ABSTRACT
In 1991, the Greater Vancouver Regional District (District) began design of the upgrade of the Annacis Island Wastewater Treatment Plant to secondary treatment. During the design, the District adopted a policy requiring that all biosolids generated from treatment be used beneficially. Removing debris from the biosolids is considered crucial to developing market acceptance of the product. In addition, removing or grinding material larger than 12 mm in the solids stream is required to avoid blocking process equipment. In North America, grinders are most commonly used to macerate sludge and scum. In Europe, screens with an opening size of 10 mm or less have been successfully used for removing debris from solids streams. The advantages of screening over grinding are lower capital cost if a significant number of grinders can be eliminated, a reduction in the quantity of debris entering the digesters and, as a result, reduced frequency of digester cleaning, reduced frequency of maintenance of equipment in the digester area, and less debris in the digested biosolids. Sludge and scum screens began operation in April 1996. This paper discusses the rationale behind the selection of these screening processes, the technology selection and design, and initial operation of the screens.

KEYWORDS
Wastewater treatment, sludge, scum, screens, screening, biosolids

INTRODUCTION
The Greater Vancouver Regional District (District) operates four major primary wastewater treatment plants in the Greater Vancouver area of southwestern British Columbia. In 1991, the District began preliminary design planning to upgrade two of their plants, the Annacis Island and Lulu Island Wastewater Treatment Plants, to secondary treatment. Both of these plants discharge to the Fraser River, the world's largest salmon fishery.

After an extensive evaluation process, the District selected the trickling filter/solids contact (TF/SC) process to treat the liquid stream at both plants. In addition, the plants will incorporate anaerobic digestion and centrifuge dewatering of solids. Other important improvements made as part of this $600 million project include new influent pumping and screening facilities, ultraviolet disinfection, dissolved air flotation thickeners, new operations centres, and extensive odour control.

This paper will focus on the Annacis Island plant. When completed, the Annacis Island plant will handle an average dry weather flow (ADWF) of 5.59 m³/s (480 ML/d) from a population of about 1 million with plans for future expansion to an ultimate capacity of 11.2 m ³/s (970 ML/d). Table 1 shows design parameters for the Annacis Island plant.

Table 1. Annacis Island Wastewater Treatment Plant Design Parameters
Parameter	Value
Design Population	995,000
Average Dry Weather Flow, ML/d	480
Peak Wet Weather Flow, ML/d	1,090
Average Annual SS Loading, kg/d	91,500
Average Annual BOD Loading, kg/d	101,500

Design of secondary facilities at the Annacis Island plant is complete. The Annacis Island plant is under construction with anticipated completion in late-1998. Design and construction of the anaerobic digestion facility was fast-tracked and was completed in late-1996.

During the course of planning for these upgrades, the District identified the need to remove debris from the solids streams in the plant and the ultimate biosolids product. Raw sludge and scum screening were selected. The design and construction of these processes were fast-tracked, and sludge and scum screens began operation in April 1996.

This paper discusses the rationale behind the selection of these screening processes, the technology selection and design, and initial operation of the screens.

RATIONALE FOR SLUDGE AND SCUM SCREENING
The Greater Vancouver Regional District Board has adopted a policy requiring that all biosolids be used beneficially, except under emergency conditions, when the District will seek approval for landfill burial. To provide a reliable biosolids utilization program, the District's philosophy is to ensure that treated biosolids may be used in as many market areas as possible. An aesthetically-acceptable product is required to maximize the availability of markets and to provide flexibility to divert biosolids to other markets if the demand or acceptability in one market area changes. Removing identifiable debris from the biosolids is considered crucial to developing market acceptance of the product; raw sludge and scum screening was seen as one way of accomplishing this.

Another aspect of the District's philosophy to ensure multiple markets is the decision to produce high-grade biosolids as defined by the Province of British Columbia's new strict biosolids regulations, which are similar to USEPA Class A requirements. To provide high-grade biosolids, the District selected the extended thermophilic anaerobic digestion process which will provide the required pathogen destruction and improved dewaterability of the wastewater solids. This process incorporates numerous heat exchangers to transfer and recover heat from various solids streams. After evaluation of various heat exchanger technologies, spiral heat exchangers were selected. Spiral heat exchangers consist of two steel plates, rolled into a cylindrical spiral to form alternate passages for the heat sink (incoming sludge) and the heat source (hot digested sludge or hot water). Spiral heat exchangers provide the highest heat transfer efficiency and, therefore, the smallest space requirements compared to alternatives. This saved overall construction cost by reducing building space requirements.

To maximize heat transfer capability, a plate spacing of 13 mm was selected for the heat exchangers. Debris in the solids stream larger than the plate spacing can block the heat exchangers; therefore, removing or grinding material larger than 13 mm is required.

In North America, wastewater treatment plants most commonly use grinders to macerate the influent sludge and scum, and avoid blocking heat exchangers and other process equipment. In Europe, wastewater treatment facilities have successfully employed screens with an opening size of 10 mm or less for removing debris from these solids streams. The advantages of screening over grinding sludge and scum are:

• Lower capital cost if a significant number of grinders can be
  eliminated.
• A reduction in the quantity of debris entering the digesters and,
  as a result, reduced frequency of digester cleaning.
• A reduction in the frequency of maintenance of equipment in
  the digester area.
• A reduction in the quantity of debris in the digested biosolids,
  thus improving its aesthetic quality for beneficial use.

Pending confirmation that reliable technologies were available, selected over grinding for removal of debris from the solids streams.screening was To minimize the required hydraulic capacity of the screens, primary sludge is gravity thickened prior to screening. Screened, thickened primary sludge is then pumped to mixed sludge tanks where it is blended with thickened waste secondary sludge (WSS), the excess biological solids from the TF/SC process, prior to anaerobic digestion. Figure 1 shows the sludge and scum screening process.

Scum at the plant is collected in a very dilute state, making scum thickening also necessary. However, because screening thickened scum would increase the likelihood of screen blinding, it was decided to screen unthickened scum prior to thickening. Screened scum is then combined with WSS and thickened in dissolved air flotation thickeners. WSS is not expected to contain significant debris which could cause blockage of the spiral heat exchangers and, therefore, is not screened.

SLUDGE SCREEN EVALUATION
Sludge Screen Technology
Although sludge screening is rare in North America, one of the most commonly used sludge screens in Europe is the in-line StrainPress, manufactured by Parkson Corporation. Among the available alternatives, the StrainPress, which both removes and dewaters screenings, was selected for detailed evaluation. Figure 2 shows a diagram of the screen.

Sludge is pumped into the screen. Pumping and an internal screw auger transport the sludge through the screen, and squeeze sludge through a perforated screen basket which surrounds the auger. The screw auger compacts the remaining debris as it carries the debris to the screenings discharge end of the unit. Here, a spring-loaded end plug provides pressure for compaction. Compacted screenings force the end plug open, and fall through an enclosed chute to a receptacle.

The StrainPress housing, screw, and screen basket are each constructed in two pieces and bolted together, allowing the unit to be taken apart for maintenance of the unit. The StrainPress housing provides total enclosure of influent sludge and screened sludge for the entire screening process; the discharge chute provides total enclosure screenings. This design minimizes potential odours.

A moisture probe is mounted on the screenings discharge end of the unit. Detection of excessive moisture signifies that liquid sludge is being carried out with the screenings, and that the screen basket may be plugged. Pressure gauges are also provided on the sludge inlet and screened sludge outlet for unit monitoring; high pressure indicates potential plugging.

Parkson Corporation offers the StrainPress in three models with capacities of 4.2, 8.3 and 16.6 L/s. Screen openings of 5 mm and 10 mm are available.

As part of the evaluation of the StrainPress, a number of plants in Europe were contacted to determine their success with the unit. The four plants contacted had a total of seven units in service for a period between 18 months and 3 years. The units screened primary sludge or blended primary and secondary sludge with and without primary scum. All operators praised the unit for its reliability, reporting no significant blockages of the units, and recommended it highly.

Pilot Tests
After discussion with references, pilot tests were conducted in 1992 at the Lulu Island plant using the StrainPress model SP3, initially with a 5 mm screen. This unit has a rated hydraulic capacity of 8.3 L/s. Thickened primary sludge and scum with a combined dry solids content of approximately 5 percent were pumped through the unit at rates as high as 16.5 L/s, about twice the rated capacity, because it was not possible to control sludge pumping rates to the screen during pilot testing. The screen effectively and reliably removed debris from the sludge/scum mixture at the pumped rate.

Pressure drop across the screen during the tests was about 25 to 30 percent of the manufacturer's recommended maximum allowable. The amount of debris removed in this initial trial was about 4.5 percent of the total dry solids. The screen compacted debris to a dry solids content of approximately 40 percent. Screened sludge was homogeneous and free of deleterious materials. It should be noted that, at the time of the tests, the Lulu Island plant did not have any influent screening or other upstream debris removal process. Influent screens are being added to the plant as part of the secondary expansion.

During initial pilot tests, it was suggested that using a screen with 10 mm openings would lower the screenings generation rate, while providing the required degree of blockage protection for the spiral heat exchangers. The tests were repeated with both 5 mm and 10 mm screens. The objective of the tests was to quantify the mass of screenings removed from primary sludge by the two screens. Tests showed that there was little difference in the quantity of debris removed between the screens. The percentage of total solids removed by the 10 mm screen was 5.9 percent and by the 5 mm screen was 6.0 percent.

Based on the pilot tests, the StrainPress with a 10 mm sieve was selected for primary sludge screening at the Annacis Island plant.

Assuming that screenings will be compacted to a dry weight solids content of 40%, the average mass of wet screenings produced by the StrainPress at the Annacis Island plant was estimated at 4000 kg/d.

SCUM SCREEN EVALUATION
The manufacturer of the sludge screen did not recommend its use for scum screening due to concerns with screen blinding. A number of alternative scum screening technologies were evaluated before selecting the Rotostep manufactured by Hycor Corporation for pilot testing. The Rotostep, which resembles an escalator in appearance, was tested because of its self-cleaning abilities, a particular benefit with scum due to expected build-up of grease on the screen surfaces. The Rotostep screen has two sets of step-shaped, alternating, vertical plates, one fixed and the other movable. The movable plates are driven in a vertical, circular motion which cleans the plates as it moves captured screenings up the fixed steps. When the screenings reach the top of the fixed plates, they fall onto an inclined chute and into a hopper.

This in-channel step screen was available in 3 mm and 6 mm screen openings and with a clean water hydraulic capacity ranging from 87 to 540 L/s. The step screen performed well at screening primary scum during pilot tests. However, pursuant to testing, Hycor stopped manufacturing this screen.

As a result, alternate scum screening technologies were re-evaluated. Rotary drum screens, which are available from a number of manufacturers, have been used for screening scum. Rotary drum screens were originally rated lower in our evaluation when compared to the step screen due to the potential problems of screen blinding at high grease loadings and of removing scum with the screenings, thus reducing the amount of scum going to the digesters.

Review of scum screening operations at the District's Lions Gate Wastewater Treatment Plant showed that rotary screens could be used successfully for screening scum. At the Lions Gate plant, primary scum is being screened through two externally-fed rotary drum screens: a coarse screen with 10 mm openings and a fine screen with 0.75 mm openings. The intent of the two-stage process is to remove inorganic screenings with the 10 mm screen and other organics with the 0.75 mm screen. Coarse screenings are trucked to landfill. Fine screenings are fed to the digester. Effluent from the screen is returned to the plant headworks. A high pressure spray wash prevents screen blinding. Plant staff report that this screening operation has been successful.

On externally-fed drum screens, solids can become lodged between the doctor blade and the screen, allowing screenings to press through the gap and fall into the effluent stream. Therefore, screenings could carry-over and clog the heat exchangers. Internally-fed rotary drum screens were believed to provide similar performance while eliminating this carry-over, and were therefore selected for detailed evaluation.

A number of plants and manufacturers were contacted to determine their success with internally-fed drum screens. In general, the plants contacted were screening a mixed solids stream with high grease loading (not specifically primary scum). Operators reported no problems with screen blinding and minimum maintenance on their units. At the Renton wastewater treatment plant near Seattle, screens with 1.5 mm screen openings operated without blinding, even though the external and internal sprays were not being used.

As part of a separate study, the District had conducted a successful pilot test of an internally-fed screen at the Lulu Island plant. They were, therefore, confident of the selection of this screen for this application without further pilot testing.

Figure 3 shows an internally-fed rotary drum screen. Scum is pumped into the screen through a T-shaped inlet diffuser in the interior of the drum screen. The screen slowly rotates and scum passes through the wedge-wire screen onto an integral tray and flows to an outlet pipe from where it is pumped to the thickening process prior to digestion. Debris is retained on the internal surface of the screen. As the drum rotates, flights attached to the internal surface help mix and dewater the screenings, and convey them to a discharge chute at the end of the drum. A stainless steel housing encloses the entire unit, which helps to minimize over-spray and odours. The unit includes external and internal spray wash systems for screen cleaning.

OPERATIONAL RESULTS
Two Parkson StrainPress units for screening primary sludge and one internally-fed rotary drum screen for screening scum have been in operation at the Annacis Island plant since April 1996.

Sludge Screens
Table 2 shows the performance of the Parkson StrainPress sludge screens compared to pilot test results. The screens have performed well at flows up to 35 L/s, 40% more than their rated capacity. Screens reduce the percent solids in the screened sludge by approximately 10% compared to unscreened sludge. Screened sludge is homogeneous and free of debris. The screens operate as much as 18 hours/day and remove 20 to 25 m³/d of screenings. Screenings have about 40% dry solids content.

The screens have successfully reduced the debris entering the anaerobic digesters. When plant operators drained the anaerobic digesters for cleaning after six months of operation, they found no debris.

Operators at the Annacis Island plant report they are very satisfied with the operation of the sludge screens. The screens have performed reliably with no major problems. The only improvement that operators suggest for the Parkson StrainPress is the capability to connect a hot water backwash; grease builds up on the screens, which have to be taken apart to clean.

Table 2. Performance of Sludge Screens
Parameter	Pilot Performance at Lulu Island WWTP	Full-scale Performance at Annacis Island WWTP
Screen opening, mm	10	10
Rated screen capacity, L/s	10	25
Feed rate, L/s	20	20 - 35
% solids, unscreened sludge	4.35	5.1
% solids, screened sludge	4.15	4.6
% solids, screenings	39	37
% solids removal	4.6	10

Scum Screens
Table 3 shows the performance of the internally-fed rotary drum screen. The screen operated at 17 L/s, approximately 80% of its rated capacity. Initially, the screen performed well and plant operators were satisfied with its operation. The screen removed approximately 1 m³/d of screenings with 26% dry solids.

Operators were concerned about the quantity of spray water carried over with the screenings. They subsequently reduced the volume of water carry-over by operating only one of the two spray wash headers. When the screen blinded, both spray wash headers were turned on to clear plugging.

Over time, the screen began to have problems with hair-pinning. Hair became enmeshed in the wedge wire and expanded mesh backing which supports the wedge wire on this particular screen; grease attached to the hair, and the screen clogged. The spray wash system could not clear the clogging, and wash water and scum would carry over into the screenings.

To clear the screen, operators had to remove the screen housing and hose the screen with a high pressure wash. The high pressure wash effectively removed the hair and grease; however, the required maintenance was labour intensive and frequent; the screen would perform well for only a few hours after cleaning before water began to carry-over into the screenings. The District decided against installing a high pressure or hot spray wash to clean the screen. Rather, the District decided to test scum screening using the Parkson StrainPress.

Scum was diverted to the standby sludge screen. The discharge cone tension on the screen was adjusted to suit scum screening. The StrainPress has operated as a scum screen for two months, and has performed effectively and reliably, with little required maintenance. Scum screenings have similar dry solids to sludge screenings. As a result of this testing, the District has purchased an additional StrainPress for scum screening.

Table 3. Performance of Internally-fed Drum Screen for Scum Screening
Parameter	Value
Screen opening, mm	10
Rated screen capacity, L/s	22
Feed rate, L/s	17
Screenings generated, m3/d	1
% solids of screenings	26

CONCLUSIONS

1. Debris removal from wastewater solids streams can effectively be used to improve aesthetic acceptability of biosolids products and protect process equipment.
   
   
2. Pilot tests and full scale operational testing show that raw primary sludge and scum can be cost effectively and reliably screened to remove identifiable debris.

ACKNOWLEDGEMENTS
This work was conducted by the Greater Vancouver Regional District staff and ABR Consultants, a joint venture between Associated Engineering (B.C.) Ltd., Brown and Caldwell Consultants Canada Ltd., and Reid Crowther and Partners Ltd.