PET Food Processing - December 2020 - 42

EQUIPMENT

Left: Today's smart
detection systems can
alert operators and
self-diagnose to ensure
safety program parameters are being met.
Eriez Magnetics

Right: In any food processing environment,
inspection equipment
must be designed to
withstand washdown.
Mettler Toledo

Spurgeon explains smart systems accomplish this in
several ways including a verification check that signals an
alarm when the detector needs to be tested; self-diagnostics
that alert operations if there are issues with the equipment;
integration with information systems (IS) for data pushes
for record keeping and system monitoring; programmable
logic controller (PLC) input enabling the detector to take
commands for product changes; and reject confirmation
that ensures off-quality product is indeed rejected.
" Additionally, the pet food industry has adopted many
of the safety standards used by food companies and now,
many producers use metal detectors in their HACCP program, " Spurgeon said. " Therefore, plant personnel need to
be more attuned to the proper setup and operation of metal
detectors to be certain they are meeting their food safety
program requirements. "
Eric Garr, regional sales manager, Fortress Technology,
Toronto, Canada, said testing of in-line inspection usually
entails dropping test samples into the product flow, which
can be difficult to get consistent results, plus that introduces
contamination into the line. Therefore, Fortress has recently introduced Halo, an auto-touch system that helps measure the performance of the metal detector without having
to insert metal into the system.
" It works using internal interference that it creates, which
in turn creates a disturbance exactly like metal, so it alleviates the need for those test samples, " Garr explained. " It's a
huge improvement on the quality of test performance. "

Employing optics
TOMRA Foods, Leuven, Belgium, has launched a pet food
application that sorts kibble and guards against contamina-

" The goal is to find foreign
matter - including bones -
prior to fine grinding. "
Lanel Menezes, Mettler Toledo

PET FOOD PROCESSING | December 2020 | www.petfoodprocessing.net

tion by employing optical sensors, noted Lars Povlsen, global
sales manager, proteins, TOMRA Foods' pet food division.
These sensors differentiate between good product and foreign material or cross-contamination based on color, structure or composition. Air-valves on the platforms remove undesired material with precise and powerful air bursts, keeping the processor's yield to an absolute maximum.
" It is a new application we expect to see grow strongly in
the next couple of years, " Povlsen said.
John Lawrence, chief executive officer, Sightline Process
Control, Ontario, Canada, said the most important hardware improvements he has seen are related to advancements in camera technology.
" Camera resolution, processing speed, and frame rates
have come a long way in the past five years allowing us to
analyze more details and smaller objects with precision and
to process images at up to 100 objects per second, " he said.
" With more image capture options, our list of applications
has grown to include products like kibble and pet treats. "
Additionally, a few of the more desired measurements
possible for products prior to packaging are color, shape,
length and width, volume, counts, and detection of broken
pieces. Important post-packaging inspections include bag
seal inspection and verification of packaging elements such
as bar/lot/date codes.
" A prime example is a treat simulating a bone with center
'marrow,' " Lawrence said. " With a vision system, any geometric measurements can be extracted, plus the system can
detect if the center 'marrow' is present and if it meets specifications for color and volume. "
Lawrence described Sightline systems as non-contact
inspection systems that use a series of high-resolution/
high-speed cameras along with imaging software to detect and measure random objects as they move along a
conveyor belt.
" Virtually any size, shape or color attribute can be extracted from the object images and converted into standard
units, providing real-time 100% inspection of any production line, " he said. " Products can be configured with any
combination of measurements from the included library of
over 100+ pre-loaded algorithms. "

http://www.petfoodprocessing.net

PET Food Processing - December 2020

Table of Contents for the Digital Edition of PET Food Processing - December 2020