°C), under the maximum operating temperature. If the surrounding temperature is 90°F (32.2°C) to represent using the LEDs out of water on a summer day Tpad is 170.66°F (77°C). This second temperature is an underestimate because of the large difference in convective capabilities between water and air, but it is still more than 100°F below the maximum LED temperature.

While all the other components individually do not produce as much heat as the LEDs they do still need to be managed thermally or risk overheating. We plan on tying all of the ground traces to the standoff mounting holes on the PCBs, thereby providing a conductive path through the boards to the housing which will dissipate the heat into the water. This, in addition to the free convection inside the housing will be enough to handle the much lower heat generation of the rest of the circuitry inside the housing.

Housing Stress Analysis:

The three main components of the light housing have been simulated to a depth of 500 ft using Solidworks Cosmos FEA program. The shared housing unit (the real portion), which houses most of the LED light and motor electronics, had a maximum stress of 1.829E3 psi, and an overall minimum factor of safety of 6. The lens cap, that the LED board is fixed to, under the same conditions, had a maximum stress of 6.275E2, and a minimum factor of safety of 15. The borosilicate glass lens, which is commonly sold under the trademark: Pyrex, has a max stress of 2300 psi, which is not must for a metal, but because the glass can not flex, and distribute the internal pressure, it has a low yield strength. This yield strength can vary from 1000-4000 psi because this is our lowest factor of safety; further testing will be done with both modeling and actual pressure tests conducted in the Hydroacoustics Inc. pressure chamber.

[pic]

Figure 1: Von Mises stress profile for main housing unit

[pic]

Figure 2: Von Mises stress profile for Lens holder.

[pic]

Figure 3: Von Mises stress profile for Glass Lens.

Assembly:

The ability for the unit to be assembled has been proven with an assembly model of our designed and OEM parts. The main change that has occurred is moving from using a hex nut on the inside of the housing to fix the bulkhead connector, to having the ¾-16 threads cut directly into the housing. This allows for the bulkhead connector to mate directly with the housing, and allows for a permanent seal with a waterproof thread locking compound. There are two main benefits of this sealing method: 1) the bulkhead connector is no longer a possible penetration point for water, and 2) more surface area is left on the inside of the housing for the mounting of the circuit boards.

The main construction requirements for the light unit are for the aluminum housing pieces, and the PCBs. RIT has the capability to machine the housing, which was a driving factor in picking Aluminum as the housing material, it is mush easier to machine that stainless steel, and will not require as much of a labor charge from the CNC lab. The PCB design is under way, but none of the boards require more than 70% of their surface area to be populated with surface mount parts. The largest part will actually be the board-board connectors used (not shown on assembly).

[pic]

Figure 4: Assembly cross-section. NOTE: PCB modeled w/o parts population.

[pic]

Figure 5: Isometric view of exterior of light housing unit.

Galvanic Corrosion:

The corrosion of one of two dissimilar metals is referred to as galvanic corrosion, or dissimilar metal corrosion, and uses the property of electrolysis to diminish the weaker of the two metals. The weaker metal, or anode, has a lower number on the galvanic table in relation to the second metal, the cathode. The further these two metals are apart from each other, the stronger the electrolytic process. For this reason, we are using 7075 Al for the housing, and 304 Stainless (18-8) for the hardware. The aluminum housing will also be anodized to further increase its resistance to oxidization and corrosion. As a second possibility, a previously anodized part can be chromated to increase the corrosion FOS by 2. Typically a difference of 0.15 V or less on the Anodic Index is recommended for harsh environments (sea water included), 7075 and 304 have anodic numbers of 0.90 V and 0.85 V respectively, for a difference of 0.05 V. (Corrosion-)

1. Magnesium

2. Mg alloy AZ-31B

3. Mg alloy HK-31A

4. Zinc (hot-dip, die cast, or plated)

5. Beryllium (hot pressed)

6. Al 7072 clad on 7075

7. Al 2014-T3

8. Al 1160-H14

9. Al 7079-T6

10. Cadmium (plated)

11. Uranium

12. Al 218 (die cast)

13. Al 5052-0

14. Al 5052-H12

15. Al 5456-0, H353

16. Al 5052-H32

17. Al 1100-0

18. Al 3003-H25

19. Al 6061-T6

20. Al A360 (die cast)

21. Al 7075-T6

22. Al 6061-0

23. Indium

24. Al 2014-0

25. Al 2024-T4

26. Al 5052-H16

27. Tin (plated)

28. Stainless steel 430 (active)

29. Lead

30. Steel 1010

31. Iron (cast)

32. Stainless steel 410 (active)

33. Copper (plated, cast, or wrought)

34. Nickel (plated)

35. Chromium (Plated)

36. Tantalum

37. AM350 (active)

38. Stainless steel 310 (active)

39. Stainless steel 301 (active)

40. Stainless steel 304 (active)

41. Stainless steel 430 (active)

42. Stainless steel 410 (active)

43. Stainless steel 17-7PH (active)

44. Tungsten

45. Niobium (columbium) 1% Zr

46. Brass, Yellow, 268

47. Uranium 8% Mo.

48. Brass, Naval, 464

49. Yellow Brass

50. Muntz Metal 280

51. Brass (plated)

52. Nickel-silver (18% Ni)

53. Stainless steel 316L (active)

54. Bronze 220

55. Copper 110

56. Red Brass

57. Stainless steel 347 (active)

58. Molybdenum, Commercial pure

59. Copper-nickel 715

60. Admiralty brass

61. Stainless steel 202 (active)

62. Bronze, Phosphor 534 (B-1)

63. Monel 400

64. Stainless steel 201 (active)

65. Carpenter 20 (active)

66. Stainless steel 321 (active)

67. Stainless steel 316 (active)

68. Stainless steel 309 (active)

69. Stainless steel 17-7PH (passive)

70. Silicone Bronze 655

71. Stainless steel 304 (passive)

72. Stainless steel 301 (passive)

73. Stainless steel 321 (passive)

74. Stainless steel 201 (passive)

75. Stainless steel 286 (passive)

76. Stainless steel 316L (passive)

77. AM355 (active)

78. Stainless steel 202 (passive)

79. Carpenter 20 (passive)

80. AM355 (passive)

81. A286 (passive)

82. Titanium 5A1, 2.5 Sn

83. Titanium 13V, 11Cr, 3Al (annealed)

84. Titanium 6Al, 4V (solution treated and aged)

85. Titanium 6Al, 4V (anneal)

86. Titanium 8Mn

87. Titanium 13V, 11Cr 3Al (solution heat treated and aged)

88. Titanium 75A

89. AM350 (passive)

90. Silver

91. Gold

92. Graphite

[pic]

The overall cost for parts for each light unit will be approx. $352.55. Some of this cost will be able to be spread out over the other projects within the project track though because they will share the same materials. For example, the PCB manufacturing is $33.00 for one 60 in2, two PCB sheets may be enough for the LED light, Thruster, and Robotic Platform teams, so the cost would be spread proportionally between them. The raw materials required for the entire project- Aluminum stock for housing and PCB manufacturing @ $33.00 per 60 in2 (2 layer board), is estimated to be $397.41. The portion of the thruster unit housing that will be supplied by this team will cost $205.00, plus a portion of the raw materials, as stated above. The longest lead time, and highest priced item is the connector and cabling system; because the connectors are made to order, there is an approx. 4 week lead time.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download