Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An investigation of in-vessel composting process control strategies Fraser, Bud S. 1997

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1997-0529.pdf [ 3.68MB ]
JSON: 831-1.0058599.json
JSON-LD: 831-1.0058599-ld.json
RDF/XML (Pretty): 831-1.0058599-rdf.xml
RDF/JSON: 831-1.0058599-rdf.json
Turtle: 831-1.0058599-turtle.txt
N-Triples: 831-1.0058599-rdf-ntriples.txt
Original Record: 831-1.0058599-source.json
Full Text

Full Text

A N INVESTIGATION OF IN-VESSEL COMPOSTING PROCESS CONTROL STRATEGIES by Bud S. Fraser B. Sc. (E. E.), The University of Manitoba, 1988 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES Bio-Resource Engineering Program We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA 30 September 1997 ©Bud S. Fraser, 1997 In present ing this thesis in partial fulf i lment of the requ i rements for an a d v a n c e d d e g r e e at the University of British C o l u m b i a , I agree that the Library shall make it freely available for re ference and study. I further agree that p e r m i s s i o n for extensive c o p y i n g of this thesis for scholar ly p u r p o s e s may b e granted by the h e a d of m y depar tment or by his o r her representat ives. It is u n d e r s t o o d that c o p y i n g o r pub l ica t ion of this thesis for f inancial gain shall not b e a l l o w e d wi thout m y writ ten p e r m i s s i o n . D e p a r t m e n t of CfamhcJ < W Bft-fksovfCt T h e Universi ty of British C o l u m b i a V a n c o u v e r , C a n a d a Date Qgf. R . D E - 6 (2/88) Abstract Composting provides environmental benefits including an alternative method for organic waste disposal, as well as improving soil fertility. However poor compost quality and odour emission are often significant problems in the composting industry. Composting process control can potentially help reduce both of these problems. In spite of the recent development of a number of process control strategies, very few direct comparisons have been made between these, particularly in terms of compost quality and odour emission. To help address this need, a series of experiments was conducted to evaluate the effects of several in-vessel process control strategies on compost quality and odour emission. The strategies tested focussed on aeration control, which in turn affects temperature, oxygen, and moisture, the most important conditions for microbial activity. Fixed aeration (Beltsville method), temperature feedback (Rutgers method), oxygen feedback, and combined temperature/oxygen feedback were tested. A modified algorithm based on temperature, called linear temperature feedback, was also developed and tested. Results showed that while many quality parameters such as C:N ratio and organic matter loss were similar between process control methods, compost oxygen was maintained more consistently using oxygen feedback or linear temperature feedback methods; these methods may also provide greater nitrification and lower phytotoxicity. Linear temperature feedback is preferable to oxygen feedback in that it does not require oxygen sensors to operate. Odorous gas (including ammonia) mass emission rates were typically found to increase with higher aeration rates, such as those used to limit temperature, though the gas concentration was lower. For maximum retention of nitrogen, adequate supply of readily biodegradable carbon in the feedstock is vital. i i Table of Contents Abstract i i List of Figures v List of Tables vi Acknowledgments y i i i 1 Introduction 1 1.1 Background 1 1.2 Problem Statement 2 1.3 Objectives 3 2 Literature Review 5 2.1 Compost Quality 5 Standards 5 Additional Quality Criteria and Discussions 7 2.2 Odour U 2.3 Process Control 12 Beltsville Process 13 Rutgers Process 13 Temperature and Oxygen Feedback 14 Oxygen Feedback 15 "Leeds" System 15 Air Recirculation 16 Future Needs 16 3 Materials and Methods 17 3.1 Experimental Design 17 3.2 Analytical Methods 20 Compost Measurements 20 Gas and Odour Measurements 21 Data Analysis 22 3.3 System Design 23 Control Algorithms 23 Aeration Rate Determination 24 3.4 Feedstock Selection and Processing 28 4 Results and Discussion 29 4.1 Run 1 29 4.2 Run 2 33 4.3 Run 3 40 ii i 4.4 Run 4 45 4.5 Run 5 50 4.6 Discussion 57 5 Conclusions and Recommendations 65 5.1 Conclusions 65 5.2 Recommendations for Further Research 67 Bibliography 68 Appendix A Sample Spreadsheets 73 Appendix B Aeration Calculations 76 Appendix C Source Code 78 Oxygen Feedback 79 Linear Temperature Feedback 83 Appendix D Aeration vs. Temperature Modelling Data 88 iv List of Figures Figure 3-1. Compost Reactor Process Control Schematic Diagram 25 Figure 3-2. Temperature/Oxygen Feedback Flowchart 26 Figure 3-3. Linear Temperature Feedback Flowchart 27 Figure 4-1. Run 1 Compost Temperature vs. Time 31 Figure 4-2. Run 2 Compost Temperature vs. Time 36 Figure 4-3. Run 2 Peak Odorous Gas Concentrations 39 Figure 4-4. Run 2 Peak Odorous Gas Emission Rates 39 Figure 4-5. Run 3 Compost Temperature vs. Time 43 Figure 4-6. Run 4 Compost Temperature vs. Time 47 Figure 4-7. Run 4 Compost Oxygen Concentration vs. Time 48 Figure 4-8. Run 5 Compost Temperature vs. Time 52 Figure 4-9. Run 5 Compost Oxygen Concentration 53 Figure 4-10. Organic Matter Loss vs. Change in C:N Ratio 63 Figure D - l . Run 4 Aeration and Oxygen Uptake vs. Temperature - Increasing Trend 88 Figure D-2. Run 4 Aeration and Oxygen Uptake vs. Temperature - Decreasing Trend 88 v List of Tables Table 4-1. Run 1 Summary of Setup 31 Table 4-2. Run 1 Feedstock Composition 31 Table 4-3. Run 1 Compost Quality Parameters 32 Table 4-4. Run 1 Odorous Gas Characteristics 33 Table 4-5. Run 2 Summary of Setup 35 Table 4-6. Run 2 Feedstock Composition 36 Table 4-7. Run 2 Compost Quality Parameters 37 Table 4-8. Run 2 Odorous Gas Characteristics 38 Table 4-9. Run 3 Summary of Setup 41 Table 4-10. Run 3 Feedstock Composition 42 Table 4-11. Run 3 Compost Quality Parameters 44 Table 4-12. Run 3 Nitrogen Balance 45 Table 4-13. Run 4 Summary of Setup 46 Table 4-14. Run 4 Feedstock Composition 47 Table 4-15. Run 4 Compost Quality Parameters 49 Table 4-16. Run 4 Nitrogen Balance 50 Table 4-17. Run 5 Summary of Experimental Setup 52 Table 4-18. Run 5 Feedstock Composition 52 Table 4-19. Run 5 Compost Quality Parameters 54 Table 4-20. Run 5 Nitrogen Balance 55 Table 4-21. Run 5 Odorous Gas Characteristics 55 Table 4-22. Run 5 Cress Seed Germination 56 vi Table 4-23. Process Control Comparison Matrix 64 Table A - l . Nitrogen Measurement Spreadsheet 73 Table A-2. Moisture, VS, TOC, and O M Measurement Spreadsheet 74 Table A-3. Mass Balance Measurement Spreadsheet 75 vii Acknowledgments I would like to thank first and foremost Dr. Anthony Lau for his support, encouragement, and dedication as my thesis supervisor. Thanks also to Dr. Victor Lo, and Prof. Jim Atwater for their participation as committee members. During the course of my work, much assistance was provided by other faculty, staff and students of Bio-Resource Engineering, in particular Dr. Ping Liao, Mr. Monty Bruce and Mr. Raymond Wong. Funding was generously provided in part from the B. C. Science Council, the Natural Sciences and Engineering Research Council of Canada, and the Solid Waste Association of North America. viii 1 Introduction 1.1 Background One of many definitions of composting is "the biological decomposition and stabilisation of organic substrates, under conditions that allow development of thermophilic temperatures as a result of biologically produced heat, to produce a final product that is stable, free of pathogens and plant seeds, and can be beneficially applied to land." (Haug, 1993). A variety of different methods may be used for composting, ranging widely in complexity. The most traditional and simplest method is the windrow or static pile, consisting of a heap of compostable material that is aerated by natural diffusion and convection of air, and may be turned or mixed occasionally. To help speed the process, air may be forced into the pile, increasing the availability of oxygen to microbes and resulting in an increased rate of degradation. This process is known as aerated static pile. For more intensive composting, an in-vessel process may be used, where compost is placed in an open or closed container, with active ventilation (usually forced air) and typically some type of agitation or mixing. Process control is most commonly applied to the in-vessel system and sometimes the aerated static pile. In general the purpose of process control is to optimise the environmental conditions for the desired microbes, providing maximum degradation of substrate while maintaining compost quality and minimising odour (Person and Shayya, 1994; Finstein and Hogan, 1993). Two significant environmental problems to which composting is being suggested and applied as a solution are solid waste disposal, and agricultural soil depletion and degradation. In an effort to improve solid waste disposal and reduce the landfill waste stream, a program of the Canadian Council of Ministers of the Environment (CCME) has set the objective of reducing 1 landfilling of waste by 50% by the year 2000 (Ag. Canada, 1995). Since organic materials make up a significant portion of municipal solid waste (MSW), for example close to a third in the Greater Vancouver Regional District (Waste Program Consortium, 1992), alternative methods for disposal of organic materials such as composting would help achieve this objective. In addition, under intensive farming, agricultural soil quality becomes degraded mainly due to loss of organic matter, resulting in compaction, poor water infiltration and retention, oxygen depletion, and depleted fertility (Ag. Canada, 1995). Compost can improve soil by providing organic matter, improving soil structure and moisture retention, reducing plant pathogens, and providing nutrients (Haug, 1993). Characteristics of compost or the composting process that are considered beneficial or desirable can vary according to the application. This means that different criteria for measuring compost quality may be applied, depending on the situation. For example, compost can be used as an agricultural fertilizer due to high nutrient content. However according to Swiss regulations, when compost is used as a soil improver (not fertilizer) to improve structure, water holding capacity and organic content, maximum limits are imposed on some nutrients including nitrogen, in order to prevent eutrophication of surface waters from runoff (Brinkman et al., 1997). Quality requirements for products used for soil reclamation may be less stringent in other respects such as stability compared to products used for potting soil (Brodie et al., 1993). 1.2 Problem Statement There are a number of technical challenges facing mid- and large-scale composting operations, including odour management, quality control, and efficiency. Odour nuisance is a widespread problem in waste composting (Finstein and Hogan, 1993). In Ontario, some commercial compost products tested have been found to be immature and phytotoxic (Mathur, 2 1996). Improving process control provides a potential method to address these technical challenges. Accordingly, a number of process control strategies have been developed; however some strategies appear to have limitations. For example, the temperature feedback control (Rutgers) method, which is currently widely used, does not supply optimal oxygen at all times (Elwell et al., 1996). Also, though individual process control strategies have been tested, there is little information available directly comparing the different strategies in terms of their effects on compost quality and odour. Such information would help producers and system designers select the most appropriate methods for their needs. Though most composting odour is treatable, reduction of source odour can reduce sizing and cost of treatment facilities and can have other benefits such as improved worker health and safety. 1.3 Objectives Differences in compost quality and odour emission were expected based on the hypotheses that more precise supply of oxygen will maintain more consistently aerobic conditions in the compost, which will increase substrate degradation, affect other quality parameters, and reduce emission of odorous gases; and that exhaust gas recirculation will reduce the emission rate of odorous gases from the system. Also, more precise oxygen control may be achievable without the need for direct oxygen monitoring equipment. The specific objectives of this thesis were as follows: 1. To develop a set of criteria for measuring compost quality. 2. To review existing process control strategies and determine the most relevant process parameters relating to compost quality and odour. 3. To develop modified process control algorithms. 3 4. To formulate a suitable feedstock recipe, including the appropriate substrate and amendments. 5. To measure and compare the compost quality and odour emissions for the most common process control strategy, as well as several modified process control strategies. 4 2 Literature Review 2.1 Compost Quality Standards In Canada, three organizations are responsible for compost and composting standards: Agriculture and Agri-Food Canada (AAFC), provincial and territorial governments, and the Standards Council of Canada (SCC) through the Bureau de normalisation du Quebec (BNQ) (BNQ, 1996). The Canadian Council of Ministers of the Environment (CCME) assists to coordinate provincial and territorial initiatives. C C M E guidelines for Category A compost (can be used for all applications including agriculture and horticulture) include the following requirements: - carbon:nitrogen (C:N) ratio <, 25, - oxygen uptake rate < 150 mg 0 2/kg volatile solids per hour, and - germination of Cress (Lepidium sativum) seeds and of radish (Raphanus sativus) seeds in compost must be greater than 90 percent of the germination rate of the control sample, and growth rate of plants grown in a compost/soil mixture must not differ more than 50% from a control sample. Other requirements can be substituted, including minimum curing time, maximum reheating upon standing, or minimum organic matter reduction. In British Columbia, the B. C. Regulation 334/93 requires the following conditions to be satisfied for municipal solid waste compost: Pathogen Destruction Temperatures of at least 55 °C must be maintained during the in-vessel composting process for a minimum of three days to ensure adequate destruction of potentially harmful pathogens. 5 Organic Content Reduction To ensure sufficient degradation of organic components so that the product is stable, organic content of the feedstock must be reduced by 60% according to the following formula: % Reduction = [ 1 - (%A(100-%B))/(%B(100-%A))] x 100 (2-1) where % A = % organic matter content of dry matter after decomposition, and %B = % organic matter content of dry matter before decomposition. In this report, organic matter loss is based on mass balance and percent organic matter measured: % O M loss = (Organic mass B - Organic Mass A)/Organic Mass B [ (%B * Mass B) - (%A * Mass A) ]/(%B * Mass B) (2-2) where organic matter content and mass are dry solids basis. Because O M loss is based on mass balance rather than % O M alone, there could be some difference between the % O M loss (Eq. 2-2) and the %Reduction in O M (Eq. 2-1). Eq. 2-1 assumes that inorganic matter is conserved; however in practice some is lost to leachate. Since dry solids and organic matter as a percentage of dry solids were both measured, the most appropriate method here is Equation 2-2. Self-Reheating Self-reheating of the compost must be less than 20 °C above ambient after being aerated and put in a pile 2 m wide by 1.5 m high for 3 days. This indicates that biological activity is low, and therefore the compost is relatively stable and mature. Heavy Metals Content While this measurement is important for compost quality and applicability, compost metal content depends mainly on the feedstock, rather than the composting process control. Therefore, 6 this rneasurement is not applicable for assessment of process control strategies. pH Level The compost pH level must be 5 to 8. This test ensures the compost is not excessively acidic or basic, due to feedstock imbalances or process problems. Acidic compost often indicates anaerobic conditions during the process. Additional Quality Criteria and Discussions In addition to the tests listed in the national and B.C. Regulations, quality criteria and discussions of various quality aspects have been identified from other sources as follows. Phytotoxicity Phytotoxin levels should be low enough to promote the germination of seeds and growth of plants. Plant bio-assays, such as seed germination and plant growth, can be done to indicate the presence of plant toxins as well as nutrient value in the compost product (Garcia et al., 1990; Jimenez and Garcia, 1989). Phytotoxicity of compost is primarily due to organic acids, particularly acetic acid, and possibly butyric or other acids (Manios et al., 1989). Ammonium and phenols can also be factors (Garcia et al., 1990). Typically, phytotoxicity will disappear after sufficient curing, as it is associated with intermediate stages of composting (Zucconi et al., 1981). However, anaerobic conditions can lead to increased formation of volatile organic acids (Miller, 1993). It is •therefore reasonable to assume that process control could influence the levels of organic acids and therefore phytotoxicity and the length of time required to remove it. Nitrogen Content Retention of nutrients required for plant growth, in particular nitrogen, is important for compost applied in agriculture; losses of nitrogen during composting are often high, and can easily amount to 50% (Witter and Lopez-Real, 1987). Nearly all nitrogen lost is due to ammonia 7 volatilisation in the early stages of composting (Witter and Lopez-Real, 1988). The main factors in high ammonia production are nitrogen-rich feedstock and high decomposition rate; volatilisation is increased by high pH, high temperature, and high ammonia concentration. Addition of carbon-rich materials, particularly those that are more readily degradable and/or have been ground or chopped to increase surface area, tends to reduce ammonia loss by immobilisation of nitrogen into microbiafbiomass (Witter and Lopez-Real, 1987). In some cases, nitrogen has been found to increase during composting (Hansen et al., 1991; Bemal and Navarro, 1996); the cause of these increases was not confirmed but in the case of Bemal and Navarro, biological nitrogen fixation was suggested as a possible cause. In another case a partial recovery from nitrogen loss during composting was attributed to biological nitrogen fixation (de Bertoldi et al., 1982). In that experiment, nitrogen fixation was found to occur in the latter stages (after 20 days) of composting, during mesophilic temperature periods, as it is inhibited by temperatures over 40°C and high ammonia levels. Many nitrogen-fixing bacteria were isolated from the compost, including Azomonas (aerobic), Klebsiella, Enterobacter, and Bacillus (facultative), and Clostridium (anaerobic). High temperatures (over 40 °C) and high levels of ammonia have also been shown to inhibit nitrification (Bemal et al., 1996; de Bertoldi et al., 1982), so that nitrification generally occurs in the curing phase (Witter and Lopez-Real, 1987). Carbon to Nitrogen Ratio A carbon to nitrogen (C/N) ratio of less than 20 is often used as an indicator of compost maturity. However, the actual C/N ratio will depend largely on the C/N ratio of the starting material as well as the proportion of biodegradable carbon, so a final C/N ratio of less than 20 is a necessary, but not sufficient, condition for compost maturity (Jimenez and Garcia, 1989). 8 Nitrate to Ammonia Ratio During the composting process, ammonia production decreases in the latter stages, and production of nitrates increases. The presence of nitrates and absence of ammonia have therefore been used as an indicator of compost maturity. In some cases, the presence of nitrates alone has been used as an indicator (Finstein and Miller, 1985). Physical Characteristics The compost should have a dark, humus-like appearance, and an "earthy" odour which is not unpleasant. Lack of recognisable ammonia smell indicates that the ammonia emissions have reached a low or near zero level, which is expected for mature, stable compost. (Jimenez and Garcia, 1989). Humus Content Humification has been studied in many cases, and a number of different tests developed to indicate compost maturity. Various types of humus extractions can be analysed for carbon, however this has not proved a good indicator for all types of compost due to variability with the nature of raw materials and variability of extractability with humus age or presence of certain minerals and metals. Fractionation of extracts into humic and fulvic acids may also be done, but results have not been reliable for all compost types. N M R spectroscopy has also been used with limited success. These humus extraction tests tend to be expensive and time consuming to carry out, and the results are often difficult to interpret (Ciavatta et al., 1993; Mathur, 1992). Pathogen Reduction For application of compost to agriculture or horticulture, pathogen content is an especially important issue. Though other types of pathogen reduction are possible, heat inactivation is the most widely used and accepted method. In spite of the simplicity of the B. C. Regulation on this 9 issue, the effectiveness of pathogen reduction depends on the type of pathogens concerned, and is a function of both time and temperature. A high temperature for a short duration can be equivalent to a lower temperature for a longer duration. Most human pathogens found in sewage sludge or municipal waste can be destroyed within one hour at temperatures above 55 °C; however some such as Mycobacterium tuberculosis, spore-forming bacteria, or viruses may require higher temperatures and longer durations (Haug, 1993). Fungal plant pathogens in general can survive at higher temperatures than most animal pathogens (Golueke, 1982); some plant pathogens have shown a marked resistance to high temperatures and composting conditions, such as Tobacco Mosaic Virus (TMV), and some Fusarium species (Brinkman et al., 1997). While sterilization at high temperature (eg., wet sterilization at 110°C) can destroy all pathogens, this is normally not required or desirable. However, it is reasonable to conclude that a temperature regime significantly higher than the B. C. Regulations require during composting can decrease the survival rate of pathogens, particularly those affecting plants. Feedstock Biodegradability Feedstock characteristics have a major impact on the nature and quality of the compost product. For example, choice of bulking agent affects the composition of the compost product; degradation of ligno-cellulosic material is very slow (Martin et al., 1993). Leaves may provide a carbon source with high biodegradability. Organic carbon biodegradability can be estimated based on lignin content (Kayhanian and Tchobanoglous, 1992) using the following formula: Biodegradability factor = 0.83 - 0.028 x %lignin (2-3) Leaves have a lignin content of approximately 10% (Misksche and Yasuda, 1977), whereas many woods have lignin contents ranging from 19 to 33% (Crawford, 1981). Red Alder for example has a lignin content of approximately 25%. These values give a biodegradability factor 10 of 0.55 for leaves compared to 0.13 for Alder wood. In addition, lignin in leaves is concentrated in the structural "veins"( Solomon et al., 1993), leaving other largely cellulosic parts of the leaf unprotected and more readily degradable, which may increase the biodegradability of the leaf further. However it is important to note that some leaf types such as Oak contain compounds that inhibit microbial activity. Fish waste and vegetable waste have biodegradability factors of approximately 0.80 to 0.83 (Kayhanian, 1992; Haug, 1993). Empirical data on feedstock biodegradability appears to be limited primarily to results from anaerobic digestion studies. To ensure sufficient free air space in the compost, some bulking agent is usually required, and wood chips are most commonly used for this purpose (Haug, 1993). Due to the low biodegradability of wood chips, most will remain after composting and may be screened out for reuse. Smaller wood particles such as shavings or shredded wood may also be used, but may be more difficult to screen out. 2.2 Odour The most significant sources of odour from solid wastes are probably fatty acids and sulphur compounds, in particular methyl mercaptan (MM) and dimethyl sulfide (DMS) (Kamiya and Ose, 1984). DMS has been used as an index for compost odour potential (Often and Dennison, 1997). Sulphur gases are characteristic of anaerobic decomposition (Haug, 1993); however it has been established that odorous gases, including M M and DMS, are produced under aerobic composting conditions (Banwart and Bremner, 1976; Kuroda et al., 1996). Anaerobic microbes in anaerobic micro-environments of aerobic composts may be responsible for a significant portion of the overall metabolic activity (Atkinson et al., 1996). It has also been reported that increased process aeration resulted in a reduction of concentration but an increase in emission 11 (total mass) of odorous compounds; conversely, decreased aeration resulted in increased concentration but decreased emission rate (Walker, 1993). 2.3 Process Control The most critical conditions for achieving maximum microbial activity by maintaining conditions favourable for micro-organisms are temperature, oxygen, and moisture (Finstein et al., 1992; Miller, 1993). The most common method of controlling the process temperature is through aeration. Aeration controls temperature primarily through evaporative heat loss (Finstein and Hogan, 1993). Oxygen level may be maintained through natural air convection and diffusion (windrow configuration), agitation (agitated windrow or in-vessel configuration), or forced aeration (aerated static pile or in-vessel configuration). In general, only forced aeration provides sufficient aeration to prevent oxygen limitation. (Finstein and Hogan, 1993). A number of studies, summarised by Haug (1993), have investigated oxygen consumption. These studies have found that oxygen consumption increases with increasing temperature. Some data show temperature optima of 40 to 60 °C after which oxygen consumption decreases; other data show increasing consumption rates up to about 70 °C. It was suggested (Regan and Jeris, 1973) that the differences in temperature optima may be in part due to differences in the cellulosic content of the substrate. The data suggest that temperature could be a useful indicator on which to model oxygen consumption and therefore control oxygen level in the compost. Minimum desirable interstitial oxygen concentration has been reported as 12 to 14% (Miller, 1993) to prevent a decrease in microbial activity. Moisture content has been typically either set initially by feed conditioning (de Bertoldi et al., 1988; MacGregor, 1981), controlled by manual water addition (Finstein et al., 1983), or maintained by open loop control of water addition (Tseng et al., 1995). Successful examples of 12 automatic control of moisture content are not common, perhaps due to the relative difficulty of in-situ moisture measurements. Process control has been successfully used to reduce odours from composting operations (Murray et al., 1991). Traditionally, the main mechanism used to achieve this is maintenance of aerobic conditions through aeration; however maintenance of aerobic conditions does not yield odour-free compost, due in part to odorous intermediate products of the composting process (Haug, 1993). Agitation is a common element of composting systems, particularly for in-vessel composting. However, no literature was found specifically addressing different agitation strategies in terms of process control. Finstein et al. (1992) stated "...the central issue in composting process design is temperature control". Accordingly, the most commonly utilised process control strategies focus on aeration as the process variable, as described in the following sections. Beltsville Process In the Beltsville process, aeration of a static pile is provided by a timer which turns a blower on and off on a predetermined schedule (Finstein et al., 1983). The process control objective is to maintain oxygen at between 5 and 15%. This strategy, while being relatively simple, proved to have some limitations (see Rutgers process). Rutgers Process In the Rutgers process, the control objective is to maximise microbial activity by regulating temperature via controlled ventilation of the compost (Finstein et al.,1983). In the initial stage, rapid temperature ascent is encouraged by timer-controlled aeration; once temperature reaches 13 the desired range, it is limited to a maximum value through temperature feedback-controlled ventilation. In a comparison between Beltsville and Rutgers strategies, (Finstein et al., 1992), the materials balance showed that the Rutgers process resulted in decomposition of 4.7 times more volatile solids in % of the time, compared to the Beltsville process. It was concluded that the lower degradation rate in the Beltsville pile was mainly due to limitation of microbial activity due to high temperatures. Temperature and Oxygen Feedback A modified Rutgers strategy was tested that combines temperature feedback control with oxygen feedback control (Vallini et al., 1989). In this control strategy, when the temperature exceeds the setpoint, the blower works continuously. At temperatures lower than the setpoint, the blower is controlled by the oxygen level. When the oxygen level drops below the oxygen setpoint, the blower works continuously. When the oxygen level is above the oxygen setpoint, the blower operates on a low duty cycle timer. In this experiment the thermophilic (greater than 45 °C) stage was completed within 21 days, and the C:N ratio, pH, and phytotoxicity tests all indicated that the final compost was of satisfactory quality and maturity. Finstein and Hogan (1993) stated however that oxygen feedback is not necessary. During temperature feedback control, a high level of oxygen is a result of temperature-induced aeration. During come-up (increasing temperature) and come-down (decreasing temperature) stages when the temperature does not demand aeration, maintenance of oxygen level can be achieved using an adequate fixed baseline schedule. A study using only baseline ventilation during come-up in conjunction with feedback temperature control demonstrated this (Vallini and Pera, 1989). Also, equipment currently available for measuring oxygen concentration tends to be more expensive 14 and less reliable than equipment for measuring temperature, due to the higher complexity. In spite of these arguments against combined temperature and oxygen feedback control, no literature could be found directly comparing this method with others such as Rutgers. Oxygen Feedback Use of oxygen feedback aeration control was tested in the composting of MSW in a "closed" reactor (de Bertoldi et al., 1988). Forced aeration was provided, with the aeration rate controlled by oxygen level feedback, using a setpoint of 15%. During the experiment, temperatures peaked at close to 70°C, although temperatures above 60°C only occurred in the upper 10 cm of the compost. Compost quality was judged to be good and pathogen reduction was judged to be adequate, however organic matter (dry) reduction was only 14%. Processing time using this scheme is "medium": longer than Rutgers, shorter than Beltsville. There may be energy and cost savings, since aeration is done only when oxygen demand is high, and there is no agitation. However oxygen monitoring equipment is required, and since there is no temperature feedback, the potential for excessively high temperatures remains. "Leeds" System Leeds University (Leton and Stentiford, 1990) developed a temperature feedback aeration control system that uses temperature measurement to implement a simple model of the process. When the temperature is above the setpoint, a fan is turned on for a fixed period. However when the temperature is below the setpoint, the current temperature reading is compared with the previous temperature reading. Depending on whether the temperature is increasing or decreasing, the fan is turned on for a longer or shorter period of time. This allows the aeration rate to be adjusted to three different rates for different portions of the process: the initial come-15 up portion, the high temperature portion, and the final come-down portion. It is not clear what benefit, i f any, is derived from this system compared to other systems. A i r Recirculation An alternate aeration strategy using air recirculation was tested in a composting tunnel, using temperature feedback aeration control (Miller et al., 1990). Dry matter loss, physical consistency, pathogen content, odour production, nutrients, and composting time were all judged to be satisfactory. Air recirculation design allows separate control of temperature and oxygen, by using a heat exchanger and variable fresh air intake respectively. A potential benefit offered by this system is improvement of the manageability of odour emissions by reducing the volume of exhaust air that must be handled and treated. Using recirculation, emission rates of some odorous gases (reduced sulphur gases and ammonia) may be lower, based on the longer residence time of exhaust gases in the compost (Mathur, 1994). Future Needs Several areas have been identified in publications where further research is needed, including the use of higher "low" fan settings during the earliest composting phase to avoid anaerobic conditions (Elwell et al., 1996), investigating the effects of different aeration schemes on odour production (Mathur, 1994) and further investigating the effect of oxygen, temperature, and aeration variations (Finstein and Hogan, 1993). These areas are the focus of the testing in this thesis. 16 3 Materials and Methods 3.1 Experimental Design In order to evaluate the effects of different process control strategies, two or three in-vessel compost reactors were run in parallel, generally using the same feedstock but different process control methods. Compost quality and odour measurement results were then analysed and compared between the reactors. The active (thermophilic) phase, where the most process control is applied, was the focus of measurements; however in later runs, the curing phase was also included in quality analysis. Once the active phase was completed and temperature returned to near ambient levels, the compost was removed from the reactors and placed in bags for 1 to 2 months. The curing bags used were 25C capacity, Rayon fabric, allowing air diffusion through the material, and once filled were placed side by side on a shelf or floor. Based on the research found from the literature review, the most commonly used and accepted in-vessel process control strategy is the temperature feedback (Rutgers) aeration control method. Therefore, this method was used as a baseline for comparison of process control strategies: in each experiment, one compost reactor was configured for temperature feedback. A series of laboratory experiments was conducted over a Wi year period. First, several experiments were used to design and test a laboratory scale in-vessel composting system, with computer monitoring. Once this was completed, more extensive data collection began (run 1). Runs 1 and 2 were done as part of a project in conjunction with a commercial composting facility; selection of feedstock materials was done largely in accordance with those used at the commercial facility. For run 1, two reactors were set up with temperature feedback (Rutgers type) aeration, with temperature setpoints of 58°C and 68°C in reactors 1 and 2 respectively. Feedstock was a mixture of fish, vegetables, paper sludge, and other materials. Compost samples 17 were taken at the beginning of the experiment, and at the end of the active phase, when compost temperature returned to near ambient. A curing phase was not studied during this experiment. In run 2, two variations of temperature feedback were compared with fixed rate aeration. Reactors 1 and 2 were set up for temperature feedback, however reactor 1 was to maintain more ideal conditions in terms of oxygen concentration. This was done by adjusting the instantaneous flow during the process according to estimated oxygen demand; in other words a type of manual oxygen feedback control was used. A third reactor was constructed and configured for fixed rate aeration using a lower flow rate than the other reactors, in order to create a low oxygen concentration during the process. The temperature setpoint was 58°C for reactors 1 and 2. Feedstock was a mixture of fish, vegetables, and other materials similar to those used in Run 1. Following the active phase, the compost was removed to fabric bags, and stored indoors for 50 days for curing. Feedstock was sampled at beginning of the experiment and the compost was sampled at the completion of the active and curing phases. Since the extent of biodegradation as indicated by loss of organic matter was still unsatisfactory in run 2, it was of interest to investigate whether feedstock composition is an important factor. Therefore, several different feedstocks were compared in run 3. In addition to the fish and vegetable waste, mixtures of leaves, shredded yard and wood waste, and Alder pulpwood chips were used in different proportions. Reactor 1 used leaves and wood chips; reactor 2 used a combination of leaves, wood chips and yard/wood waste; and reactor 3 used yard/wood waste. A l l three reactors were configured for temperature feedback aeration control. The compost was mixed by hand on day 7, to improve porosity and break up any large particles. Following the active phase, compost was screened manually using a LA" screen, then cured for 43 days in fabric bags. 18 For runs 4 and 5, oxygen monitoring capability was added to the system, allowing periodic oxygen measurements to be taken and recorded automatically, and used for oxygen feedback aeration control. Run 4 compared temperature feedback, oxygen feedback, and combined temperature/oxygen feedback. The desired oxygen range for reactors 2 and 3 was set at 14 to 17%. A range rather than a setpoint was used, to avoid the need for very frequent sampling and adjustment of aeration rate. The control algorithm for reactor 3 is shown in Figure 3-2; the reactor 2 algorithm is the same with the exception that the temperature feedback portion is not included. Also, temperature setpoints were increased to 60 °C to assist in satisfying the minimum pathogen reduction requirement, since temperatures often fluctuate several degrees during temperature feedback. Based on the results from the previous run, leaves were added to the feedstock. However in this case more than half of the leaves used were shredded, to further increase the availability of carbon. The remainder of the leaves were left whole, as these appeared to add significant porosity to the feedstock. Red Alder wood chips were also included to provide structure. The compost was mixed manually on day 7. After the active phase, the compost was stored in synthetic bags for 42 days . The compost was not screened so that a larger curing mass could be maintained, and water was added after 2 weeks of curing to maintain the moisture content near 50 %. For Run 5, the oxygen feedback algorithm was modified to work using only temperature inputs; this method was called linear temperature feedback (refer to System Design section) and implemented in reactor 3. Reactor 1 was configured for fixed rate aeration, and reactor 2 temperature feedback. This setup was intended to test the linear temperature feedback model as well as collect more data on using fixed aeration as was done in Run 2. In this experiment, a feedstock similar to that of Run 4 was desired. Due to limited supply of some wood chips and 19 leaves, no whole leaves were used and shredded yard/wood waste was used in place of some of the Alder wood chips. The compost was mixed manually on day 9. Following the active phase it was stored for 41 days in fabric bags. More detail on the setup and feedstock composition for each run is shown in summary tables in the Results and Discussion section. Ambient air temperature (used for cooling compost) was typically 24±2°C. 3.2 Analytical Methods Compost Measurements With the exception of run 1, compost was generally sampled at three points: at the beginning of the active phase (feedstock), following the active phase, and following the curing phase. Sampling techniques were done so that they are as representative as possible (Johnson et al., 1993). Starting in run 2, 5 grab samples were taken and mixed in a container, then this composite sample was sampled for analysis. Storage time of samples was minimised, and all samples were stored in airtight bags in a cold room at 4 °C. Compost temperature was measured using type T thermocouples sealed in stainless steel probes, connected to the PC monitoring system. Gravimetric moisture content was determined by evaporation at 105 °C for 24 hours (Amer. Soc. Agron., 1982). pH was measured using distilled water dilution as per standard methods (Amer. Soc. Agron., 1982). Volatile solids content was estimated using ignition at 550 °C for 15 minutes (APHA, 1985), and organic matter and total organic carbon (TOC) were estimated using ignition for a further 60 minutes at 660 °C (Gallardo et al., 1987; Golueke, 1977). Starting at run 4, total-organic carbon (TOC) was measured by combustion at 680° C and C 0 2 measurement, using a Shimadzu TOC-5050 Total Organic Carbon Analyser with SSM-5000 Solid Sampling Module. 20 Water soluble ammonium and nitrite/nitrate were measured using water extraction and the Technicon Autoanalyser II industrial method. Total nitrogen was measured using ignition at 950°C in the Leco FP228 Nitrogen Determinator, with the exception that in run 1, T K N acid digestion method from a commercial laboratory (Norwest) was used. Gas and Odour Measurements Compost off-gas was collected in 50 Tedlar sampling bags prior to analysis to ensure adequate sample size. Gastec detector tubes and hand sampling pump were then used for ammonia and methyl mercaptan measurements on the gas samples, while Drager detector tubes were used for dimethyl sulfide, with the exception of run 1 where gas chromatography was used. Accuracy tolerance of the Gastec tubes is given as ±25%, and standard deviation of the Drager tubes is given as ±20%. Interstitial compost oxygen concentration was measured using a galvanic type ESD6100 compost oxygen probe/controller. A Cress (Lepidium sativum) seed germination test was used for determining compost phytotoxicity (Zucconi et al., 1981) for run 5 cured material. Seeds were placed on filter paper in a petri dish, then 3 m# of compost water extract (10 g wet solids to 90 mf distilled water) was added and the dish covered. The dishes were incubated at 27°C for 67 hours. This differs from the published procedure in that more extract was used than 1 mH (to help avoid premature drying), and the incubation period was much longer than 24 hours, since the seeds used were found to germinate much more slowly. Six replicates (dishes) were used per compost sample, with 10 seeds per dish. A set of controls of equal size using only distilled water was included. After the incubation period, germinated seeds were counted in all dishes. Germination rate as 21 well as germination index (germination rate divided by germination rate of control) were calculated. Data Analysis Based on the compost quality parameters found in the literature review, the following parameters were analysed: - carbon to nitrogen (C:N) ratio, - organic matter loss, - ammonium to nitrate ratio, - emission rate of odorous gases, - duration of pathogen reduction (>55 °C) period, and - nitrogen retention. Gas emission rates were calculated by multiplying the measured concentration by the airflow rate and converting to mass flow rate. Measurements were entered into spreadsheets using Microsoft Excel. Statistical analysis consisted of calculation of arithmetic means, as well as standard deviations as a measure of variability of results. The number of replicates used was typically 3 to 5, though in run 1 some were only 2; this should be kept in mind when looking at the standard deviations. Some sample spreadsheets are included in Appendix A. For calculation of C:N and ammonium to nitrate ratios, the arithmetic means were used and standard deviation of the result is estimated based on error calculated as per A S H R A E methods (ASHRAE, 1985). 22 3.3 System Design The in-vessel laboratory composters were constructed from 120 litre plastic barrels, with a perforated acrylic plenum which allows upward forced airflow, a removable lid with seal, and headspace exhaust and sampling ports. Insulated sides with an integral controlled heater reduce the heat loss from the compost mass. Temperature monitoring was provided by removable thermocouple probes. A PC-based process control system monitored and logged data from the thermocouples and oxygen probe, while controlling the aeration solenoid valves and diaphragm air pumps. The process control system is based on a I B M PC-compatible 486 platform, running Labtech Control V.9.02 software. Interfaces to external measuring and control devices are provided through Advantech data acquisition and control cards. A schematic diagram of the process control system is shown in Figure 3-1. Control Algorithms For oxygen feedback and linear temperature feedback functions, programs were written in ANSI C and integrated into Labtech Control, since the more complex algorithms could not be implemented using the built-in features of Labtech Control. The source code for this software is included in Appendix C. Flowcharts describing operation of these algorithms are shown in Figures 3-2 and 3-3. The linear temperature feedback algorithm was designed so that oxygen could be maintained within a desirable range (approximately 14 to 17%), without requiring oxygen feedback. Based on Run 4 data showing average aeration duty cycle and temperature for reactors 2 and 3 (oxygen feedback), a simple first order model was created, relating aeration duty cycle to temperature. 23 Aeration duty cycle and temperature data showed a coefficient of correlation of 0.81 for reactor 2 and 0.83 for reactor 3. The model was based on two data regions: the increasing temperature region and the decreasing temperature region. Linear regression was used on each of these data sets to determine an equation relating aeration duty cycle to temperature. Appendix D shows graphs of of the oxygen model data from reactor 3. Once the equations were determined, a C language computer program was written and compiled using Microsoft Visual C++ V. 1.52 compiler, and integrated into Labtech Control to perform the aeration control function. Source code for the program is listed in Appendix C. To save money, the linear temperature feedback algorithm could also be implemented using a much simpler computer system and a standalone C program, if desired, rather than Labtech Control. Aeration Rate Determination Required aeration rate for supply of oxygen and cooling was determined by estimating the oxygen uptake from empirical data and calculating heat output of the compost (Haug, 1993). The calculations are shown in Appendix B. A value of 7 to 8 {/min was used for cooling based on the capacity of the diaphragm pumps and the calculated value. When reviewing the aeration rates used for each run in Section 4, which have been converted to {/min»kg DS units (litres per minute per kilogram dry solids), differences are often visible from one run to the next as well as between one reactor to the next. This is mainly due to the unpredictability of the actual airflow rate achievable by the air pumps, except where airflow was changed intentionally as noted. After adding the compost to the reactors, the maximum achievable airflow may vary between reactors, or from the previous run, due to varying compost porosity; the rate was normally decreased such that it was approximately equal in all reactors. Slightly different masses of compost in each 24 reactor would then lead to unequal aeration rates in terms of flowrate per unit dry solids. These differences were not expected to have a significant impact on results. P C I/O Board Ambient Thermocouple Air Pump R Relay <8>—(gy-Solenoid Valve T o Biof i l ter Oxygen Sensor ®-Airflow ,_£ Meter Compost Reactor Jacket Heater t'. Jacket Thermo-couple V -Thermo-couple Probe Figure 3-1. Compost Reactor Process Control Schematic Diagram. 25 A I R O N O R O F F 1 MIN . A C C O R D I N G T O C U R R E N T D C S T A R T ( D C = D E F A U L T I V A L U E ) M E A S U R E C O M P O S T T E M P E R A T U R E T E M P > S E T P O I N T ? , A I R O N Y E S " 1 M I N . 1 MINUTE CYCLE M E A S U R E O X Y G E N C O N C E N T R A T I O N D E C R E A S E D C 30 MINUTE CYCLE Figure 3-2. Temperature/Oxygen Feedback Flowchart. 26 START (DOOEFAULT ; VALUE) I AIR ON OR OFF 1 MIN. ACCORDING TO CURRENT DC MEASURE COMPOST TEMPERATURE w AIR ON YES*" 1 MIN. 1 MINUTE CYCLE CALCULATE 30 MIN RUNNING AVERAGE TEMP CALCULATE NEW DC BASED ON EQ.1 (INCREASING TEMP) CALCULATE NEW DC BASED ON EQ.2 (DECREASING TEMP) I 30 MINUTE CYCLE Figure 3-3. Linear Temperature Feedback Flowchart. Note: "DC" indicates duty cycle. 27 3.4 Feedstock Selection and Processing As mentioned earlier, feedstock composition was initially designed to duplicate, as closely as possible, that of a commercial composting facility that handles M S W as well as industrial and agricultural wastes. The feedstock included wood/yard waste, paper (mechanical pulp waste or "sludge"), manure, animal rendering waste (in this case fish), and vegetable waste. Starting in run 3, the feedstock composition was reworked, maintaining components of vegetable and fish waste and wood/yard waste, while removing the paper sludge component. Fish, vegetables and leaves were shredded using a yard shredder, while woody materials were typically picked up pre-shredded from the composting facility. Materials were mixed in buckets using an auger as well as by hand. 28 4 Results and Discussion Figures and tables are shown at the end of each subsection. For each experiment run, a short tabular summary of the setup for that run as well as feedstock composition is provided. 4.1 Run 1 Figure 4-1 shows the temperature profiles for both reactors. Once the temperature setpoint of 57°C for reactor 1 was reached, high rate aeration started and the temperature was maintained near this value for 2 days. The temperature then decreased to near ambient after 7 days. Baseline aeration resumed after the temperature dropped below the setpoint. However, reactor 2 temperature never reached the setpoint of 67 °C, and therefore only the baseline aeration was maintained. The temperature for reactor 2 peaked near 60 °C, was maintained between 55 °C and 60 °C for 3 days, then decreased to near ambient after 8 days. Compost quality parameters are summarized in Table 4-3. The total organic carbon (TOC) of the initial feedstocks showed a marked difference between the two reactors, though the same feedstock materials were used. Small standard deviations for these measurements suggest that the differences were due to unequal feedstock composition in the two reactors, or problems with the sampling. Feedstocks for the two reactors were mixed separately in two containers, and therefore each was sampled and analysed separately. Note that errors in TOC would affect other parameters, particularly C:N ratio and organic matter loss. The C:N ratio decreased to between 22 and 23 for both reactors. Organic matter loss was low, in the range 7 -11% for both reactors. Ammonium-nitrogen increased, and nitrate-nitrogen decreased slightly for both reactors, giving similar ammonium:nitrate ratios for both cases. pH increased to near neutral from slightly acidic for both cases. 29 Table 4-4 shows that the oxygen concentration in reactor 2 dropped to only 2.7% by day 2 and remained below 5% until day 4. This low oxygen condition was due to the low aeration rate in reactor 2. At the same time, concentration of methyl mercaptan (MM) and dimethyl sulfide (DMS) was higher than in reactor 1. However, the emission rates of these gases were similar from the two reactors; for DMS, the emission rate from reactor 2 was lower. The peak ammonia gas concentration was similar for both reactors, near 450 ppm, though the trends over time were very different; reactor 1 peaked sharply early (day 3), while reactor 2 slowly increased to peak on day 9. The early peak of reactor 1 occurs during the highest temperature period, when oxygen supply is highest due to temperature feedback cooling. This would allow a high metabolism rate of the nitrogen-containing proteins, releasing ammonia. In reactor 2 however, metabolism rate may be more limited by the low oxygen early in the process; also pH could potentially be lower during the low-oxygen period (Miller, 1993). High pH tends to encourage ammonia loss (Witter and Lopez-Real, 1987). The ammonia gas emission rates were also very different between reactors: reactor 1 peaked at a value 4 times higher than reactor 2 on day 3, but decreased quickly to a much lower level by day 4, while reactor 2 slowly increased to its peak value on day 9. Higher airflow contributed to the higher peak for reactor 1. 30 Reactor Process Control Control Parameters 1 Temperature Feedback Setpoint = 57 °C Aeration 0.17 C/min«kg DS Baseline 1.38 C/min«kg DS Maximum 2 Temperature Feedback Setpoint = 67 °C Aeration 0.18 0/min«kg DS Baseline 1.44 #/min«kg DS Maximum Table 4-1. Run 1 Summary of Setup. Contribution Reactor Material (% wet mass) 1,2 Shredded Herring Fish 14 Shredded Vegetables 27 Horse Manure 5 Paper Sludge 18 Shredded Wood Waste 21 Compost Recycle 10 Carbon (Ash) 5 Total 100 Table 4-2. Run 1 Feedstock Composition. 31 Parameter Moisture (% Wet Basis) Day: Reactor 1 2 Mean 63.1 64.4 SD 3.1 3.1 15 Mean 56.5 55.1 SD 6.0 2.5 Total Organic Carbon 1 43.7 0.7 43.7 1.6 (LOI, % Dry Basis) 2 47.1 1.3 43.6 2.6 Total Nitrogen 1 1.6 0.6 1.9 0.7 (% Dry Basis) 2 1.0 0.3 2.0 0.9 Ammonium-Nitrogen 1 0.122 0.017 0.751 0.020 (% Dry Basis) 2 0.136 0.005 0.752 0.066 Nitrate-Nitrogen 1 0.019 0.003 0.009 0.001 (% Dry Basis) 2 0.020 0.002 0.009 0.001 C:N Ratio 27.3 47.1 10.3 14.2 23.0 21.8 8.5 19.9 Ammonium:Nitrate Ratio 1 6.4 1.4 2 6.8 0.7 83.4 83.5 9.5 11.8 pH 5.8 5.7 0.0 0.0 7.8 7.8 0.0 0.0 Organic Matter Loss 1 (% Dry Basis, From 2 Day 1) 7.7 10.9 Table 4-3. Run 1 Compost Quality Parameters. Note: "-" indicates not measured. 32 Parameter Hour: Day: Reactor 27 2 50 3 75 4 126 6 196 9 286 12 Temperature (°C) 1 58 57 54 36 26 26 2 56 59 57 45 28 26 Oxygen(%) 1 13.5 15.5 9.2 17.6 19.5 20.2 2 2.7 4.0 4.6 13.8 19.5 20.3 Ammonia Concentration (ppm) 1 20 440 180 270 270 250 2 ND 140 300 400 450 400 Ammonia Emission Rate (ng/min) 1 66 1327 129 194 194 180 2 ND 101 215 287 323 287 Methyl Mercaptan Concentration (ppm) 1 6.6 - - - - -2 >31.1 - - - - -Methyl Mercaptan Emission Rate (|ig/min) 1 61.7 - - - - -2 63.2 - - - - -Dimethyl Sulfide Concentration (ppm) 1 2.7 - - - -2 6.7 - - - - -Dimethyl Sulfide Emission Rate (ng/min) 1 32.6 - _ - -2 17.6 - - - - -Table 4-4. Run 1 Odorous Gas Characteristics. Note: "ND" indicates none detected. 4.2 Run 2 For this and subsequent runs the feedstock was mixed thoroughly, sampled, and divided amongst the three reactors. Therefore the initial compost quality parameters are the same for all three reactors. By day 2, as temperature increased and oxygen decreased in reactor 1, the aeration rate was increased manually; unfortunately the maximum low rate that could be achieved due to pump limitations was only 1.10/min average, rather than the 2.4 H/min desired, based on oxygen uptake 33 estimates to maintain oxygen near 15%. Later after 6 days, temperature decreased to 30 °C and the reactor 1 low aeration rate was decreased to 0.13 Q/min average. Temperature profiles for reactors 1 and 2 (Figure 4-2) were similar, maintaining over 55 °C for only 2 to 2.5 days. Reactor 3 however achieved much higher temperatures, with a peak of approximately 78 °C, and over 55 °C for 7 days. It is clear that the high temperatures of reactor 3 are mainly due to lack of a high airflow rate for cooling. Reactors 1 and 2 therefore did not achieve the minimum requirement for pathogen reduction of three days at 55 °C or greater, while reactor 3 far exceeded it. Table 4-7 summarizes the compost quality parameters for this run. Moisture trends are similar for all reactors, with limited drying during the active and curing phases. TOC decreased slightly during each phase as well. Total nitrogen decreased significantly, particularly in reactor 2. The nitrogen decrease resulted in an increase in C:N ratio during the active phase, contrary to the decrease in C:N considered desirable (Jimenez and Garcia, 1989). In spite of a reasonable starting C N ratio (20-30), much of the carbon in the feedstock may not be readily biodegradable, due to the paper sludge and wood waste (see discussion for next run). This conclusion is supported by the low organic matter loss observed. The lack of readily degradable carbon would also contribute to ammonia volatilization and loss of nitrogen (Witter and Lopez-Real, 1988). Reactor 3 had the highest nitrogen at the end of the active phase, which is expected due to the lower aeration rate and ammonia emission rate (Table 4-8). It is not clear why reactor 2 lost more nitrogen than reactor 1 - it may be due to higher emission rates during the latter two thirds of the run; however there is not enough data to determine this. Nevertheless the differences in nitrogen are relatively small and may not be indicative of differences in the process control. 34 Arnmonium-nitrogen results were similar for all reactors, however reactor 1 showed a much higher nitrate-nitrogen level (560 ppm) than 2 and 3 (40 to 50 ppm) after curing. This resulted in a lower ammonium:nitrate ratio for reactor 1. pH for reactor 1 was lower than the other two reactors, after both the active and curing phases, and closer to a desirable neutral value. Odorous gas characteristics are shown in Tables 4-8. Compost oxygen concentration sharply decreased to below 3% during day 1, but rebounded to near 15% on day 2, once the cooling high rate aeration started. The decrease in oxygen near the beginning of the process is characteristic of temperature feedback aeration control (Elwell et al., 1996), as discussed in the Literature Review section. Measurements of M M and DMS gases showed a clear relationship with the process control, as illustrated in Table 4-8 and Figures 4-3 and 4-4. Peak gas concentrations were highest for reactor 3, with the lowest aeration; however emission rates were higher in reactors 1 and 2, due to the higher aeration rates. Peak ammonia emission rates were also higher in reactors 1 and 2, however the concentration was also much higher in these reactors, resulting in an even larger difference in emission rate compared to reactor 3. These results support the theory of greater ammonia generation under aerobic conditions. Reactor Process Control Temperature Feedback with Manual Oxygen Feedback Temperature Feedback Fixed Rate Table 4-5. Run 2 Summary of Setup. Control Parameters Setpoint = 58 °C Aeration 0.18 d/min'kg DS Baseline 1.28 e/min«kg DS Maximum Setpoint = 58 °C Aeration 0.18 5/min'kg DS Baseline 1.28 0/min-kg DS Maximum Aeration 0.11 <!/min«kg DS 35 Content Reactor Material (% wet mass) 1,2,3 Shredded Herring Fish 14 Shredded Vegetables 26 Horse Manure 5 Paper Sludge 19 Shredded Wood Waste 22 Compost Recycle 10 Carbon (ash) 4 Total 100 Table 4-6. Run 2 Feedstock Composition. 10 0 - - J — I I I—J I L—J I * • ' • ' • ' L__| | • ' ' I ' • • • • • » 0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 Time (h) Figure 4-2. Run 2 Compost Temperature vs. Time. 36 Day: 1 15 65 Parameter Reactor Mean SD Mean SD Mean SD Moisture 1 64.8 0.5 60.8 0.2 51.4 1.8 (% Wet Basis) 2 64.8 0.5 61.0 0.8 48.4 1.2 3 64.8 0.5 60.9 0.7 48.4 1.2 Total Organic Carbon 1 45.0 0.9 42.7 0.6 41.2 1.2 (LOI,% Dry Basis) 2 45.0 0.9 42.8 1.6 41.0 0.1 3 45.0 0.9 43.0 1.7 42.5 3.4 Total Nitrogen) 1 2.1 0.2 1.8 0.0 1.6 0.2 (% Dry Basis) 2 2.1 0.2 1.6 0.1 1.6 0.1 3 2.1 0.2 1-9 0.1 1.8 0.1 Ammonium-Nitrogen 1 0.456 0.051 0.515 0.027 0.022 0.002 (% Dry Basis) 2 0.456 0.051 0.529 0.045 0.025 0.000 3 0.456 0.051 0.540 0.006 0.024 0.001 Nitrate-Nitrogen 1 0.015 0.001 0.038 0.001 0.056 0.013 (% Dry Basis) 2 0.015 0.001 0.039 0.001 0.005 0.002 3 0.015 0.001 0.038 0.001 0.004 0.002 C:N Ratio 1 21.4 2.1 23.7 0.3 25.8 3.3 2 21.4 2.1 26.8 1.9 . 25.6 1.6 3 21.4 2.1 22.6 1.5 23.6 2.3 Ammonium:Nitrate Ratio 1 30.4 4.0 13.6 0.8 0.4 0.1 2 30.4 4.0 13.6 1.2 5.0 2.0 3 30.4 4.0 14.2 0.4 6.0 3.0 pH 1 5.3 0.0 7.9 0.0 7.3 0.0 2 5.3 0.0 8.2 0.0 7.6 0.0 3 5.3 0.0 8.2 0.0 7.6 0.0 Organic Matter Loss 1 - 16.7 (% Dry Basis, From 2 - 14.6 -Day 1) 3 - 15.7 -Table 4-7. Run 2 Compost Quality Parameters. Note. TOC, TN sampled at 75 days instead of 64 37 Hour: 0 21 42 142 310 Day: 1 1 2 5 12 Parameter Reactor Temperature (°C) 1 18.2 47.4 57.3 30.7 27.1 2 20.1 47.4 57.8 28.9 25.9 3 18.6 47.8 77.5 65.7 27.6 Oxygen(%) 1 20.9 2.4 16.5 17.0 14.1 2 20.9 1.5 14.7 18 20.1 3 20.9 0.9 1.5 9.7 20.1 Ammonia Concentration (ppm) 1 ND ND 920 138 40 2 ND ND 840 325 350 3 ND ND 50 125 33 Ammonia Emission Rate (ng/min) 1 ND ND 4624 13 4 2 ND ND 2533 233 251 3 ND ND 22 323 90 Methyl Mercaptan Concentration (ppm) 1 ND ND 5 ND ND 2 ND ND 12 ND ND 3 ND ND 33 2 ND Methyl Mercaptan Emission Rate ((ig/min) 1 ND ND 71 ND ND 2 ND ND 102 ND ND 3 ND ND 40 2 ND Dimethyl Sulfide Concentration (ppm) 1 10 4 28 1 1 2 10 4 24 1 1 3 10 3 125 3 1 Dimethyl Sulfide Emission Rate (ng/min) 1 26 12 514 ND ND 2 26 10 264 3 2 3 16 5 197 5 1 Table 4-8. Run 2 Odorous Gas Characteristics. 38 140 120 E a a. c o '•5 2 * : 60 c a> o c o o 100 80 40 20 Reactor No. • D imethyl Su l f ide IS Methyl M e r c a p t a n Figure 4-3. Run 2 Peak Odorous Gas Concentrations. —. 600 c E "35 500 E 2 O) 400 2 0 1 or c o U> (A E LU 300 200 100 U Dimethyl Sul f ide B Methyl M e r c a p t a n Reactor No. Figure 4-4. Run 2 Peak Odorous Gas Emission Rates. 39 4.3 Run 3 Reactors 2 and 3 heated to 60°C by the end of day 2, and remained at 55 to 60°C for 3 days, whereas reactor 1 did not reach 60°C until day 4, but remained above 55°C for 4.5 days (Figure 4-5). The longer high temperature phase of reactor 1 is presumably due to the increased proportion of readily degradable carbon; however an explanation for the slower initial heating is less clear. One possibility is the presence of an inhibitory substance in the leaves. When the compost was mixed on day 7, reactors 2 and 3 reheated 10 to 15°C over the next 1 to 2 days. A l l reactors satisfied the minimum requirement for pathogen reduction. In terms of quality parameters, significant differences were observed between reactors, as shown in Table 4-11. The most apparent difference was the increase of nitrogen for reactor 1; along with the decrease in TOC, this resulted in a decrease in C:N ratio from 23.2 to 11.3, while in the other cases the C:N ratio stayed near constant or increased slightly. During the active phase, moisture increased in reactor 1, while in reactors 2 and 3 moisture decreased. The decrease in C:N ratio as well as the much higher organic matter loss indicate a higher degree of degradation in reactor 1; this may also account for the increased moisture and higher ammonium-nitrogen concentration, since water and ammonia are both produced during the degradation of organic material (Haug, 1993). Since nitrate-nitrogen concentration in reactor 1 was also higher at the end of the active phase, ammonium:nitrate ratios were similar in all reactors. Table 4-12 shows the nitrogen balance in detail. Reactor 1 showed a net increase of near 24% in total nitrogen mass during the active phase, compared to losses in the other reactors, with maximum loss in reactor 3. The reason for a net gain in nitrogen is not clear; it could be due to nitrogen fixation, though much of the fixed nitrogen would have to have been converted 40 to organic form (e.g., microbial biomass), since the higher ammonium- and nitrate-nitrogen can only account for a portion of the increase. Some other factor related to the increased leaf content could potentially have increased nitrogen-fixing activity. These results, along with the higher degradation, suggest that reactor 1 contains more readily degradable carbon in the form of leaves, which helps to immobilize nitrogen. On the other hand, reactor 2 and more so reactor 3 contain less readily degradable carbon which allows more nitrogen to be lost. Due to screening, the mass of compost cured was significantly reduced. The amount of cured material was much less from reactor 1 than reactors 2 and 3, since it had the largest amount of leaves, many of which were screened out along with some attached compost. This small mass resulted in excessive drying during curing, particularly from the reactor 1 material. Note that the larger than usual differences in aeration rate between reactors is due mainly to the exact dry solids content of each feedstock not being measured until after the reactors were loaded with material. Reactor Process Control Control Parameters Feedstock Amendment 1 Temperature Feedback Setpoint = 58 °C Aeration 0.28 C/min«kg DS Baseline 1.41 <!/min«kg DS Maximum Leaves Wood Chips 2 Temperature Feedback Setpoint = 58 °C Aeration 0.27 C/min«kg DS Baseline 1.37 <?/min«kg DS Maximum Leaves Yard Waste Wood Chips 3 Temperature Feedback Setpoint = 58 °C Aeration 0.23 <!/min«kg DS Baseline 1.13 (!/min»kg DS Maximum Yard waste Table 4-9. Run 3 Summary of Setup. 41 Content Content TOC TN Moisture C:N (% Wet (% Dry (% Dry (%Dry (% Wet Ratio Reactor Material Basis) Basis) Basis) Basis) Basis) 1 Shredded Vegetables 42.3 11.5 47.5 1.68 89.3 28.2 Mixed Shredded Fish 21.5 19.6 46.0 6.66 64.1 6.9 Leaves 12.1 16.6 51.9 1.23 45.7 42.1 Shredded yard/wood waste 0.0 0.0 46.3 0.65 16.8 71.3 Red Alder Pulp Chips 24.1 52.3 55.3 0.28 14.6 196.8 Total 100.0 100.0 28.2 2 Shredded Vegetables 41.4 10.7 47.5 1.68 89.3 28.2 Mixed Shredded Fish 21.1 18.3 46.0 6.66 64.1 6.9 Leaves 7.2 9.5 51.9 1.23 45.7 42.1 Shredded yard/wood waste 21.1 42.5 46.3 0.65 16.8 71.3 Red Alder Pulp Chips 9.2 19.0 55.3 0.28 14.6 196.8 Total 100.0 100.0 26.4 3 Shredded Vegetables 39.6 9.4 47.5 1.68 89.3 28.2 Mixed Shredded Fish 20.1 16.1 46.0 6.66 64.1 6.9 Leaves 0.0 0.0 51.9 1.23 45.7 42.1 Shredded yard/wood waste 40.3 74.5 46.3 0.65 16.8 71.3 Red Alder Pulp Chips 0.0 0.0 55.3 0.28 14.6 196.8 Total 100.0 100.0 27.1 Table 4-10. Run 3 Feedstock Composition. 42 80 <D ° - 30 a> I- 20 10 0 \ R eactor 1 eactor 2 eactor 3 - R \ R tl '- II /I j — IP / Mix ng Se< —I } Nol es ; 0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 384 408 Time (h) Figure 4-5. Run 3 Compost Temperature vs. Time. Notes: The region from approximately 230 to 300 hours was interpolated due to data logging failure. 43 Day: 1 18 61 Parameter Reactor Mean SD Mean SD Mean SD Moisture 1 67.2 3.4 70.9 2.5 16.8 0.6 (% Wet Basis) 2 65.9 1.3 59.7 0.8 29.9 3.4 3 59.8 2.1 55.4 2.5 30.6 4.1 Total Organic Carbon 1 51.2 1.5 47.2 2.2 46.1 1.3 (LOI, % Dry Basis) 2 47.1 1.1 47.1 0.5 46.4 2.8 3 44.8 1.9 45.0 2.7 42.6 3.1 Total Nitrogen) 1 2.2 0.2 4.2 0.2 4.3 0.1 (% Dry Basis) 2 2.5 0.3 2.6 0.4 2.9 0.1 3 2.6 0.4 1.9 0.2 2.2 0.1 Ammonium-Nitrogen 1 - 0.981 0.083 0.116 0.014 (% Dry Basis) 2 - - 0.452 0.102 0.106 0.017 3 - - 0.448 0.033 0.110 0.014 Nitrate-Nitrogen 1 - 0.019 0.001 0.009 0.001 (% Dry Basis) 2 - - 0.011 0.001 0.015 0.003 3 - - 0.010 0.000 0.008 0.002 C:N Ratio 1 23.3 2.2 11.2 0.7 10.7 0.4 2 18.8 2.3 18.1 2.8 16.0 1.1 3 17.2 2.7 23.7 2.9 19.4 1.7 Ammonium:Nitrate Ratio 1 - 51.6 5.1 12.9 2.1 2 - - 41.1 10.0 7.1 1.8 3 - - 44.8 3.3 13.8 3.9 pH 1 - 8.4 0.0 6.9 0.0 2 - - 8.4 0.0 7.4 0.0 3 - - 8.6 0.0 7.6 0.0 Organic Matter Loss 1 - 40.1 -(% Dry Basis, From 2 - 13.6 -Day 1) 3 12.5 -Table 4-11. Run 3 Compost Quality Parameters. 44 Parameter Day: Reactor 1 18 TN (% Dry Basis) 1 2.2 4.2 2 2.5 2.6 3 2.6 1.9 Mass of Solids (g) 1 4960 3220 2 5120 4420 3 6210 5410 Mass of Nitrogen (g) 1 109 135 2 128 115 3 161 103 Change in N-Mass (%) 1 _ +23.8 2 - -10.2 3 - -36.0 Table 4-12. Run 3 Nitrogen Balance. 4.4 Run 4 Due to equipment problems, the run could not be started right away and the feedstock mixture was stored for several days. As Figure 4-6 shows, temperatures in all three reactors increased very quickly to reach 60 to 70 °C within 1 day. For reactors 1 and 3 this means that the cooling aeration was not sufficient to maintain the setpoint temperature for a short period. However, the temperatures fell below 55 °C within 2 to 3 days of reaching 55 °C, failing to satisfy the minimum requirement for pathogen reduction. Table 4-15 indicates that the starting C:N ratio was actually very low, and did not change appreciably after composting. In addition, organic matter loss during the active phase was relatively low at 19 to 26%. Since the predicted C:N ratio (based on measured carbon and nitrogen of individual feedstock components) was nearly 27, these results suggest that the feedstock may have been partially degraded at startup time, due to the lengthy storage period. The fast temperature rise may have been due to an active microbial population already 45 established in the material, plus the fact that the feedstock materials, particularly fish and vegetables, were shredded finely to give a large surface area. The oxygen feedback control algorithms for reactors 2 and 3 maintained oxygen concentrations above 14% for nearly all of the active phase (Figure 4-7). In reactor 1, oxygen concentration made its characteristic decrease, but only to near 10% (compared to 15 to 17% for reactors 2 and 3); presumably this is because the setpoint temperature was reached so quickly, causing the cooling aeration to start and boost oxygen concentration. Reactor 2 showed a decrease in percent nitrogen after the active phase, but an increase after curing. However the nitrogen balance in Table 4-16 shows a net loss in nitrogen mass for all reactors in both the active and curing phases. Note that the increase in total carbon percentages in Table 4-15 is due to the relatively high nitrogen mass loss coupled with much lower carbon mass loss. C:N ratios, ammonium:nitrate ratios, and pH were similar for all three reactors, with the exception that nitrate-nitrogen was higher in reactor 3 after curing, resulting in a lower ammonium:nitrate ratio. Reactor Process Control Control Parameters 1 Temperature Feedback Setpoint = 60 °C Aeration 0.33 C/min«kg DS Baseline 1.64 0/min«kg DS Maximum 2 Oxygen Feedback Oxygen range=14 -17 % Aeration 0.33 C/min»kg DS Starting 1.64 C/min«kg DS Maximum Setpoint = 60 °C Oxygen range 14 -17 % Aeration 0.33 l!/min*kg DS Starting 1.64 l!/min»kg DS Maximum Table 4-13. Run 4 Summary of Setup. 3 Temperature/Oxygen Feedback 46 Content Content TOC TN Moisture C:N (% Wet (% Dry (%Dry (%Dry (% Wet Ratio Reactor Material Basis) Basis) Basis) Basis) Basis) 1,2,3 Shredded Vegetables 39.3 10.8 47.5 1.68 89.3 28.2 Mixed Shredded Fish 20.0 18.3 46.0 6.66 64.1 6.9 Leaves 7.5 10.4 51.9 1.23 45.7 42.1 Shredded Leaves 11.3 15.6 51.9 1.23 45.7 42.1 Red Alder Pulp Chips 18.8 40.9 55.3 0.28 14.6 196.8 Compost Recycle 3.1 4.0 50.0 2.00 50.0 25.0 Total 100.0 100.00 26.9 Table 4-14. Run 4 Feedstock Composition. Reactor 1 'Reactor 2 24 48 72 96 120 144 168 192 216 240 264 288 312 336 Time (h) Figure 4-6. Run 4 Compost Temperature vs. Time. 47 25 20 £ 15 c o> O A « 4 ^ — ~ Reactor 2 Reactor 3 0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 Time (h) Figure 4-7. Run 4 Compost Oxygen Concentration vs. Time. 48 Day: I 15 57 Parameter Reactor Mean SD Mean SD Mean SD Moisture 1 68.2 1.7 67.4 2.0 52.6 3.4 (% Wet Basis) 2 68.2 1.7 68.0 1.7 51.2 4.7 3 68.2 1.7 67.9 2.9 44.1 3.4 Total Organic Carbon 1 41.9 2.1 36.0 1.2 42.3 0.8 (% Dry Basis) 2 41.9 2.1 35.1 0.5 40.1 1.1 3 41.9 2.1 37.1 2.0 37.8 0.4 Total Nitrogen) 1 3.3 0.6 3.4 0.1 3.3 0.1 (% Dry Basis) 2 3.3 0.6 3.1 0.2 3.6 0.1 3 3.3 0.6 3.4 0.2 3.1 0.2 Ammonium-Nitrogen 1 0.442 0.111 0.483 0.036 0.046 0.005 (% Dry Basis) 2 0.442 0.111 0.479 0.027 0.060 0.017 3 0.442 0.111 0.515 0.022 0.045 0.003 Nitrate-Nitrogen 1 0.011 0.001 0.015 0.001 0.009 0.002 (% Dry Basis) 2 0.011 0.001 0.014 0.001 0.014 0.005 3 0.011 0.001 0.014 0.001 0.018 0.001 C:N Ratio 1 12.7 2.4 10.6 0.5 12.8 0.5 2 12.7 2.4 11.3 0.7 11.1 0.4 3 12.7 2.4 10.9 0.9 12.2 0.8 Ammonium: Nitrate Ratio 1 40.2 10.7 32.2 3.2 5.1 1.3 2 40.2 10.7 34.2 3.1 4.3 2.0 3 40.2 10.7 36.8 3.1 2.5 0.2 pH 1 7.4 0.0 8.8 0.0 6.9 0.0 2 7.4 0.0 8.7 0.0 7.0 0.0 3 7.4 0.0 8.8 0.0 7.0 0.0 Organic Matter Loss 1 25.6 36.3 (% Dry Basis, From 2 - 19.0 34.4 Day 1) 3 - 21.2 24.8 Table 4-15. Run 4 Compost Quality Parameters. 49 Parameter Day: Reactor 1 15 57 T N (% Dry Basis) 1 3.3 3.4 3.3 2 3.3 3.1 3.6 3 3.3 3.4 3.1 Mass of Solids (g) 1 4270 3490 2740 2 4270 3610 2950 3 4270 3640 3230 Mass of Nitrogen (g) 1 141 119 90 2 141 112 106 3 141 124 100 Change in N-Mass (%) 1 _ -15.6 -36.2 (From Day 1) 2 - -20.6 -24.8 3 - -12.1 -29.1 Table 4-16. Run 4 Nitrogen Balance. 4.5 Run 5 Temperatures in all three reactors reached the setpoint by the end of day 2, as shown in Figure 4-8. Reactor 1 continued to heat until reaching 71 °C on day 3, while reactors 2 and 3 were kept near 60 °C for 2 to 3 days by the temperature feedback aeration. A l l three reactors satisfied the minimum time and temperature requirement for pathogen reduction, though for reactor 1 it was farexceeded, and reactor 3 was only just satisfied. After cooling to 30 to 40°C, the compost was mixed at which point it reheated to near 50 °C before cooling again. Figure 4-9 shows that reactor 1 oxygen dropped to near 5% and remained much lower than reactors 2 and 3 until day 6. Reactor 2 also dropped briefly to near 7% on day 2. Reactor 3 stayed closest to the desired oxygen range of 14 to 17%. In terms of quality parameters (Table 4-19), again it is found that the starting C:N ratio was lower than expected, and increased slightly in all cases. The reason for the low starting C:N ratio is not known; it may be due to an error in feedstock preparation, or in measurement of C:N for individual feedstock components. As with the previous run, the organic matter loss is relatively 50 low for all reactors. Table 4-20 shows that there was significant nitrogen loss in all reactors, in both the active and curing phases. Nitrogen loss during the active phase was greatest for reactor 2. Reactor 1 showed higher ammonium-nitrogen concentration at the end of the active phase, while reactor 3 showed higher nitrate-nitrogen at the end of the curing phase. Drying was greatest for reactors 2 and 3, which is expected due to the higher overall airflow. Table 4-21 shows the odorous gas characteristics. Trends in gas concentrations and emission rates are not as clear as in Run 2. However, peak M M and D M S emission rates are highest from reactor 3, which had the highest aeration rate and oxygen concentration at that time. DMS concentration is highest from reactor 1, which had the lowest aeration rate and oxygen concentration. Seed germination tests (Table 4-22) on the cured compost showed a small difference between reactors, with the highest germination for reactor 3 and lowest for reactor 1. Due to large standard deviations, these results are not conclusive. However, the errors were primarily due to placement of the dishes in the incubator. Upper dishes were dried prematurely, causing lower germination rates; this probably contributed to the average germination indices being lower than the 90% requirement of the C C M E . Since the four groups of dishes were arranged symetrically in the horizontal and vertical planes, in theory the drying effect is the same for each group, so that the errors should not contribute to differences in germination rates between the reactors. 51 Reactor Process Control Control Parameters 1 Fixed Rate Aeration 0.17 e/min«kg DS 2 Temperature Setpoint = 60 °C Feedback Aeration 0.24 C/min»kg DS Baseline 1.22 5/min«kg DS Maximum 3 Linear Temperature Setpoint = 60 °C Feedback Aeration 0.24 C/min-kg DS Starting 1.19 5/min*kg DS Maximum Table 4-17. Run 5 Summary of Experimental Setup. Content Content TOC TN Moisture C:N (% Wet (% Dry (%Dry (% Dry (% Wet Ratio Reactor Material Basis) Basis) Basis) Basis) Basis) 1,2,3 Shredded Vegetables 37.9 9.1 47.5 1.66 89.3 28.2 Mixed Shredded Fish 23.5 19 46.0 6.66 64.1 6.9 Shredded Leaves 7.2 13.7 51.9 1.23 15.0 42.1 Red Alder Pulp Chips 15.7 30.0 55.3 0.29 14.6 196.8 Shredded Yard/Wood Waste 15.7 28.2 46.3 0.65 20.0 71.2 Total 100.0 100.00 26.9 Table 4-18. Run 5 Feedstock Composition. 80 70 ^ 60 0 "jT 50 1 40 CD g 30 CD I- 20 10 0 '-_ i i Reactor 1 Reactor 2 Reactor 3 — \ \ i \ 1 T '- M 1 Mixi ng \ 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 Time (h) Figure 4-8. Run 5 Compost Temperature vs. Time. 52 25 20 C 15 C a> O 5 0 \ I : Reactor 1 — R e a c t o r 2 Reactor 3 - —I 0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 Time (h) Figure 4-9. Run 5 Compost Oxygen Concentration. 53 Day: 1 15 56 Parameter Reactor Mean SD Mean SD Mean SD Moisture 1 61.5 1.3 63.8 2.2 55.2 1.4 (% Wet Basis) 2 61.5 1.3 58.9 2.1 52.0 1.5 3 61.5 1.3 58.5 0.8 48.0 0.9 Total Organic Carbon 1 41.3 3.0 41.2 2.0 41.2 3.4 (% Dry Basis) 2 41.3 3.0 38.9 2.9 42.2 2.3 3 41.3 3.0 40.7 1.4 40.8 1.3 Total Nitrogen) 1 2.6 0.5 2.6 0.1 2.2 0.1 (% Dry Basis) 2 2.6 0.5 2.2 0.1 2.3 0.1 3 2.6 0.5 2.4 0.2 2.2 0.1 Ammonium-Nitrogen 1 0.081 0.018 0.697 0.015 0.159 0.022 (% Dry Basis) 2 0.081 0.018 0.407 0.032 0.188 0.038 3 0.081 0.018 0.397 0.044 0.154 0.006 Nitrate-Nitrogen 1 0.012 0.002 0.017 0.003 0.012 0.001 (% Dry Basis) 2 0.012 0.002 0.013 0.002 0.018 0.004 3 0.012 0.002 0.015 0.002 0.027 0.001 C:N Ratio 1 15.9 3.3 15.8 1.0 18.7 1.8 2 15.9 3.3 17.7 1.5 18.3 1.3 3 15.9 3.3 17.0 1.5 18.5 1.0 Ammonium:Nitrate Ratio 1 6.8 1.9 41.0 7.3 13.3 2.1 2 6.8 1.9 31.3 5.4 10.4 3.1 3 6.8 1.9 26.5 4.6 5.7 0.3 pH 1 5.1 0.0 8.3 0.0 7.0 0.0 2 5.1 0.0 8.3 0.0 6.7 0.0 3 5.1 0.0 8.4 0.0 6.7 0.0 Organic Matter Loss 1 - 23.6 36.2 (% Dry Basis, From 2 - 23.5 35.8 Day 1) 3 - 19.5 35.1 Table 4-19. Run 5 Compost Quality Parameters. 54 Day: 1 15 56 Parameter Reactor TN (% Dry Basis) 1 2.6 2.6 2.2 2 2.6 2.2 2.3 3 2.6 2.4 2.2 Mass of Solids (g) 1 5740 4490 3920 2 5740 4600 3820 3 5860 4780 4000 Mass of Nitrogen (g) 1 149.2 116.7 86.2 2 149.2 101.2 87.9 3 152.4 114.7 88.0 Change in N-Mass (%) 1 - -21.8 -42.2 (From Day 1) 2 - -32.2 -41.1 3 - -24.7 -42.3 Table 4-20. Run 5 Nitrogen Balance. Hour: 43 91 283 Day: 2 4 12 Parameter Reactor Temperature (°C) 1 52.6 65.7 47.5 2 54.0 60.5 34.4 3 55.0 60.2 39.7 Oxygen (%) 1 8.7 9.7 17.1 2 10.3 16.8 19.8 3 14.5 17.0 17,2 Methyl Mercaptan Concentration (ppm) 1 11 3 1 2 6 1 ND 3 11 1 ND Methyl Mercaptan Emission Rate (ug/min) 1 25 7 2 2 17 12 ND 3 63 11 ND Dimethyl Sulfide Concentration (ppm) 1 60 - -2 25 - -3 42 - -Dimethyl Sulfide Emission Rate (ug/min) 1 173 - -2 92 - -3 308 - -Table 4-21. Run 5 Odorous Gas Characteristics. 55 Day: 56 Parameter Reactor Mean SD Germ. Index Germination Rate 1 0.35 0.43 0.66 2 0.4 0.35 0.75 3 0.42 0.39 0.79 Control 0.53 0.36 1.00 Table 4-22. Run 5 Cress Seed Germination. 56 4.6 Discussion Run 1 illustrated that when using temperature feedback, the overall amount of aeration (total mass or volume of air provided) can be greatly affected by the choice of temperature setpoint. The amount of aeration will depend largely on the proportion of time that the compost is above the temperature setpoint, which in turn depends on feedstock characteristics. This is a potential disadvantage for using temperature feedback, since the heating characteristics of the feedstock are not always known. In runs 2 and 5, the maximum temperature in one reactor exceeded 70°C. As described in the Literature Review section, high temperatures (over 60 °C) have been reported to decrease microbial activity (MacGregor et al., 1981; Finstein et al., 1983). While no direct evidence of this was observed, it is possible that in run 2, i f the compost samples were taken earlier, organic matter loss may have been higher for the temperature feedback reactor since the active phase appears to have completed (returned to near ambient temperatures) earlier. In this case, though higher temperature in the fixed aeration reactor would seem to suggest higher activity, heat was not being adequately removed by aeration, allowing the temperature to increase; also, the metabolic pathways used by the thermophilic micro-organism community above 60 °C may be different and result in a lower substrate utilisation rate. In practice, the earlier completion of the active phase means the compost could more easily be removed from the reactor and transferred to the curing phase at an earlier time. Based on the discussion in the Literature Review section, results from run 2 also show that the choice of aeration control method potentially affects the degree of pathogen reduction. Depending on the aeration control algorithm and the temperature setpoint (if any), the minimum temperature and time characteristic for pathogen reduction may not be satisfied, or may be far exceeded. It is quite possible that in a larger commercial system, 57 heat loss per unit mass of compost will be less than the laboratory scale system, and therefore the minimum requirement would be satisfied in all the cases tested; however the choice of process control will still greatly affect the temperature profile and potentially the occurrence of pathogens in the product. In terms of feedstock formulation, for runs 1 and 2, paper sludge and wood waste were used both as carbon sources and bulking agents; however it was clear from visual inspection as well as the low organic matter loss (7 to 11% for run 4, 15 to 17% for run 5) that they were not readily biodegradable during the active phase. Wood is well known to be resistant to breakdown during composting due mainly to lignin content; mechanical paper pulp, which was the type used here, is less degradable than chemical (Kraft) pulp but more degradable than raw wood (Haug, 1993). Both appeared relatively unchanged following the active and curing phases (compost cured for 6 weeks, run 5). The results from run 3 showed a clear improvement in biodegradability of the feedstock as well as nitrogen retention when leaves were used as the primary carbon source (OM loss 40% with more leaves, 12% with no leaves). For runs 2, 3,4, and 5, organic matter loss and change in C:N ratio were compared. Run 1 was not included due to problems in TOC measurement. As shown in Figure 4-10, a larger decrease in C:N ratio over the active phase was accompanied by a larger loss in organic matter. These results suggest a correlation between feedstock degradation (indicated by larger organic matter loss) and nitrogen retention (indicated by lower final C:N ratio). As discussed previously, both feedstock degradability and nitrogen retention are greater when more readily degradable carbon is present. In most runs, it is likely that organic matter loss and the decrease in C:N ratio would have been greater if more readily degradable carbon had been present, such as shredded leaves; 58 however it appears that the 60% reduction in organic matter required by the B. C. regulations is difficult to achieve, at least for the type of system and feedstocks used here. The results confirm that low starting C:N ratios result in rapid composting, but with increased loss of nitrogen due to ammonia volatilization (Bishop, 1983). Nitrification (producing nitrate-nitrogen) appeared to be greater when using an aeration control algorithm that supplied oxygen more in accordance with demand, in other words, some form of oxygen feedback. Though nitrate-nitrogen levels were relatively low overall (40 to 560 ppm, typically less than 200 pm), this was observed in run 2 (temperature feedback with manual oxygen feedback), run 4 (oxygen feedback), and run 5 (linear temperature feedback). The reason for these observations is not clear. Since nitrification tends to follow the thermophilic stage (Witter and Lopez-Real, 1987), the only difference between process control methods at this stage is that the oxygen feedback-type control results in a much lower aeration rate in the latter stages of the active phase. In comparing the total nitrogen (ranging from 1.6 to 4.3%) with several other composts from published papers, it was found to be in a similar range, depending largely on the compost feedstock. For example, in-vessel compost from M S W organic fractions was 1.2% (de Bertoldi et al., 1988); windrow compost from M S W organic fractions plus sludge was 1.7% (de Bertoldi et al., 1992); in-vessel compost from swine manure and peat moss was 3.9 to 4.5% (Lau et al., 1992), and aerated static pile compost from fish and sawdust was 2.3 to 3.0% (Liao et al., 1994). As a fraction of the total nitrogen, ammonium-nitrogen ranged from 1.4 to 8.2%, which is much lower than some composts recently measured (Mathur et al., 1990), where manure-based compost was 14 to 22% and shellfish waste based compost was 26 to 36%. Nitrate-nitrogen was 0.2 to 3.5%, compared to ranges of 1 to 6% and 2 to 8% for the manure- and shellfish-based 59 composts respectively. The lower ammonium may indicate more mature compost and/or differences in feedstock. In runs where measurements were made on cured compost, it can be seen that in many cases, differences in compost quality parameters, such as total nitrogen, observed at the end of the active phase may change or disappear by the end of the curing phase. Also, organic matter loss can be significant during the curing phase. This confirms the importance of the curing phase to the overall composting process. Run 5 illustrated that compost oxygen concentration can be maintained at a more constant level, in a desirable range such as 14 to 17%, using only temperature feedback and a simple first order model, compared to traditional temperature feedback. This method will help prevent the onset of anaerobic or near-anaerobic conditions early in the process that can occur with a temperature feedback system, without requiring more expensive oxygen monitoring equipment. However design of the algorithm does require knowledge of the oxygen and temperature versus time characteristics of the compost. While the seed germination test for run 5 was not conclusive, it suggested that compost phytotoxicity may be lower when using an oxygen-demand based aeration algorithm, compared to low, fixed rate aeration or temperature feedback. This could be due to maintenance of more consistent oxygen concentration and reduction of oxygen-limiting conditions that produce more organic acids (Miller, 1993). For runs 1, 2 and 5, in most cases increased aeration (for example due to temperature feedback) decreased ammonia and odorous gas concentrations but increased the emission rate of these gases. These results parallel some found previously (Walker, 1993), and confirm that maintaining high (near or above 15%) interstitial oxygen concentration by aeration does not 60 prevent the emission of odorous sulphur gases, and therefore unpleasant odour. Another potential factor in odour production is moisture; the relatively high compost moisture may have limited oxygen diffusion into particles, in spite of high interstitial oxygen concentrations, increasing the extent of anaerobic microenvironments and potentially production of sulphur gases. Depending on the type of composting system implemented, the gas concentrations and emission rates are important for system design. For example, in an enclosed agitated-bay system, high concentrations of ammonia or odorous gases can be a health threat to workers; in these cases, maintenance of aerobic conditions and minimization of gas concentrations would be desirable. On the other hand, increased emission rates of gases would require increased biofilter size where biofiltration is used to treat off-gases based on increased loading (Haug, 1993). Table 4-23 compares key compost quality and odour results as well as operating characteristics for the process control strategies tested. For the first four parameters, rankings were based on experimental results from runs 2, 4, and 5 where different strategies were directly compared. For each run, results for each process control type were ranked with 1 representing the most desirable result; differences of less than 10% were considered equal ranking. The individual rankings were then averaged for each process control type, and these averages ranked. This method allows each run to be ranked separately, taking into account differences in feedstock between runs. Cost and power consumption rankings were projected based on the literature and the laboratory equipment setup. Trends mentioned earlier with respect to odorous gas concentration and emission are visible, with fixed aeration resulting in higher concentrations and lower emission rates. No clear trend in organic matter loss was evident, but nitrogen retention appears to be better for process control that does not implement cooling aeration, which supports the theory that aeration increases the ammonia emission rate. However, the 61 differences in nitrogen retention are still relatively small. Pathogen reduction and active phase duration have an inverse relationship in that higher temperatures (above 60 °C) should improve pathogen reduction but increase the length of time for the active phase to complete (temperatures returning to near ambient). The pathogen reduction and duration data in Table 4-23 are based on the ability of the process control to limit the temperature, assuming that temperature feedback does not use an excessively high setpoint (over 60 °C). Temperature feedback (on its own or combined with oxygen or linear temperature feedback) limits the temperature the most; oxygen feedback and linear temperature feedback (with no cooling function) provides some limiting of temperature since aeration is increased along with oxygen demand, and fixed aeration provides minimal cooling and therefore minimum limiting of temperature. In terms of cost, oxygen sensing equipment is considered the most expensive and complex; in addition, cooling aeration increases the fan size and cost over fixed aeration. Linear temperature feedback may require more expensive control equipment such as a PC, compared to a timer for fixed aeration or a simple feedback controller for temperature feedback. Power consumption is greatest where cooling aeration is implemented, though this may be offset somewhat by decreased process duration. Overall, it appears that there is little to justify the use of oxygen feedback; linear temperature feedback should provide similar operation with lower equipment cost. Linear temperature feedback with cooling aeration may be most appropriate where process duration, maintenance of consistent aerobic conditions, and lower odorous gas concentrations are desired. Fixed aeration may be the most appropriate when process duration and odorous gas concentrations are not major concerns. 62 The discussions here are in the context of a batch-type, in-vessel composting process. However some of the principles and findings could be applied to continuous in-vessel processes as well. Oxygen feedback, and linear temperature feedback could in theory be applied to a continuous system, where better control of oxygen concentration is desired. Findings on feedstock composition as well as odorous gas emissions in relation to process aeration control would also apply. 45 i 40 - - • • 5 - -0 4— 0.25 0.50 0.75 1.00 1.25 1.50 C:N Final/C:N Initial Figure 4-10. Organic Matter Loss vs. Change in C:N Ratio. 63 Parameter Fixed Temp. Oxygen Temp./ Linear Aeration Feedback Feedback Oxygen Temp. Feedback Feedback Nitrogen Retention 1 2 1 3 1 Organic Matter Loss 1 1 1 2 1 Odorous Gas Concentration (peak measured sum of M M and DMS) 3 1 1 2 Odorous Gas Emission (peak measured sum of M M and DMS) 1 1 ~ 2 2 Pathogen Reduction2 1 3 2 3 2-33 Duration2 3 1 2 1 1-23 Control Equipment Cost 1 2 3 3 2 Power Consumption (based on aeration volume) 1 2 1 2 1-23 Table 4-23. Process Control Comparison Matrix. Notes: 1. "1" indicates most desirable result. 2. Based on temperature limiting ability of process control. 3. Depends whether temperature limiting (cooling aeration) is implemented. 64 5 Conclusions and Recommendations 5.1 Conclusions The objectives of this study were to determine relevant process parameters and compost quality criteria, develop modified process control strategies, compare process control strategies in terms of compost quality and odour emissions, and formulate a suitable feedstock recipe. The process parameters most important for composting process control were found to be temperature, oxygen, and moisture. The experimental results indicated that temperature and interstitial oxygen content were successfully controlled via aeration; however moisture content could have limited diffusion of oxygen within compost particles, in turn limiting the effectiveness of process control based on oxygen content. While many different criteria can be used to measure compost quality, those most widely accepted and easy to apply were adopted for use in the experiments; these were organic matter reduction, nitrogen retention, decrease in carbon to nitrogen ratio, decrease in ammonia to nitrate ratio, minimum time and temperature for pathogen reduction, pH, and seed germination. In-vessel process control strategies were compared in terms of their effect on compost quality and odour emission. In terms of quality, all the process control strategies tested produced compost that was similar in many aspects, including organic matter reduction, nitrogen retention and C:N ratio, particularly after the curing phase. However, there were differences in process control system operation and compost quality that are relevant to composting system design as follows. Temperature feedback (Rutgers type) aeration control led to very low oxygen concentrations early in the composting process, which produced higher concentrations of odorous gases. The 65 overall amount of aeration depended on the choice of temperature setpoint and heating characteristics of the compost. Linear temperature feedback aeration control, a modified temperature feedback algorithm developed for this study, provided an alternative process control method with some advantages over the Rutgers method. Both oxygen feedback and linear temperature feedback aeration control methods provided more consistent compost oxygen concentration compared to temperature feedback and may have caused an increase in nitrification. However, linear temperature feedback does not require expensive oxygen sensors, provided the oxygen uptake and temperature characteristics of the compost can be estimated. Linear temperature feedback and oxygen feedback control may also have provided compost of lower phytotoxicity than temperature feedback. The simpler fixed aeration strategy (Beltsville type) seemed to have provided increased pathogen reduction, but it also may increase the time required to produce equivalent quality compost compared to temperature-limiting processes such as temperature feedback or linear temperature feedback. Fixed aeration may also provide energy savings over a temperature-limiting process, though the savings may be offset by the increased process duration. In terms of odour emissions, maintenance of aerobic conditions in compost through process control did not remove the generation of strong odours. Higher aeration rate (as used in temperature-limiting control) and the associated higher oxygen concentration typically resulted in lower odorous gas (including ammonia) concentrations, but higher mass emission rates of these gases, which may increase the required size of odour treatment facilities. Conversely, fixed aeration typically resulted in higher concentration but lower emission rates, potentially allowing reduced sizing of odour treatment facilities. 66 Based on the results, among the processes tested, linear temperature feedback with cooling aeration is most appropriate where process duration, maintenance of consistent aerobic conditions, and lower odorous gas concentrations are desired; fixed aeration is most appropriate when process duration and odorous gas concentrations are not major concerns. Finally, feedstock composition had a significant effect on compost quality. When composting nitrogen-rich feedstocks such as those used in the experiments, providing sufficient readily biodegradable carbon was critical to conserving nitrogen and maximising organic matter reduction. Though a consistently performing, ideal feedstock recipe was not finalised, a mixture of whole and shredded dry leaves provided a suitable carbon source and bulking agent for nitrogen-rich substrates such as fish and vegetable waste. 5.2 Recommendations for Further Research Further research is recommended in the following areas: 1. The effect of exhaust gas recirculation on odorous gas emissions and compost quality, particularly nitrogen retention. Unfortunately, sufficient time and resources were not available to test the effects of air recirculation. 2. The relationship of compost nitrification and phytotoxicity with aeration strategies. 3. The effect of different curing methods and conditions on compost quality. 4. Further tests to determine the extent and timing (active phase vs. curing phase) of nitrogen fixation in compost. 5. Investigation of additional compost quality indices that provide more detailed information on fertilizer value, maturity and phytotoxicity such as humus content and organic acid content. 6 7 Bibliography Ag. Canada. 1995. A Program to Assess the Impacts and Benefits of Composted Source-Separated Solid Wastes (CSSSW) Applied to Agricultural Lands: National Agricultural Compost Trial, Technical Bulletin 1995-9E. Research Branch, Agriculture and Agri-Food Canada, Ottawa, Can. American Society of Agronomy. 1982. Methods of Soil Analysis, 2nd. Ed. Soil Science Society of America, Madison, Wis. APHA. 1985. Standard Methods for the Examination of Water and Wastewater, 16th ed. American Public Health Association, Washington, D. C. ASHRAE. 1985. Fundamentals Handbook. American Society of Heating, Refrigerating, and Air Conditioning Engineering, Inc. Atlanta, USA. Atkinson, C. F., Jones, D. D., and Gauthier, J. J. 1996. Putative Anaerobic Activity in Aerated Composts. Journal of Industrial Microbiology 16:182-188. Banwart, W. L., and Bremner, J. M . 1976. Evolution of Volatile Sulfur Compounds from Soils Treated With Sulfur-Containing Organic Materials. Soil Biology and Biochemistry 8:439-443. Bernal, M . P., Navarro, A. F., Roig, A. , Cegarra, J., and Garcia, D. 1996. Carbon and Nitrogen Transformation During Composting of Sweet Sorghum Bagasse. Biology and Fertility of Soils 22:141-148. Bishop, P. L., and Godfrey, C. 1983. Nitrogen Transformations During Sludge Composting. Biocycle July/August 1983, 34-39. Brinkman, J., Baltissen, T., and Hamelers, B. 1997. Development of a Protocol for Assessing and Comparing the Quality of Aerobic Composts and Anaerobic Digestates, Final Report. International Energy Agency Bioenergy Anaerobic Digestion Activity, Environment Canada, Ottawa, Can. Brodie, H. L., Gouin, F. R., and Carr, L. E. 1993. Quality Criteria for Marketable Compost. American Society of Agricultural Engineers, Paper No. 934031, St. Joseph, USA. BNQ. 1996. Organic Soil Conditioners - Compost: CAN/BNQ 0413-200, National Standard of Canada. Le Bureau de Normalisation du Quebec, Quebec, Can. Crawford, R. L. 1981. Lignin Biodegradation and Transformation. John Wiley and Sons Inc., New York, USA. de Bertoldi, M . , Vallini, G., Pera, A. , and Zucconi, F. 1982. Comparison of Three Windrow Compost Systems. Biocycle 23(2):45-50. 68 de Bertoldi, M . , Rutili, A. , Citterio, B. , and Civillini, M . 1988. Composting Management: A New Process Control Through 0 2 Feedback. Waste Management and Research (6):239-259. Elwell, D. L. , Keener, H. M . , and Hansen, R. C. 1996. Controlled, High Rate Composting of Mixtures of Food Residuals, Yard Trimmings and Chicken Manure. Compost Science & Utilization 4(1):6-15. Finstein, M . S., Miller, F. C , Strom, P. F., MacGregor, S. T., and Psarianos, K. M . 1983. Composting Ecosystem Management for Waste Treatment. Bio/Technology 1:347-353. Finstein, M . S., and Miller, F. C. 1985. Principles of Composting Leading to Maximization of Decomposition Rate, Odor Control, and Cost Effectiveness. Composting of Agricultural and Other Wastes, ed. Gasser, J. K. R. Elsevier Applied Science Publishers, London & New York: 13-26. Finstein, M . S., Miller, F. C , MacGregor, S. T., and Psarianos, K. M . 1992. The Rutgers Strategy for Composting: Process Design and Control. Acta Horticulturae 302:75-86. Finstein, M . S., and Hogan, J. A. 1993. Integration of Composting Process Microbiology, Facility Structure and Decision-Making. Science and Engineering of Composting Design, Environmental, Microbiological and Utilization Aspects, Hoitink, H. A. J. and Keener, H. M . , eds. Proceedings of the International Composting Research Symposium, May 1992, Columbus, USA. Gallardo, J. F., Saavedra, J., Martin-Patino, T., and Millan, A. 1987. Soil Organic Matter Determination. Communications in Soil Science and Plant Analysis 18(6):699-707. Garcia, C , Hernandez, T., and Costa, F. 1990. Phytotoxicity Suppression in Urban Organic Wastes. Biocycle, June 1990: 62-63. Golueke, C. G. 1977. Biological Processing: Composting and Hydrolysis. Handbook of Solid Waste Management, D. Wilson, Ed., Van Hostrand Reinhold Co., New York, USA: 202-203. Golueke, C. G. 1982. When is Compost Safe?. Biocycle Mar/Apr 1982:28-36. Hansen, R. C , Keener, H. M . , Marugg, C , and Dick, W. A. 1991. Nitrogen Transformations During Poultry Manure Composting. American Society of Agricultural Engineers, Paper No. 914014, St. Joseph, USA. Haug, R. T. 1993. The Practical Handbook of Compost Engineering. Lewis Publishers, Boca Raton, FL, USA. Jimenez, E. I., and Garcia, V. P. 1989. Evaluation of City Refuse Compost Maturity: A Review. Biological Wastes 27:115-142. 69 Johnson, G., Crawford, S., and Stark, S. A. 1993. Sampling Municipal Solid Waste Compost. Biocycle December 1993:61-64. Kamiya, A. , and Ose, Y. 1984. Study of Odorous Compounds Produced by Putrefaction of Foods. Journal of Chromatography 292:383-391. Kayhanian, M . , and Tchobanoglous, G. 1992. Computation of C/N Ratios for Various Organic Fractions. Biocycle May 1992:58-60. Keener, H. M . , Marugg, C , Hansen, R. C , and Hoitink, H. A. J. 1993. Optimizing the Efficiency of the Composting Process. Science and Engineering of Composting Design, Environmental, Microbiological and Utilization Aspects. Hoitink, H. A. J. and Keener, H. M . , eds. Proceedings of the International Composting Research Symposium, May 1992, Columbus, USA:59-94. Kuroda, K. , Osada, T., Yonaga, M . , Kanematu, A. , Nitta, T., Mouri, S., and Kojima, T. 1996. Emissions of Malodorous Compounds and Greenhouse Gases from Composting Swine Feces. Bioresource Technology 56:265-271. Lau, A. K. , Lo, K. V. , Liao, P. H. , and Yu, J. C. 1992. Aeration Experiments for Swine Waste Composting. Bioresource Technology 41:145-152. Leton, T. G. And Stentiford, E. I. 1990. Control of Aeration in Static Pile Composting. Waste Management and Research 8:299-306. Liao, P. H. , Vizcarra, A. T., Chen, A. , and Lo, K. V. 1994. Passively Aerated Layered Composting of Salmon Farm Mortalities. Biological Agriculture and Horticulture 10:265-270. MacGregor, S. T., Miller, F. C , Psarianos, K. M . , and Finstein, M . S. 1981. Composting Process Control Based on Microbial Heat Output and Temperature. Applied Environmental Microbiology, 41:1321-1330. Manios, V. I., Tsikalas, P. E., and Siminis, H. I. 1989. Phytotoxicity of Olive Tree Leaf Compost in Relation to the Organic Acid Concentration. Biological Wastes 27:307-317. Martin, A. M . , Evans, J., Porter, D., and Patel, T. R. 1993. Comparative Effects of Peat and Sawdust Employed as Bulking Agents in Composting. Bioresource Technology 44(l):65-69. Mathur, S. P., Schnitzer, M . , and Schuppli, P. 1990. The Distribution of Nitrogen in Peat-Based Composts of Manure Slurries and Fisheries Wastes. Biol. Agric. AndHort. 7:153-163. Mathur, S. P. 1992. Evaluation of Compost Biomaturity Tests and an Experimental Approach to a New Test. Unpublished report, WCI Waste Conversion Inc., Ottawa, Can., January, 1992. Mathur, S. P., Cook, D. G., and Meyboom, P. A. 1994. A Blueprint for Compost Research and Development Activities in Ontario. Prepared under contract for the Waste Reduction Branch of 70 the Ontario Ministry of Environment and Energy, at the Composting Council of Canada, September 1,1994:14-16. Mathur, S. P. 1996. The Use of Compost as a Greenhouse Growth Media. Waste Reduction Branch, Ontario Ministry of Environment and Energy, Toronto, Can. Miller, F. C , and Finstein, M . S. 1985. Materials Balance in the Composting of Wastewater Sludge as affected by Process Control Strategy. Journal Water Pollution Control Federation 57(2): 122-127. Miller, F. C , Harper, E. R., Macauley, B. J., and Gulliver, A . 1990. Composting Based on Moderately Thermophilic and Aerobic Conditions for the Production of Commercial Mushroom Growing Compost. Australian Journal of Experimental Agriculture 30:287-296. Miller, F. C. 1993. Composting as a Process Based on the Control of Ecologically Selective Factors. Soil Microbial Ecology: Applications in Agriculture and Environmental Management, Metting, B., ed. M . Dekker, New York, pp.517-544. Misksche, G. E., and Yasuda, S. 1977. Lignin in Leaves of Several Softwoods and Hardwoods. Holzforschung 31:57-59. Murray, C. M , Thompson, J. L. , and Ireland, J. S. 1991. Process Control Improvements at Composting Sites. Biocycle 32(12):54-58. Often, L., and Dennison, L. 1997. Dimethyl Sulphide and Dimethyl Disulphide Emissions from the Caledon Composting Facility. Unpublished report, Municipality of Peel, Brampton, Can. Ozisik, M . N . 1985. Heat Transfer A Basic Approach. McGraw-Hill, New York, USA. Person, H. L. , and Shayya, W. H. 1994. Composting Process Design Computer Model. Applied Engineering in Agriculture 10(2):277-283. American Society of Agricultural Engineers, St. Joseph, USA. Solomon, E. P., Berg, L. R., Martin, D. W., and Villee, C. V. 1993. Biology, 3rd Ed. Saunders College Publishing, Orlando, USA. Jeris, J. S. and Regan, R. W. 1973. Controlling Environmental Paramaters for Optimum Composting - Part III. Compost Science, May/June 1973. Tseng, D. Y. , Chalmers, J. J., Tuovinen, O. H. , and Hoitink, H. A. J. 1995. Characterization of a Bench-Scale System for Studying the Biodegradation of Organic Solid Wastes. Biotechnology Progress 11: 443-451. Vallini, G., and Pera, A. 1989. Green Compost Production from Vegetable Waste Separately Collected in Metropolitan Garden-Produce Markets. Biological Wastes 29:33-41. 71 Vallini, G., Pera, A. , Briglia, M . , and Perghem, F. 1989. Compost Detoxification of Vegetable-Tannery Sludge. Waste Management and Research 7:277-290. Van Lier, J. J. C , Van Ginkel, J. T., Straatsma, G., Gerrits, J. P. G., and Van Griensven, L. J. L. D. 1994. Composting of Mushroom Substrate in a Fermentation Tunnel: Compost Parameters and a Mathematical Model. Netherlands Journal of Agricultural Science 42-4. Walker, J. M . 1993. Control of Composting Odors. Science and Engineering of Composting Design, Environmental, Microbiological and Utilization Aspects, Hoitink, H. A. J. and Keener, H. M . , eds. Proceedings of the International Composting Research Symposium, May 1992, Columbus, USA. Waste Program Consortium. 1992. Waste Analysis Sampling, Testing and Evaluation, Effect of Waste Stream Characteristics on Municipal Solid Waste Incineration: the Fate and Behavior of Metals. (Mass Burn M S W Incineration: Burnaby, BC), Table ES-5. Produced for Environment Canada, US EPA, International Lead Zinc Research Organization, and the Greater Vancouver Regional District. Consortium members include: A. J. Chandler and Assoc. Ltd.; Regal and Rego Assoc. Inc.; Environmental Research Group - University of New Hampshire; and the Wastewater Technology Centre, Burlington, Ont. Witter, E. and Lopez-Real, J. M . 1987. The Potential of Sewage Sludge and Composting in a Nitrogen Recycling Strategy for Agriculture. Biological Agriculture and Horticulture 5:1-23. Witter, E. and Lopez-Real, J. M . 1988. Nitrogen Losses During the Composting of Sewage Sludge, and the Effectiveness of Clay Soil, Zeolite, and Compost in Adsorbing the Volatilized Ammonia. Biological Wastes 23:279-294. Zucconi, F., Forte, M . , and Monaco, A. 1981. Biological Evaluation of Compost Maturity. Biocycle July/August 1981:27-29. 72 Appendix A Sample Spreadsheets This section provides a few samples of data collection and analysis spreadsheets used. Compost Total Nitrogen (Leco method), NH3-N and N03-N (soluble, autoanalyser method) T e s t O F 1 name: Date: 97/03 Sample T N T N S o l i d smpl. %wb % d b Dilution N H 4 - N ppm wb (dilute) N H 4 - N ppm db N 0 3 - N ppm wb (dilute) N 0 3 - N ppm db Moisture % wb Sample Desc : 1a 0.97% 1b 1.31% 1c 1.13% 1d 0.98% 1e 0.82% Average : Std. Dev: Sample Desc : 2a 0.97% 2b 1.31% 2c 1.13% 2d 0.98% 2e 0.82% Average : Std. Dev: Sample Desc : 3a 0.97% 3b 1.31% 3c 1.13% 3d 0.98% 3e 0.82% Average: Std. Dev: R 1 Initial 3.05% 4.12% 3.55% 3.08% 2.58% 3.28% 0.58% R 2 Initial 3.05% 4.12% 3.55% 3.08% 2.58% 3.28% 0.58% R 3 Initial 3.05% 4.12% 3.55% 3.08% 2.58% 3.28% 0.58% 10 133.4 10 176.7 10 177.9 10 105.4 10 108.8 10 133.4 10 176.7 10 177.9 10 105.4 10 108.8 10 133.4 10 176.7 10 177.9 10 105.4 10 108.8 4195.7 5557.5 5595.3 3315.0 3422.0 4417.1 1111.6 4195.7 5557.5 5595.3 3315.0 3422.0 4417.1 1111.6 4195.7 5557.5 5595.3 3315.0 3422.0 4417.1 1111.6 3.5 110 .7 | 68 .2% 3.7 116.4 68.2% 3.7 116.4 68 .2% 3.5 110.1 68 .2% 3.3 103.8 68 .2% 111.5 5.2 3.5 110.7|T 68.2% 3.7 116.4 68 .2% 3.7 116.4 68.2% 3.5 110.1 68.2% 3.3 103.8 68 .2% 111.5 5.2 3.5 H O j f 68 .2% 3.7 116.4 68 .2% 3.7 116.4 68 .2% 3.5 110.1 68 .2% 3.3 103.8 68 .2% 111.5 5.2 Table A - l . Nitrogen Measurement Spreadsheet. 73 Compost Moisture, VS, TC (LOI method) Test name: O F 2 Date: 97/04 Sample C r u c Cruc+ws Cruc+ds Cruc+ash1 Cruc+ash2 Moisture Moisture V S A s h Total C O M 550C/15min 660C/60min % wb % d b % d b % d b % d b % d b Sample D e s c : R1 initial 1a 44.081 51.341 46.914 44.448 44.349 61,0% 156.3% 87.0% 9.5% 50.3% 90.5% 1b 42.415 50.599 45.511 43.014 42.779 62.2% 164.3% 80.7% 11.8% 49.0% 88.2% 1c 44.803 54.011 48.233 45.471 45.239 62 .7% 168.5% 80.5% 12.7% 48 .5% 87.3% 1d 42.419 49.658 45.32 42.824 42.672 59.9% 149.5% 86.0% 8.7% 50.7% 91.3% A v e r a g e 1: 61 .5% 159.6% 83.6% 10.7% 49.6% 89.3% Std. Dev. 1.3% 6.2% 3.7% 1.7% 0.9% 1.7% S a m p l e Desc : R2 Initial 2a 44.081 51.341 46.914 44.448 44.349 61.0% 156.3% 87.0% 9.5% 50.3% 90.5% 2b 42.415 50.599 45.511 43.014 42.779 62.2% 164.3% 80.7% 11.8% 49.0% 88.2% 2c 44.803 54.011 48.233 45.471 45.239 62 .7% 168.5% 80.5% 12.7% 48 .5% 87.3% 2d 42.419 49.658 45.32 42.824 42.672 59.9% 149.5% 86.0% 8.7% 50.7% 91.3% A v e r a g e 2: 61 .5% 159.6% 83.6% 10.7% 49.6% 89.3% Std. Dev. 1.3% 6.2% 3.7% 1.7% 0.9% 1.7% S a m p l e Desc : R3 Initial 3a 44.081 51.341 46.914 44.448 44.349 61.0% 156.3% 87.0% 9.5% 50.3% 90.5% 3b 42.415 50.599 45.511 43.014 42.779 62.2% 164.3% 80.7% 11.8% 49.0% 88.2% 3c 44.803 54.011 48.233 45.471 45.239 62 .7% 168.5% 80.5% 12.7% 48 .5% 87.3% 3d 42.419 49.658 45.32 42.824 42.672 59.9% 149.5% 86.0% 8.7% 50.7% 91.3% A v e r a g e 3: 61 .5% 159.6% 83.6% 10.7% 49.6% 89.3% Std. Dev. 1.3% 6.2% 3.7% 1.7% 0.9% 1.7% Table A-2. Moisture, VS, TOC, and O M Measurement Spreadsheet. 74 Compost Mass Balance Initial Mass (kg) 13.44| 13.44 13.44 Initial to Final Active Moisture (%wb) 68.2% 68.2% 68.2% Mass loss (kg) 2.71 2.15 2.11 Volatile Solids (%db) 86.3% 86.3% 86.3% Water loss (kg) 1.93 1.49 1.48 Organic Matter (%db) 91.7% 91.7% 91.7% Solids loss (kg) 0.78 0.66 0.63 Water (kg) 9.17 9.17 9.17 Solids loss (%) 18.2% 15.5% 14.8% Solids (kg) 4.27 4.27 4.27 Volatile solids loss (%) 27.7% 19.6% 23.8% Volatile solids (kg) 3.69 3.69 3.69 Org. matter loss (%) 25.6% 19.0% 21.2% Final Active Mass (kg) 10.73| 11.29 11.33 Final Active to Cured Moisture (%wb) 67.4% 68.0% 67.9% Mass loss (kg) 4.79 5.10 5.41 Volatile Solids (%db) 76.3% 82.1% 77.2% Water loss (kg) 4.09 4.48 5.05 Organic Matter (%db) 83.5% 88.0% 84.9% Solids loss (kg) 0.70 0.62 0.36 Water (kg) 7.24 7.68 7.69 Solids loss (%) 20.4% 17.3% 10.1% Solids (kg) 3.49 3.61 3.64 Volatile solids loss (%) 13.0% 20.9% 4.7% Volatile Solids (kg) 2.67 2.96 2.81 Org. matter loss (%) 14.4% 19.0% 4.5% Curing Starting Mass (kg) 10.57 11.14 11.19 Initial to Cured Water (kg) 7.13 7.58 7.60 Solids loss (%) 34.9% 30.1% 23.5% Solids (kg) 3.44 3.56 3.59 Volatile solids loss (%) 37.1% 36.4% 27.4% Volatile Solids (kg) 2.63 2.92 2.77 Org. matter loss (%) 36.3% 34.4% 24.8% Ending Mass (kg) 5.78 6.04 5.78 Moisture (%wb) 52.6% 51.2% 44.1% Volatile Solids (%db) 83.3% 78.5% 81.8% Organic Matter (%db) 89.8% 86.2% 90.2% Water (kg) 3.04 3.09 2.55 Solids (kg) 2.74 2.95 3.23 Volatile Solids (kg) 2.28 2.31 2.64 Table A-3. Mass Balance Measurement Spreadsheet. 75 Appendix B Aeration Calculations Feedstock parameters: Volatile solids (VS) 80% dry basis (estimated from early results) Moisture 65% Mass 16 kg (wet basis) 1. Determination of aeration for peak oxygen demand (Haug (1993) page 269). From Haug Figure 10-17, peak oxygen consumption at 60 °C for mixed garbage and refuse is approximately 5 mg 0 2 /g VS»h. Air is 23.2% oxygen by weight (Haug p.263). Air demand = 5 x 10"3 g 0 2 /g VS»h * (1/0.232) g air/g 0 2 2.16 x 10'2g air/g VS 'h 1.8 x 1 0 5 m 3 air/g VS«h VS = 16 x l 0 3 g solids * (l-0.65)(solids content)* 0.8 (VS content) 4 . 4 8 x l 0 3 g .-.Air demand = 1.8 x 10"5 m 3 air/g VS»h * 4.48 x 103 g VS 0.081 m3/h 1.3 tf/min 2. Determination of aeration for peak heat removal. Conditions: Ambient temp.(inlet) 25 °C Ambient relative humidity 60% Exhaust temp, (outlet) 5 5 ° C AT (inlet - outlet) 30°C Set heat release = heat "demand" of air a. Determine water removed with exhaust gas. At outlet, assume exhaust saturated: Specific humidity W o u t = 0.12 gH 2 0/gdry air (@ 55°C) At inlet: Water vap. pressure P v = P v s * relative humidity 76 Sat. water vap. press. P v s Pv Water removed b. Heat demand of air. i . Vaporisation heat i i . Heating dry air i i i . Heating water vapour Total c. Heat production. Assume 3260 cal/g 0 2 consumed. Peak heat output .-. Peak air demand for cooling 10 x(-2238/T(°K)+ 8.896) 24.32 mm Hg 24.32 * 0.6 14.6 mm Hg 0.6221 ( P ^ - P v ) 0.0122 gH 20/g dry air W - W V Y out Y V in 0.1078 gH 20/g dry air water removed * heat of vaporisation 0.1078 g H 20/g dry air * 585.81 cal/g H 2 0 63.1 cal/g air AT * spec, heat of air 30°C * 0.24 cal/g«°C 7.2 cal/g air AT * (water removed + inlet vapour) * spec, heat of water vapour 30°C * (0.1078+0.0122) g H 20/g dry air * 0.44 cal/g* °C 1.6 cal/g air 71.9 cal/g air 3260 cal/g 0 2 * 5 x 10"3 g 0 2 /g VS 'h * 4.48 x 103 g VS 73024 cal/h 73024 cal/h * 1/(71.9 cal/g air) 1015.6 gair/h 0.846 m3/h 14.1 0/min Note that this calculation assumes that all heat is lost to aeration, whereas there is also heat loss through the reactor; therefore the actual required airflow for cooling will be less. Value used is for 20°C, however the difference in heat of vaporisation for 25 °C is not significant. 77 Appendix C Source Code The source code provided in this section was written in ANSI C. 78 Oxygen Feedback /* 02_CTRL1 */ /* This program accepts an input of oxygen level between 0 and 100 (Xin) as well as temperature (Yin), and outputs a flag (0 or 1) indicating whether the aeration should be off (0) or on (1) during each cycle based on the oxygen level. Note the output is always 0 when the temperature is above the setpoint. (Temperature feedback control is a separate module) It is based on the sample code rbv.c and example, c */ #define tempsp UsDefVarl /* temperature setpoint (int) */ #define duty_percent UsDefVar2 /* duty cycle in percent (int) */ #define o2_high UsDefV ar3 /* high end of o2 range (float) */ #define o2_low UsDefVar4 /* low end of o2 range (float) */ #include "windows.h" #include "math.h" #include "ltuserw.h" /* global variables */ int counter; int maxcount; int air_interval; /* interval (#counts) between air on cycles */ int temp_interval; /* #counts since temperature exceeded setpoint */ int o2_delay; /* #counts temp must be < setpoint before adjusting aeration duty cycle float o2_surge; /* maximum allowable o2 increase for changing duty cycle */ float o2_dead; float o2_curr; float o2_prev; float tempcurr; int air_flag; /* minimum o2 level for changing duty cycle */ /* current o2 level */ /* previous o2 level */ /* current temp */ /* output (0 or 1)*/ 79 int dutycount; int duty_count_incr; int duty_count_min; int air_array[10][10]= {1,0,0,0,0,0,0,0,0,0, /* array that determines air on/off cycle */ 1,0,0,0,0,1,0,0,0,0, /* a given duty cycle uses a single row */ 1,0,0,1,0,0,1,0,0,0, /* eg. duty cycle 20% uses row 2 */ 1,0,1,0,0,1,0,1,0,0, 1,0,1,0,1,0,1,0,1,0, 1,0,1,0,1,1,0,1,0,1, 1,1,0,1,1,0,1,1,0,1, 1,1,1,1,0,1,1,1,1,0, 1,1,1,1,1,1,1,1,1,0, 1,1,1,1,1,1,1,1,1,1}; struct ReadBlocklnput Datablock; struct ReadBlockOutput Value; void FAR PASCAL LT_CIcon_init(fh_name, ptr, callback) char *fh_name; CICONDATA *ptr; void *call_back(int m, void far *inp_data, void far *outp_data); { int i; Ciconptr = ptr; /* initialize algorithm variables */ counter = 1; max_count= 10; o2_delay = 15; temp_interval = o2_delay + 1; dutycountrnin = 1; duty_count_incr = 1; o2_surge = 5; o2_dead = 0.5; o2jprev=100; duty_count = floor((duty_percent/10)+0.5); /* duty cycle represented by integer eg 20% is 2 */ 80 if (duty_count == 0) duty_count = duty_count_min; } void FAR PASCAL LT_CIcon_cIose(handle) int handle; { } void FAR PASCAL LT_CIcon_Open(ptr) CICONDATA *ptr; { int i; Ciconptr = ptr; } void FAR PASCAL oxygenl(ptr, callback) CICONDATA *ptr; void *call_back(int fh, void far *inp_data, void far *outp_data); { /* run time */ Ciconptr = ptr; Data_block.usNum=Xin; /* set X input block as the block to read */ call_back(READ_BLOCK_VAL,(void *)&Data_block,(void *)&Value); /* get current value */ if(Value. error) /* check for error condition */ { error_status=Value. error; BlockValReadErr; /* use L A B T E C H error macro to return correct error code */ LTReturn; /* pass back error information and return to L A B T E C H */ } o2_curr = Value. dData; Data_block.usNum=Yin; /* set Y input block as the block to read */ 81 call_back(READ_BLOCK_VAL,(void *)&Data_block,(void *)&Value); /* get current value */ if{ Value, error) /* check for error condition */ { error_status=Value, error; BlockValReadErr; /* use L A B T E C H error macro to return correct error code */ LTReturn; /* pass back error information and return to L A B T E C H */ } temp_curr = Value. dData; /* start algorithm */ if ( counter = maxcount + 1 ) counter = 1; /* reset at end of cycle */ if (tempcurr >= tempsp) temp_interval = 0; /* check for temp exceeding setpoint */ else temp_interval++; /* increment temp interval if below setpoint */ if ( counter == 1 ) { /* check oxygen and potentially adjust duty cycle once per max counter cycles */ if (o2_curr > o2_high && o2_curr - o2_prev < o2_surge && dutycount > duty_count_min && tempinterval >= o2_delay) duty_count = dutycount - dutycountincr; /* decrease duty cycle */ if (o2_curr < o2_low && o2_curr > o2_dead && dutycount < 10) duty_count = duty_count + duty_count_incr; /* increase duty cycle */ } if (temp_curr >= temp_sp) air flag = 0; else air_flag = air_array[duty_count-l][counter-l]; /* air on when temp below setpoint based on duty cycle dCalcResult = air_flag; /* output air flag */ 82 o2_prev = o2_curr; counter++; /* end algorithm */ LTReturn; } Linear Temperature Feedback /* t2_CTL3 */ /* This program accepts an input of temperature (Xin), and outputs a flag (0 or 1) indicating whether the aeration should be off (0) or on (1) during each cycle based on the temperature and the temperature trend (increasing or decreasing). Air duty cycle is checked and potentially adjusted once every "long cycle" (maxcount2 counts). Aeration is off when temperature exceeds the setpoint. Based on the sample code rbv.c and example, c */ #define temp_sp UsDefVarl /* temperature setpoint (int) */ #define duty_percent UsDefVar2 /* duty cycle in percent (int) */ #include "windows.h" #include "math.h" #include "ltuserw.h" /* global variables */ int counter; int counter2; 83 int maxcount; int max_count2; int airinterval; int temp_interval; int dc_adj_delay; float temp_curr; float temp_avg_curr; float tempavgjprev; float temp_sum; float tempchange; int airflag; int dutycount; int duty_count_incr; int dutycountmin; intair_array[10][10]= /* interval (#counts) between air on cycles */ /* #counts since temperature exceeded setpoint */ /* #counts temp must be < setpoint before adjusting aeration duty cycle */ /* current temp */ /* average temperature current period */ /* average temp, from previous period */ /* sum used to calculated first temp, average */ /* change from previous to current period */; /* sum used to calculated temp, average */ /* output (0 or 1)*/ {1,0,0,0,0,0,0,0,0,0, 1,0,0,0,0,1,0,0,0,0, 1,0,0,1,0,0,1,0,0,0, 1,0,1,0,0,1,0,1,0,0, 1,0,1,0,1,0,1,0,1,0, 1,0,1,0,1,1,0,1,0,1, 1,1,0,1,1,0,1,1,0,1, 1,1,1,1,0,1,1,1,1,0, 1,1,1,1,1,1,1,1,1,0, 1,1,1,1,1,1,1,1,1,1}; /* array that determines air on/off cycle */ /* a given duty cycle uses a single row */ /* eg. duty cycle 20% uses row 2 */ struct ReadBlocklnput Data_block; struct ReadBlockOutput Value; void FAR PASCAL LT_CIcon_init(fh_name, ptr, callback) char *fh_name; CICONDATA *ptr; void *call_back(int fh, void far *inp_data, void far *outp_data); { int i; Ciconptr = ptr; 84 /* initialize algorithm variables */ counter = 1; counter2 = 1; max_count= 10; max_count2 = 30; dc_adj_delay = 60; temp_interval = 0; duty_count_min = 1; temp_sum = 0; temp_avg_prev = 0; duty_count = floor((duty_percent/10)+0.5); /* duty cycle represented by integer eg 20% is 2 */ if (duty_count == 0) dutycount = dutycountjmin; } void FAR PASCAL LT_CIcon_close(handle) int handle; { } void FAR PASCAL LT_CIcon_Open(ptr) CICONDATA *ptr; { int i; Ciconptr = ptr; } void FAR PASCAL model3(ptr, call_back) CICONDATA *ptr; void *call_back(int m, void far *inp_data, void far *outp_data); { /* run time */ Ciconptr = ptr; 85 Data_block.usNum=Xin; /* set X input block as the block to read */ call_back(READ_BLOCK_VAL,(void *)&Data_block,(void *)&Value); /* get current value */ ifTValue. error) /* check for error condition */ { error_status=Value. error; BlockValReadErr; /* use L A B T E C H error macro to return correct error code */ LTReturn; /* pass back error information and return to L A B T E C H */ } temp_curr = Value. dData; /* start algorithm */ if ( counter = maxcount + 1 ) counter = 1; /* reset at end of small cycle */ if ( counter2 = max_count2 + 1 ) counter2 = 1; /* reset at end of large cycle */ if ( counter2 <= 30 && counter2 >= 2 ) /* add temp, value to sum */ temp_sum = temp_sum + temp_curr; if (tempcurr >= tempsp) temp_interval = 0; /* check for temp exceeding setpoint */ else temp_interval++; /* increment temp interval if below setpoint */ if ( counter2 = 1 ) { /* check temp, averages and adjust duty cycle once per maxcount2 counts */ 86 temp_avg_curr = tempsum / 29; /* calculated temp, averages */ tempchange = temp_avg_curr - temp_avg_prev; tempsum = 0; /* reset temp, sum value */ if (temp_change < -0.02 && tempinterval > dc_adj_delay) /* decreasing trend */ duty_count = floor(((temp_avg_curr * 0.0286 - 1.26) * 10)+0.5); if (tempchange > 0.2 && temp_interval > dc_adj_delay) /* increasing trend */ duty_count = floor(((temp_avg_curr * 0.00686 + 0.06) * 10)+0.5); if (duty_count< 1) dutycount = duty_count_min; if (duty_count> 10) duty_count= 10; } if (tempcurr > temp_sp) air_flag = 0; /* air on when temp exceeds setpoint */ else air flag = air_array[duty_count-l][counter-l]; /* air on when temp below setpoint based on duty cycle */ dCalcResult = air flag; /* output air flag */ temp_avg_prev = temp_avg_curr; counter++; counter2++; /* end algorithm */ LTReturn; } 87 Appendix D Aeration vs. Temperature Modelling Data 350 300 r 2 0 0 150 O 100 50 0 o O 0 O R3 02 uptake • R3 Aeration o o 0 o o Jnear Linear (R3P (R3C eratic )2upt n) ake) • o o o • mm o°b o < r oo o • • >oo i : 0-9 0 . 8 : 0.7 ; o.6 a To « - 0 . 5 s. o >% c 0 O 0 . 4 2 f < ; - 0 . 3 : 0 2 ; 0.1 - 0 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 Temperature (Degrees C) Figure D-2. Run 4 Aeration and Oxygen Uptake vs. Temperature - Decreasing Trend. 88 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items