Troubling times for works
Aqua Enviro's Paul Lavender takes a look at WwTW troubleshooting
As such, it is always preferable to take a proactive approach to problem-solving because this allows a preventative course of action to be followed rather than waiting until a situation has become much more serious and expensive action is required to avoid consent failure. Treatment plant operation is analogous to running a family car where regular servicing is a far better option than a breakdown on the M25. However, many people have the mentality of "if it ain't broke, don't fix it" and if a plant complies with consent it 'ain't broke'.
Such an approach necessarily means that action is only taken when a plant is close to or actually failing consent. Regular plant monitoring and troubleshooting ensures optimum effluent quality at lowest cost. The activated sludge process is one of the most common treatment processes and it is basically very simple, with only very few options available for operator control.
These include dissolved oxygen and nutrient levels in the aeration basin, the F/M ratio, sludge age and the return and surplus sludge rates from the secondary sedimentation tank. All of these control parameters affect the status of the micro-organisms responsible for undertaking treatment and consequently the first step to investigating the process is to carry out microscopic examination. Where microscopy is carried out regularly, changes within the sample can give an early warning of problems before they develop. When a sample is received from a problem site this can often draw attention to the cause/s allowing quick remedial action. Causes can typically include problems such as low DO or nutrients, septicity, toxicity, pH and shock loads. When no obvious cause and solution is found then an on-site trouble-shooting investigation is carried out. The first question that must be answered when investigating a problematic site is "is the plant big enough to treat the flow and load it receives?" Any treatment process will have a bottleneck, which limits its actual capacity and typical bottlenecks include an aeration tank size, aeration capacity, sedimentation tank area and return and waste sludge pumping capacity. A plant operator should know the capacity of their plant and should have access to adequate data to accurately determine the flows and loads received to the plant.
The operation of each stage of the treatment process should be compared to the design upon which it was based. Where a stage of the treatment process is operating outside of its capacity, it is likely this will impact on its performance and the performance of other areas of the process. The first step to being able to understand and control the process is to characterise and quantify the wastewater at each stage of the treatment process. Auto-samplers are essential in order to obtain composite samples because spot samples are rarely representative of the daily composition. Sampling and analysis regimes vary greatly between sites and therefore the current data should be examined and further analysis carried out to complement this where required. The level of analysis required will vary depending on the nature of the site and the type of problems being experienced. For example with a domestic site or an industrial site with 24h continuous production and some upstream balancing, a 24h composite may be representative.
By contrast, a factory with regular changes in production and wash-downs using chemical cleaning agents may require determination of peak loads and changes in pH through analysis of hourly samples. Once the influent has been accurately quantified, each stage of the process should be examined as the operation of each stage has an effect downstream on the other treatment stages. If areas of the treatment process are operating outside of their design capacity then one of the following courses of action must be taken:
- reduce the flow and load to the plant through upstream waste minimisation,
- enhance the process through chemical addition - coagulants/flocculants/polymers,
- reduce load variations through upstream balancing,
- provide additional capacity through either expanding the plant or retrofitting the current plant, for example, lamella plates in the primary tanks, drop in aeration or pure oxygen units in the aeration tanks.
Once the plant is operating within its capacity, the plant should be capable of achieving a good quality effluent if the process parameters are correct. Therefore each stage of the process should be investigated to ensure optimum operation. The most common operating problems that experience are:
- filamentous bacteria, which produce either sludge bulking and create poor settlement or foaming, which leads to solids loss and solids consent failure,
- high effluent suspended, solids or turbidity,
- rising sludge in the final clarifier.
Every site is unique and there can often be multiple causes for a problem, so the common causes discussed here should not be used as the only avenues for investigation. Microscopy should highlight areas of concern as filamentous and protozoal species indicate the current process conditions within the plant. The most common causes of operational problems are discussed below. For successful treatment there must be sufficient nutrients for the micro-organisms to use, where nutrients are in short supply filamentous bulking is the most common problem. Nutrient shortages are usually only seen on municipal sites when there is a large trade component to the waste. The commonly quoted rule is a BOD:N:P ratio of 100:5:1, but in reality this is a maximum requirement only usually approached at very high loading levels and most processes can operate below this. The best way to control a process is to monitor the residual nutrients in the effluent and to aim for a level of N (ammonia and nitrate) and P of around 1-2 mg/l. However, on a number of sites where anoxic tanks are emp-loyed, increased nutrient addition can lead to nitrification and denitrification and thus no increase in residual N is observed. Therefore, if the ratio of nutrients is known then monitoring effluent P concentration provides a better control parameter.
A lack of micro-nutrients can sometimes also be a problem and if this is suspected, it is often better to trial nutrient dosing with a proprietary brand containing micro-nutrients and observe if there is a reduction in SSVI or improvement in plant performance. Attempting to analyse for the many possible compounds that may be lacking can prove an impossible task.
A lack of oxygen or poor mixing creating zones deficient in oxygen can create odours and promote filamentous bulking. Odour is a particular problem at industrial sites where the organic compounds in the wastewater generate volatile organic compounds (VOCs) and the stronger wastes treated mean that interrupted supply of oxygen will rapid lead to the formation of hydrogen sulphide.
It is desirable to maintain a DO level of 1-2mg/l in the aeration basin to enhance settleablity of the sludge flocs - due to their greater surface area filaments can effectively utilise oxygen at 0.05mg/l thus giving them a competitive advantage at lower DO levels. Low DO and/or poor mixing may be able to be altered through changing set points, refurbishing blowers and diffusers or adding drop-in mixers or blowers. The ideal F/M and sludge age will vary depending on the site and it is often a balance between different factors. The ideal growth conditions for different filaments are well documented (Jenkins et al. 1993) and the process can be altered to prevent filament proliferation. Where shocks are received to the plant (pH, toxicity, load) a higher MLSS level may provide more buffering and enhance biodegradation, while in some cases a higher MLSS may promote other problems such as foaming. Knowledge of the history of a site is advantageous in determining the optimum F/M. Other factors will restrict the range within which the MLSS can be altered, including aeration and clarifier capacity.
The clarification capacity can be calculated by determining the applied and predicted solids load to the clarifier using a recognised technique, such as the WRc nomograph (WRc Activated Sludge Design Manual 1991). While some factors affecting clarification capacity can be altered (MLSS, RAS rate), others factors, such as the SSVI, cannot.
A vicious circle can often develop whereby high SSVI values mean the MLSS must be kept low to avoid sludge blanket carry-over, thus giving a non optimum F/M, which promotes filament growth giving a higher SSVI. In these cases short-term remedial action, such as hypochlorite dosing, may be required to attempt to bring the process back under control.
Filaments proliferate in completely mixed conditions because they can out-compete floc forming bacteria for food due to their greater surface area. Where an initial high load is applied the floc formers can store the substrate, thus 'selecting' floc forming bacteria. An initial load (floc loading) of 150mg soluble BOD/g MLSS is a good benchmark value for setting up a selector tank prior to the aeration basin. These tanks may be aerobic, anoxic or anaerobic, although it is important that low DO conditions do not exist.
Where completely mixed conditions exist and filamentous bulking is a problem then it may be possible to retrofit a selector by compartmentalising the inlet end of the aeration basin. Septicity in the influent or developing within the process can promote filamentous bulking and generate odour problems.
Where septicity is present filaments are usually seen containing characteristic glowing sulphur particles. Any odour problems within the process will also generate odours when processing the sludge.
At industrial sites septicity may be tackled by looking at upstream processes and practices although this is not so easy for domestic sites where it may be a feature of the sewer network. Nitrate containing compounds can be added to the influent to reduce septicity. Maintaining sufficient DO and mixing as discussed above should stop septicity occurring within the process. Micro-organisms are sensitive to their environmental conditions and are vulnerable to both pH and toxic shocks. Although bacteria are happy operating within the pH range 6.5-8.0, they can be severely damaged if there is a rapid change, even within this range. The balance of ionised proteins on the cell surface is upset and the bacteria and flocs are damaged. Toxicity can work in a similar way or it may just inhibit the metabolism of the bacteria. Where shocks occur, then a high quantity of bacterial cells can be washed into suspension giving a turbid supernatant.
This can also occur if sudden shocks of high load are experienced, which causes bacterial growth to become dispersed. The collection of the correct data is essential to keep a treatment process working within its optimum conditions. Microscopic analysis is particularly useful as this can not only give information about the conditions within the process it can also provide an early warning before problems develop. When problems do develop the data and microscopic analysis allow a methodical approach to be taken to identifying and tackling the causative factors for poor treatment.