There's lots of interest nowadays in increased waste diversion via wet/dry systems such as the one operated by the City of Guelph, Ontario. The diversion rate (more than 50 per cent) is attractive, bu...
There’s lots of interest nowadays in increased waste diversion via wet/dry systems such as the one operated by the City of Guelph, Ontario. The diversion rate (more than 50 per cent) is attractive, but not the perceived high cost.
With funding from First Brands (Canada) Corporation, Proctor & Redfern Limited undertook an exercise to devise a system that would cost-effectively recycle the “dry” side. Meetings were held with representatives from facilities in Guelph and the County of Northumberland (which also operates the dry MRF component of a wet/dry system). They were asked a question that many municipalities must currently ponder: “If you had a clean sheet of paper to design a dry stream facility, what would you keep and what would you change from your current process?”
This led to the facility described and illustrated herein that is specifically designed to handle a mixed dry material stream such as is handled by the Guelph and Northumberland facilities, but with various improvements incorporated.
Guelph has a population of 95,000 and its facility–a sort of prototype in Canada–was built at a cost of approximately $36-million. This can be broken down into: $24-million for the building, construction and equipment capital; $12-million for the land purchase (at $1-million), permits, approvals, studies, and conceptual design. The capital cost for the dry MRF was $12.1-million.
Northumberland, with a population of 75,000, built the second dry material processing facility in Ontario for approximately $6-million (the County has not yet built the wet facility).
Outside of comparisons with these facilities, design emphasis has been placed on overcoming the common problems of MRF configuration, equipment, and sorting/sorter problems described in Table 1.
Facility description and throughput
Certain “bells and whistles” of the prototype Guelph facility (e.g., ring room for glass, viewing galleries) were removed to help increase throughput and lower costs. For practicality, the proposed optimal facility employs equipment currently in use in recycling facilities in Canada. Recognizing that incoming dry material can be difficult to sort, it strives to balance mechanical and manual sorting.
The building is approximately 10,400m2 in size (about twice the size of the Northumberland facility). The MRF uses three in-feed lines to process more than 176,000 tonnes of residential and IC&I waste annually. In all instances, throughput is based on a sorting conveyor belt speed of 1,800 cm/minute seven days per week (with 15 effective sorting hours per day). The facility does not have to operate with unsustainable belt speeds to achieve required sorting throughput rates. (The dry facility in Guelph has a throughput capacity of 91,000 tonnes per year and Northumberland 45,000 tonnes per year.)
The main line of the facility processes residential waste: more than 109,000 tonnes per year (t/y) of mixed fibres and containers, bagged or loose. A separate line was added to address the differences between residential and IC&I materials. The line also acts as a secondary sort line for residential fibres. (The Guelph facility operators stated that they wished they had that feature, which is part of the Northumberland system.) The line is designed to process approximately 28,000 t/y.
For flexibility, the proposed facility also has a separate in-feed line for commingled containers as would be found in a typical Blue Box program or from shopping centres and schools. The line is designed to handle about 39,000 t/y.
A key to successful dry stream processing is the separation flats from rounds (fibres from containers). In Guelph, a single Ballistic Separator (BaSep)–a series of variable pitch, shaking screens inside a controlled access box–is used. Because of the BaSep’s high maintenance costs, a search was done to determine what other flats/rounds separation technologies were available.
A primary sort technology chosen was one where, after the incoming stream has been cleaned up and all materials debagged, containers and fibres are moved along a separator which uses rows of “stars” that act as a screen. Fibres and bags of film pass over the top of the screen. The Mach I Separator, designed by Machinex Industries, is rated to process about 20 tonnes per hour at 96 per cent efficiency for fibres and 85 per cent for containers. (For example, fibres have four per cent containers; the containers have 15 per cent fibres.)
In Northumberland County, a Single-Bounce Adhesion Separator (BAS) is used to separate the flats and rounds. Designed as an inclined belt with a series of cams “thumping” the underside, the flat materials stay on the belt while the round materials fall off the side and move in the opposite direction.
Some fibres remain mixed in with the containers in the materials that drop out of the bottom of the Mach I. To improve capture rates, a BAS (similar to the system used by Northumberland) is used in the design as a secondary sort system. The two sorting technologies combined are expected to help better prepare the material for sorting.
Ergonomics are a concern in any process where workers do repetitive motions. In the proposed MRF design, no sorter would reach more than 91 cm. This will help reduce stress to the lower back and shoulders. Also, all conveyors are set at such a height above the floor of the sorting platform that reaching forward to grab materials is not hard on hips.
Fibre sorters drop material into chutes located immediately to each side of their station. This helps eliminate unnecessary movement and should reduce the incidence of repetitive-strain injuries. Containers are sorted by throwing the materials forward into bunkers.
There are a number of design components that improve materials management. (See chart on page 16.)
Two equipment manufacturing companies estimated the conceptual engineering capital cost for the building and equipment installed (not including land) at approximately $18- to $20-million (including 50 per cent engineering and contingencies).
The cost of the Guelph dry facility totalled $12.1-million and produces a cost per throughput tonne of $133 ($12.1-million/91,000 tonnes throughput). The cost of the Northumberland dry facility totalled approximately $6-million, producing a cost per throughput tonne of $133 ($6-million/45,000 tonnes throughput). The facility outlined in this article, assuming a capital cost of $18-million, would result in a cost per throughput tonne of $102 (based on 176,000 tonnes throughput). This is approximately 23 per cent lower than the per-throughput-tonne cost for the Guelph facility.
Daniel Lantz, B.Sc., M.Sc. is senior project consu nt at Proctor & Redfern Limited in Toronto, Ontario.
Space does not allow a complete description of all the features and methodologies used in the design. Further information is available from Procter & Redfern Limited.
MRF equipment and configuration
In-feed belt angle too high.
Plastic film management – removal of
bags and storage of film.
Sorting screen placed in wrong location to
Not removing the materials in a high volume
to low volume order to improve down-line sorting rates.
Sorting conveyor belt speed too fast
Sorting efficiencies creating need to re-sort residues.
to allow proper and maximized sorting rates.
Material movement creating double-handling
Picking ergonomics causing worker strain
and slow sorting rates.
Material storage causing cross-contamination
Chutes too small and/or back-splashes not present to
on tipping floor or in work-in-progress area.
deflect material to the bunkers, so materials end up on
Cost-Efficient Automation Sorting
Every business owner knows that one of the greatest waste management business expenses is labour. In the past, numerous workers picked material from a large pile in the middle of the tipping floor. Today, owners have the opportunity to employ advanced sorting systems that automatically stream materials such as commercial OCC, curbside news, and even single dry streams. (Single stream waste is generally defined as all household recyclable materials mixed together–small OCC, news, junk mail, bottles, cans, and glass.)
Most of the advances developed by leading edge companies such as Van Dyk Baler Corp., based in Stamford, Connecticut, have been specifically designed to reduce the number of workers required to process material.
For instance, one of Van Dyk’s developments, the Lubo Starscreen sorter, uses rubber-composite stars and spinning axles to automatically sort materials. Commercial loads of OCC were the first of these materials. The sorter allows the large OCC to pass over, while small fibers fall through. Only two laborers are used to presort trash and stringy materials.
Single streams can be a challenge, but even broken glass can be removed from co-mingled streams, and dirt and grit from office material. The biggest advantage to processing material as a single stream is the ability of haulers to collect material quickly and with the same trucks used to collect general refuse. The problems associated with single stream materials have always been the need for a huge number of sorters to separate the material and the concern of getting broken glass in with the fiber. By utilizing multiple types of the Lubo sorters, both of these issues can be addressed.
The sorter is able to process up to over 20 tonnes of single stream material per hour with less than 20 pickers.
Baler technology has also advanced. Workers are no longer needed to run the balers–photo eyes and programmable logic controllers allow Van Dyk’s Bollegraaf balers, for instance, to process up to 90 tonnes per hour, without almost any human intervention. The pre-press flap system doesn’t have to be relined. It also uses smaller, more efficient hydraulic systems that allow for comparable bale weights and densities. The information and operating system on the baler allows a processor to monitor bale weights and amounts processed. (And, it’s currently being upgraded to include tracking for individually bar-coded bales).
Written by Peter van Dyk, president of Van Dyk Baler Corp. in Stamford, Connecticut.