CHAPTER IV – METHODOLOGY
The objective of this project was to follow the system engineering process, research the field, complete a requirements analysis, design a fixture, complete the first article build, and conduct test and evaluation. Particularly, the project must determine and include the optimal light spectrum(s) and intensity for coral growth and color. The systems engineering process included the work breakdown structure, responsibility assignments, schedule, risk analysis, and other related topics. However, that information will not be presented on this website as it is not directly related to the hobby. If you are interested in this data, please contact me directly. Test and evaluation methodology included control factors, test combinations, constant variables, uncontrolled variables, test equipment, and calibration and measurement standards.
Systems Engineering Process
Project Work Breakdown Structure
The work breakdown structure for this project was not included due to its non-applicability.
The responsibility assignment was not included due to its non-applicability.
The project schedule was not included due to its non-applicability.
The sensitivity analysis was not included due to its non-applicability.
Three types of risks were analyzed: cost, schedule, and technical/performance. Based on the overall risk analysis, the majority of the project is medium to high risk, and little can be done to reduce or transfer the risk without affecting the cost and/or schedule. In order to reduce cost risk, additional time was allowed for research to ensure a solid design basis. This also helped reduced the schedule and technical/performance risk. However, most of the risk must be accepted.
Table 7 depicts the costs risks to the project. The highest risk items were the results of the basic research (probability) and the conceptual design (cost impact). Although these had one component of high risk, the other component was low. The project definition and detail design and development both were medium for probability and cost impact. All of these items affect the scope, and scope creep is one of the most typical areas of cost increase. In order to mitigate the risk as much as possible, additional time was allowed for thorough research. However, not all risk can be mitigated, so some will have to be accepted.
Another risk to cost was the cost effectiveness of the entire system. In order to be competitive, the LED lighting system has to have a threshold payback time of less than three years (per the requirements.) The options selected for this cost comparison (shown in Table 8) include the EcoTech Radion Pro, one of the leading LED lighting fixtures on the market for 2013, the ATI Powermodule, a high-end combination of T-5 lights and LEDs, the author’s current metal halide system, and an upgrade to the author’s metal halide system. The LED lighting fixture for this project was temporarily dubbed the “Quasar” for ease of comparison. The Quasar had an estimated payback of only 24 months, which met the threshold, as shown in Figure 8.
Each fixture had an initial purchase price (cost) (or overall budget for the project), a quantity required for the author’s main display aquarium, an extended initial purchase price cost, the fixture wattage (ran at full power) to determine the yearly electrical cost. The maintenance cost is yearly and includes required bulb changes.
The schedule contains a fair amount of risk since nearly all items fall on the critical path (Table 9). Changes to the scope will affect the cost, as previously mentioned, but it will also have a medium schedule impact with medium probability. Of the highest concern is the build schedule. It is the item with the longest duration, has a medium probability, and it has a high schedule impact. In order to minimize the build duration, the author has contracted with several subject matter experts to provide assistance if required, but that option would increase the cost. Additionally, the software presents a medium probability of medium schedule impacts due to the non-commercial-off-the-shelf (COTS) code. If there is a mistake in the code, it is not easy to debug.
There is a significant amount of technical and performance risk to this project (Table 10) due to the limited availability of coral and other invertebrate photosynthetic research. These invertebrates vary widely in their requirements due to their geographic location (Great Barrier Reef, Indonesia, et cetera), the collection depth (light spectrum/intensity changes with depth), water turbidity (spectrum/intensity change with turbidity), species (not all species host the same symbiotic photosynthetic algae), and health history (past injury may cause the invertebrate to host the algae differently or even different species altogether.) Therefore, this item represents the highest risk due to probability and impact. Tying for the highest risk is the interface design. The LED fixture must interface with the Neptune Apex controller, which uses proprietary software. Although its software is somewhat intuitive, it is not easy to debug. The controller also controls two EcoTech MP-40 power heads, which simulate tidal effects, storms, lagoons, and other environmental conditions. The LED fixture must be able to tie in to the power heads to simulate the storms, sunrise/sunset, and other conditions simultaneously. To minimize the risk, several subject matter experts were identified that could assist, but this would influence the cost and schedule.
Change Management and Control
Change Management and Control details are not included in this blog due to their non-applicability.
Test and Evaluation Methodology
Although the majority of test and evaluation will be performed in Phase 3 (outside the scope of the thesis project), the basic test methodology was developed. The objective of the test phase is to determine the optimal light spectrum(s) and intensity for invertebrate growth and color. The response variables (output) are mass growth and color. Mass growth is most easily determined subjectively (rating the growth rate) but is not accurate. Weighing the coral to determine mass growth is more accurate (objective test), but it involves removing the coral and killing it to weigh the skeletal mass. Initially a combination of the two tests was planned for use; however, a more representative measurement is to monitor the uptake of skeletal-building minerals. Expected consumption resembles an exponential growth rate over time, but sudden changes in growth rates should be noticeable.
Color is also subjective and slightly objective to an extent. The basic coral pigments are visible to the human eye, so a rating system will be used to determine the quality of pigmentation display. Fluorescence is harder to observe with the human eye. Coral fluorescent proteins are excited at varying wavelengths and re-emit at others. The most fluorescent activity occurs when the coral is excited with blue light. However, very little fluorescence is noted without the use of a yellow filter to block the extraneous blue light from the viewer’s eyes. In addition to coral growth and coloration, the photosynthetically active radiation (PAR), Lux, and Kelvin of the lighting beside each tested coral will be noted where possible.
Controllable factors for the experiments are extensive. Many factors affect coral growth and coloration (and many are likely still unknown). The most obvious controllable factors are the individual LED color strings (including royal blue, neutral white, cool blue, cyan/turquoise, red, and violet). The lowest setting for each color is 10% of the possible intensity (however, 15% is used to prevent inadvertent system shutoff due to low voltage). Due to coral acclimation to high light intensities, the highest intensity setting is limited to 50%. The number of combinations possible from these six control factors and eight levels is 262,144 (Table 11). Each test will take approximately two weeks to complete. At 262,144 combinations, a complete experiment would require several thousand years, which is not feasible.
Instead, royal blue (RB) and neutral white (NW) are two separate factors. All other colors are combined into a third factor. The low level intensity is 15% and the high level is 30%. A high intensity of 50% is not used due to the acclimation time require from 15% to 50%. This results in a test matrix of only eight combinations (Table 12). With two weeks per test, a 16-week test run is feasible.
Other conditions are controllable, but they will be held as constant as possible for the duration of the tests (Table 13). A small target window will be allowed with most variables (Table 14). Corals cannot grow without alkalinity, magnesium, and calcium; therefore, a low range would inhibit their growth and negatively affect the test results if not quickly corrected. Although having other variables outside the “target” range is not ideal and can negatively affect the growth rates and/or coloration, their effects should be negligible compared to the effects of low alkalinity/magnesium/calcium.
Food input can cause spikes in excess nutrient levels, so the amount and type of food is regulated. Each day a 0.1 gram sheet of seaweed (brown, red, or green) is provided. The nutritional analysis is 10.04% moisture, 46.32% protein, 3.9% fat, 1.9% fiber, and 9.17% ash. A 0.1 gram spoon of Brine Shrimp Direct Golden Pearl Reef & Larval Fish Diet (800-1000 micron size) is also fed, with a nutritional analysis of 50% crude protein, 18% crude fat, 15% crude ash, 2% fiber, 2% phosphorous, 15,000 IU/kg Vitamin A, 3,000 IU/kg Vitamin D, 350 ppm Vitamin E, 1,000 ppm Vitamin C, and 12 mg/g DHA. Its ingredients include hydrolyzed fish protein, crustacean meal, yeast, egg, soy, casein, fish oil, lecithin, cholesterol, vitamins and minerals, and antioxidants. Once each week, a 4.0-gram New Era tab is provided for grazing. Its nutritional analysis is 32% crude protein, 18% ash, 11% crude fat, 1.5% crude fiber, 23% moisture, 15,000 IU/kg Vitamin A, 2,000 IU/kg Vitamin D3, and 200 IU/kg Vitamin E. The ingredients are fish meal, dried seaweed meal, cornstarch, fish oil, krill, squid, mussel, shrimp, choline chloride, Vitamin A acetate, Cholecalciferol, dl-Alpha-Tocopherol acetate, Calcium-L-Ascorbyl-2-Monophosphate, Zinc Sulfate, Manganese Sulfate, Nicotinamide, Inositol, Copper Sulfate, d-Calcium Pantothenate, Ferrous Sulfate, Riboflavin, Calcium Iodate, Thiamine Mononitrate, Pyridoxine Hydrochloride, Menadione Sodium Bisulfite Complex, Folic Acid, Vitamin B12 Supplement, and Biotin.
Calcium, alkalinity, and magnesium are tested weekly with a Red Sea Reef Foundation Pro test kit (Figure 9). It is accurate to 5 ppm for calcium, 0.14 dKH (0.05 meq/l) for alkalinity, and 20 ppm for magnesium. (Red Sea, 2013) Calcium reagent lot numbers are 111 (part A), 191 (part B), and 331 (part C). The alkalinity reagent lot number is 321. The magnesium reagent lot numbers are 101 and 283 (part A), 303 and 531 (part B), and 141 and 953 (part C). (Bridges, 2013) These elements will be maintained within their target range using homemade additives. The recipe is located in APPENDIX D: Supplement Recipes. (Holmes-Farley, 2006)
Temperature is measured nearly continuously with the Neptune AquaController Apex Temperature Probe (permanently calibrated and National Institute of Standards and Technology (NIST) certified) (Figure 10). The temperature is regulated with the Apex through a 300 W and a 400 W heater. If the temperature range exceeds predetermined limits, an audible warning will sound and the author will receive a warning text message and email (Figure 11).
The Neptune Apex pH Probe will nearly continuously monitor pH and is accurate to 0.1 (Figure 12). However, it must be calibrated every six months with calibration fluid at 7.0 and 10.0 pH. If the measured pH range exceeds predetermined limits, an audible warning will sound and the author will receive a warning text message and email (Figure 11).
Salinity (measured in parts per thousand, ppt) or the specific gravity (measured as a ratio to pure water) is monitored during weekly testing with a Premium Blue Refractometer, RHS-10ATC (Figure 13). This refractometer is accurate to ±0.001 on the specific gravity scale. Calibration occurs weekly with zero total dissolved solids (TDS) reverse osmosis deionized (RO/DI) water. Proper salinity is maintained through the replacement of evaporated water with RO/DI water regulated with an automatic top-off system (homemade). Water changes (saltwater replacement) are performed weekly with RO/DI water and SeaChem Reef Salt mixed to the target salinity range.
Nitrate and phosphate testing occurs with a handheld Hanna Instruments meter (Figure 14). The Hanna Instruments 713 Phosphate Low Range Meter has a range of 0.00 to 2.50 parts per million (ppm), a resolution of 0.01 ppm, and is accurate to ±0.04 ppm or ±4% of the reading at 25 degrees Celsius. The reagent used is from lot number H059 and has an expiration date of 05/2016. Phosphate and nitrate levels are maintained through water changes, active skimming with a Reef Octopus Extreme 250, activated carbon in a media reactor, and granular ferric oxide in a media reactor. The activated carbon used is Bulk Reef Supply Rox 0.8, dosed at 20 tablespoons and changed monthly. The granular ferric oxide is Bulk Reef Supply, dosed at 25 tablespoons and changed monthly.
Free ammonia (NH3) and ammonium (NH4) are tested with the Seneye Reef Monitor (Figure 15). The range for NH3 is 0.000 to 0.500 ppm, and the resolution is 0.001 ppm. The accuracy is 0.005 ppm. Unfortunately, the accuracy of NH4 is not provided by the manufacturer. Regardless, NH4 is of little concern in a reef aquarium as it is a non-toxic salt, and its impacts are negated by higher pH systems.
Uncontrolled variables include the room lighting, sunlight exposure, unexpected livestock death and decay, and equipment failures. Artificial room lighting should have little to no impact on coral health or growth due to the low intensity. Sunlight exposure through uncovered windows could cause some coral bleaching due to excessive red light, or it could even wash out the LED intensity and cause the corals to turn brown (excessive population of the symbiotic algae, zooxanthallae.) Therefore, all windows in the room are covered with blinds that will be kept closed for the duration of the testing. Livestock death and subsequent decay could spike the ammonia, nitrite, nitrate, and phosphate levels. Depending on the level, the coral could bleach or turn brown. Therefore, ammonia and nitrite will be regularly tested (but uncontrolled) with a Seneye Reef Monitor (calibrated monthly). If they are detected, the test will be terminated. Lastly, equipment failures could result in diminished coral coloration or growth. For instance, if a skimmer pump quit, the corals may brown due to excessive nutrients. Or, if the chemical dosing pumps failed, the corals may stop growing due to limited minerals. Therefore, the equipment will be checked daily, and the chemicals will be tested weekly.
A three-factor design of experiment (DOE) results in the below matrix of eight tests (Table 15). The results of the tests (coral growth and coloration) will determine the main effects and interactions between the factors.
The views and opinions expressed or implied in this paper are those of the author and should not be construed as carrying the official sanction of the University of Dayton, the Engineering Department, or of individuals/groups mentioned in this paper. This project is for informational purposes only, and it should not be used replace proper electrical engineering training before attempting such a project. Any projects arising from this paper are at the reader’s own risk. Additionally, this report and analysis shall not be used for commercial and/or profit without the author’s explicit written permission and any permission required from the University of Dayton.