Metal Detection Guide

From your Metal Detection Experts at

What’s Inside?
◊ Metal Detection Principals
◊ The 4 “P”s of Metal Detection
◊ Testing Metal Detector
Performance
◊ Sources of Interference
◊ Applications
◊ Good Manufacturing Practices

Understanding the Basics of

Metal Detection Systems.
phone: 414-672-0553 | www.adsdetection.com
 Table of Contents

Home– Table of Contents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Detector Performance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2 Metal Detection– The Basic Principles. . . . . . . . . . . . . . . . . . . . 2 Reliable Reject Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Why Use Metal Detection?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Reject Device Testing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 11
Theory of Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Other Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Balanced Coil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Gravity Drop/Drop-Through Metal Detectors . . . . . . . . . . 11
Balance Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Design Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Ferrous in the Foil. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Determine Internal Pipe Dimension. . . . . . . . . . . . . . . . 11
3 Product Effect and Phasing. . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Round versus Rectangular Pipes. . . . . . . . . . . . . . . . . . . . 11
Product Effect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Overall System Length. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 12
“Phasing Out” Product Effect. . . . . . . . . . . . . . . . . . . . . . . 3 Testing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4 Metal Free Zone. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Automatic Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Sensitivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Other Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Types of Metal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Static. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5 Ferrous: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Pipeline Metal Detectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 13
Non-ferrous:. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Design Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Non-magnetic stainless steel:. . . . . . . . . . . . . . . . . . . . . . . 5 Determine System Length. . . . . . . . . . . . . . . . . . . . . . . . . . 13
Shapes and Orientation of Metal. . . . . . . . . . . . . . . . . . . . . . . 5 Selecting the Valve Style. . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Setup and Multiple Products . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Selecting the Non-Metallic Pipe . . . . . . . . . . . . . . . . . . . . 13
6 The Four “P”s of Metal Detection . . . . . . . . . . . . . . . . . . . . . . . . . 6 Testing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Preventing Contamination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Automatic Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 14
Profitable Production. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Inverse Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Protecting Customers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Good Manufacturing Practice – Guidelines. . . . . . . . . . . . . . . 14
Promoting Brand Confidence. . . . . . . . . . . . . . . . . . . . . . . . . 6 Prevention. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
7 Testing Metal Detector Performance. . . . . . . . . . . . . . . . . . . . . 7 Testing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 15
Metal Test Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Establishing operating sensitivity performance. . . . . 15
8 Test Frequency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Documenting the sensitivity standard. . . . . . . . . . . . . . 15
Test Procedure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Number of tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Test Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Rejected Product Handling. . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Automatic Test Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Documenting the Program. . . . . . . . . . . . . . . . . . . . . . . . . 15
9 Sources of Interference. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Minimal Requirements and Verification . . . . . . . . . . . . 15
Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Appendix List. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 16
Conveyor System– End of Line. . . . . . . . . . . . . . . . . . . . . . . . . 9 Appendix A – Typical Processing Line. . . . . . . . . . . . . . . . . . 16
Design Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Appendix B – Reject Devices. . . . . . . . . . . . . . . . . . . . . . . . . . . 17 17
10 Sanitary Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

1
 Metal Detection: the Basic Principles

Why Use Metal Detection? Signals from the receiving coils are connected in
opposition to each other and when no disturbance is
Metal detectors are protection. They protect present, the net signal across the coil is zero. They are
processing equipment, the products people consume, balanced. It is the electrical equivalent of a balanced
your companies brand image, and liability due to
contaminated product.

Metal detectors are both effective and inexpensive.

Processing and packaging equipment, such as crushers,
extruders, shredders, cutters, grinders, choppers, mixers
and the like are often expensive and susceptible to
damage from metal contaminants. They can be expensive
to repair and the price of downtime adds up quickly.

*See information about metal detector

placement in Appendix A.

Although 75-80% of metal detectors are used in food

related industries. They are not the only industries
who find the protection they offer useful. Aggregate,
mining, rubber, plastic,wood, textile, glass , product and weight scale.
environmental security companies rely them as well. This balance is disturbed as metal contamination enters
the aperture. The control electronics amplify and analyze
Metal Detector Usage Can be broken into three main the disturbance, then indicate a “detection” if the signal is
categories: greater than an established sensitivity threshold.

◊ Packaging– where higher sensitivity to the smallest Balance Stability

piece of metal and product purity is desired.
Very small movements in the mechanical construction
◊ B
 ulk Processing– where gravity drop or pipeline metal of the metal detector can also disturb the balance of the
detectors are used to defend processing equipment. three coils. These movements can cause the detector to
indicate that metal is present when it is not (Called “False
◊ Industrial– where industries (mining, aggregate Triggering”), or they can cause the loop to remain out
plastic, rubber, lumber, etc.) have lower sensitivity of balance and continuously “false trigger” and require
requirements however, processing equipment must re-balancing. Metal detectors which require or are prone
be protected from damage and downtime caused by to vibration provide little value to automated production
larger tramp metal. lines.
Movement can be caused by expansion of the detector
Theory of Operation from changes in temperature, vibration from nearby
equipment causing the detector to vibrate, instability in
the detector’s construction, and other causes. Robust
Balanced Coil construction, as well as good electronics, can prevent such
All metals are either magnetically conductive, electrically movement and make the detector more sensitive.
conductive or both. When they enter an electro-magnetic Good electronics design includes automatic balance
field they create a detectable disturbance or “signal”. control as part of its standard features. Good mechanical
Modern metal detectors operate on the balanced coil, full design includes enhanced potting techniques to minimize
loop system which allows detection of this signal. false triggering. Together they combine to increase
reliability and effective sensitivity.
A balanced coil consists of three coils looping inside a
static shielded head that has a built in aperture through Ferrous in Foil
which product passes. The head houses a transmitter coil
The coil balanced, full loop design does not work well
that broadcasts a radio frequency signal and generates an
when a product is packaged in foil (pure aluminum). Foil
electro-magnetic field. On either side of the transmitter
lids and trays are common examples where an alternative
are two equidistant receiver coils.
system is required. In these situations, the detector uses
a series of magnets and is referred to as a “Ferrous in Foil”
detector. Non-ferrous and stainless steel contaminants

2
Product Effect and Phasing
cannot be detected with this type of detector. Not only Product effect may change as characteristics such
Not only are all metals are either magnetically and/or as temperature and salinity change, but product
electrically conductive, but inspected products can also aging may also affect it. Product definition may differ
have one or both of these characteristics. Metal detectors between the metal detector and the people producing
“see” two signals when contaminated product passes it. For example, whole, frozen cheesecakes have a
through the aperture. One signal comes from the metal different product effect than sliced, frozen cheesecakes
which contaminates the product. The second signal of the same ingredients. Those same cheesecakes
comes from the product. The signal from the metal must may have different a product effect immediately after
be greater than the signal from the product to trigger a they emerge from a freezing tunnel than they do after
detection. sitting twenty minutes on a conveyor line.

Product Effect “Phasing Out” Product Effect

Metal detectors measure electrical conductivity and At this point, we have signals from product, large
magnetic permeability. Many inspected products pieces of metal, small pieces of metal, and vibration
inherently have one or both of these characteristics. For all vying for the attention of the detector. We also
example, iron-enriched products (e.g. cereals) create a have varying product signal because each product is
large magnetic signal that hampers the detector’s ability not exactly the same as the last. To identify a metal
to detect the magnetic signal of small pieces of metal, contaminant, the detector must remove or reduce this
and are commonly called “dry” products. Conversely, “product effect.”
products with high moisture and salt content (e.g. bread,
meat, cheese, etc.) are electrically conductive and produce An unfortunate “fix” commonly used is to reduce the
a conductive error signal, and are commonly called “wet” sensitivity of the detector. Although this makes the
products. The table below shows typical product error signals from the product smaller, but also makes the
signals and categorizes them as conductive or non- signals from the metal contaminant smaller. This can
conductive. make practical detection of metal almost impossible.
Signals from ferrous metal are larger than signals from
Typical Conductive Products: the same size piece of non-ferrous or stainless metal.
Signals caused by vibration are always along the
◊ Food: m
 eat, cheese, bread and bakery products, fish, same lines as non-ferrous metals. Special circuits can
dairy products, salads be used to amplify the signals by differing amounts,
according to phase, to improve sensitivity of the
◊ Packaging: metalized films
metal detector to the stainless signal, and reduce the
◊ Other: plastic and rubber products with high carbon sensitivity to vibration.
black content
The detector can be adjusted to “rotate” the signals
Typical Conductive Products: from the product in a way that maps out an area
where the product is expected to be. This allows
◊ Food: c
 ereal, crackers, flour and powders, biscuits, the product to be “ignored” and emphasizes
frozen food products (<10oC), peanut butter analysis of signals outside that area to identify metal
and margarine (vegetable oil is not conductive) contaminant. The technique to create this “detection
◊ Other: wood products, plastics and rubber (products envelope” is called “product compensation” or
with high carbon black content may be “phasing-out” the product effect.
considered conductive), textiles, paper
products As noted above, product effect is caused by various
factors and does not always generate the exact same
signal. The “detection envelope” must be large enough
so a product signal is normally within the envelope.
Metal contaminate generally produces a different
signal which would not appear within the “detection
envelope,” and the detector would trigger for product
reject when it sees the signal.

The mask of the “detection envelope” focuses analysis

on signals outside of where product is expected to be.
Some small pieces of metal will not provide a large
enough signal to be detected outside that envelope
and will pass through the detector without rejection.
These pieces are usually smaller than the size specified
for detection. Some product will have a signal greater
than what is expected and will be rejected even

3
though no metal is present because of variation in the
production process. Some product with no metal may be
Metal Free Zone
rejected because vibration in the system creates a signal The electro-magnetic field is created inside the detector’s
seen outside the envelope. enclosure; however, some field emanates out of the
aperture on both sides and forms the metal free zone
Product testing and accurate specification of contaminant (“MFZ”). Generally, the practical size of the leakage is 1½
are essential to determine the detector’s sensitivity and times the (smaller) aperture dimension, and no metal
product effect. This service is usually available from the should be allowed in this area. Large moving metal
metal detector manufacturer. should be kept 2 times away. Special detectors are
Creating the “detection envelope” requires experience to available for applications which demand a substantially
achieve optimum performance. If several products are reduced MFZ.
running on one production line, adjusting the detector for
each product can be inconvenient and time-consuming.
Fortunately, this experience can be programmed into the Sensitivity
detector. In theory, aperture size determines the sensitivity of a
When a detector has an automatic product compensation given metal detector. Smaller apertures generally allow
feature, product which is known to have no metal smaller pieces of metal to be detected. The smaller
contamination is passed through the detector in its “learn” dimension of rectangular apertures is used to calculate
mode. The detector’s software analyzes the signals and the sensitivity, although the length also contributes.
calculates the “detection envelope” for that particular
product. The settings can be saved so an operator can The smallest size aperture should be selected to
recall those settings the next time the product is run. maximize the sensitivity of a detector, however, there are
some exceptions including metalized film and highly
conductive product (e.g. large blocks of cheese).
Automatic product compensation
removes the effect of product variations
that reduce sensitivity The center of the aperture is the least
sensitive point of the metal detector.

*Diagram of Metal Free Zone

Sensitivity is also affected by the position of the
contaminant in the aperture. The least sensitive point is
the centerline axis of the aperture (1 in diagram), and this
is where to test performance. As metal gets closer to the
sides (and the coils), the signal it generates gets larger,
making it easier to detect (2 and 3 in diagram).

Regular testing of the detector should ensure the test

sphere passes close to the center of the aperture. If this is
not practical, a consistent position should be used so test
results will be consistent.

1
3 2

Types of Metal
Metal detector sensitivity is not the same for all types of
metal. The ease of detection depends on how easily they
are magnetized (the magnetic permeability), and the
electrical conductivity of the metal. Metal detectors are
calibrated for sensitivity to ferrous metals, non-ferrous
ProScan Pipeline Head ProScan Gravity Drop Head
metals, and stainless steel.

4
Ferrous: Shapes and Orientation of Metal
Ferrous materials are any metal easily attracted to a
Metal detection standards are based on spheres because
magnet (e.g. steel, iron, etc.). Typically, ferrous metals
spheres are the same shape from any angle. Real
are easiest to detect and usually the most common
contaminants are rarely spherical and may produce
contaminant outside of food processing plants.
a different signal depending on its orientation when
passing through the detector. Wire contaminants are the
Non-ferrous: most dramatic example of this.
Non-ferrous materials are highly conductive non-
magnetic metals (e.g. copper, aluminum, brass, phosphor The signal produced by wire shapes varies greatly
bronze, etc.). When inspecting non-conductive products, depending on the type of metal and its orientation as it
these metals produce almost the same size signal as passes through the aperture. In its worst case, a wire may
ferrous metals because they are all good conductors. produce a signal no bigger than a sphere of the same size
When inspecting conductive products, increasing the test as the diameter of the wire.
sphere size by at least 50% is a good practice.
The following diagram shows that configuration “A”
provides the biggest signal (easiest to detect), in ferrous-
Non-magnetic stainless steel: type wires, while configurations “B” and “C” provide the
High quality 300 series stainless steels (e.g. Type 304, 316, smallest signal (most difficult to detect). Similarly, in non-
etc.) are the most difficult metals to detect due to their ferrous and stainless steel wires the opposite is true (i.e. “A”
poor electrical conductive qualities and, by definition, provides the smallest signal and “B” and “C” provide the
have low magnetic permeability. These are commonly biggest signal).
used metals in the food processing and pharmaceutical
industries.
Setup and Multiple Products
When inspecting non-conductive products, a stainless Detectors which are difficult or complex to set up are
steel test sphere typically needs to be 50% larger than a generally not programmed correctly. Well-designed
ferrous sphere to produce the same size signal. When detectors’ basic instructions will allow changes to
inspecting conductive products, a stainless steel test be made correctly and easily. Many detectors have
sphere needs to be 200% – 300% larger than a ferrous “automatic setup” features which could approach what
sphere to produce the same size signal. could be achieved by an experienced operator; however, if
improperly used there can be a significant decline in the
When detection specifications include non-ferrous and/or detectors performance. While having a different product
stainless steel, the particular metals, and sizes should be setting for each product may be appropriate, more
identified. Correct identification of what particles should operator choices can mean more opportunity for incorrect
be detected is critical because these metals have many selection (see the “Product Effect” example of cheesecake
varieties and they all look slightly different to the metal above). Grouping similar products under one product
detector. setting may provide easier and more accurate inspection
(e.g. 12 oz. rice patties with 8 oz. rice patties).

Product Flow

Contaminant
Orientation
A=

B= C=

* Diagram of contaminant orientation

5
 The Four “P”s of Metal Detection

Preventing Contamination Promoting Brand Confidence

Metal in product can be a cause of consumer complaints Branding not only helps maximize sales, but also
even with metal detection systems in operation. The generates customer perception of product safety and
complaints (which usually do not include visual items such quality, and drives repeat customer purchases. Product
as bolts, washers, and pieces of blades or screens found brands are valuable, important assets which must
in product) are usually due to lack of effective controls, be protected. Contaminated products reaching the
poor working methods, and incorrect specification (and consumer can have financial consequences by product
therefore design) of the system rather than actual failure recall, from potential litigation, and can damage a brand
of the metal detector. or reputation which has been built and consumed
significant company resources. Accurate documentation
Effective metal detection programs focus on minimizing
of compliance with production standards can help
contamination through good manufacturing practices
minimize damage from customer complaints.
(“GMP”), proper equipment selection, effective testing,
and ever-improving knowledge of how industry
standards, customer requirements, and legislation affect
manufacturers.

The four “Ps” describe why metal

detection is critical for processing or
packaging lines

Effective metal detection programs focus on minimizing

contamination through good manufacturing practices
(“GMP”), proper equipment selection, effective testing,
and ever-improving knowledge of how industry
standards, customer requirements, and legislation affect
manufacturers.

Profitable Production
The cost to implement and maintain an effective metal
detection program is significantly lower than the
potential cost of metal contaminate being found in the
product. Metal contaminated product found before
shipment may result in product and packaging waste,
possible machinery damage, and loss of output. Metal
contaminated product found after shipment can affect
customers, cause product recalls, create adverse publicity
toward your company or brand, create financial liabilities,
and potential litigation. An appropriately implanted
metal detection program will lead to reduced failure costs,
improved customer and consumer satisfaction, protects
profitability as well as your company or brand.

Protecting Customers
Manufacturers have a legal and moral obligation to their
customers to minimize contamination of their product
and ensure consistent quality. They are responsible
for taking steps to protect their product’s quality and
the welfare of the consumer. Today’s manufacturers
incessantly look for ways to eliminate contamination
in their products, yet processes or procedures, or
compliance with them, break down and product becomes
contaminated. Failure to do this can be costly.

6
 Testing Metal Detector Performance

When unwanted metal can be introduced into a product,

a metal detection system is a necessary control measure Metal Test Sample
and should be considered a Critical Control Point Historically, metal detectors have been tested with a
(“CCP”). Regardless of how sophisticated or reliable a ferrous and non-ferrous, and sometimes a stainless steel
metal detection system may be, it is only as good as the test sample.
frequency and thoroughness of testing and recording
programs supporting it. Testing and recording are The size of the test sample must be established so that it
essential components of any quality HACCP (Hazard can be reliably detected inside product passing through
Analysis and Critical Control Points) system. In the the detector, down its centerline – the least sensitive point.
absence of any industry standards for detector testing,
each company must establish their own test criteria. Every application will be different and samples should
be matched to each application and detector. Samples
The HACCP plan should be validated before too small for the application will cause unnecessary
implementation, re-validated annually, and verified test failures, and create frustrated test operators. If the
frequently. Validation is the assessment of its scientific sample is too large, the performance of the detector will
and technical aspects and that the hazards, critical not be accurately tested. Establish a realistic, repeatable
limits, monitoring, and corrective actions have been operating performance level using a selection of test
properly established. Re-validation is needed to ensure sample sizes. Then choose an appropriate test sample(s)
the plan continues to be valid after considering changes for testing each application type.
in processing or manufacturing methods. Verification
determines if the plan is adhered to and is being properly
administered. Verification can include review of the
application of methods, procedures, tests, and other Typical Guidelines for Sensitivity:
evaluations in addition to monitoring, to determine
compliance with the validated plan.
Non-
Conductive
Conductive
Identifying and rejecting contamination as early as Aperture Product
Product
possible within the manufacturing process is consistent Height
with good manufacturing practices (“GMP”), and is a Ferrous & Non-
Ferrous
HACCP pre-requisite. HACCP does not rely solely on Non-Ferrous Ferrous
product testing to ensure safety, but builds safety into the Up to 2 in.
entire manufacturing process. Contamination or hazards 1.0 mm 1.5 mm 2.0 mm
(50mm)
are reduced or prevented by relying on controls in the
Up to 5 in.
process. 1.5 mm 2.0 mm 2.5 mm
(125 mm)
When creating or changing your testing program, Up to 8 in.
2.0 mm 2.5 mm 3.0 mm
consider the following points: (200 mm)

7
Test Frequency
Management must decide the frequency of testing the
detector. Typically, detectors are tested after repair or
maintenance of the equipment, at a shift change, when ADVANCED DETECTION SYSTEMS®
“Metal Detection Experts”

product changes, or on an hourly basis. 4740 W. Electric Ave. Milwaukee, WI 53219 P: (414) 672-0553 www.adsdetection.com Engineered & Manufactured in USA

Metal Detector Verification

Test Procedure Pursuant to our service call on _________, we are please to verify the following test
results for your metal detector.
Make Model Serial # Aperture Size
 eep test procedures as simple as possible, but should
K
include the following: ______”H x ______”W

The following Certified Test Samples were verified and detected successfully three (3)
◊ T
 est samples should travel through the approximate times in succession:
centerline of the aperture which is the least sensitive _________mm Ferrous Metal Test Sample Certification #:___________
point. _________mm Non-Ferrous Metal Test Sample Certification #:___________

◊ T
 est samples should be placed within the product, if _________mm 316 Stainless Steel Metal Test Sample Certification #:___________

possible. Inspection
The following performance metrics checked to validate the metal detector is able to
◊ T
 est procedures must allow the reject device to reliably meet the requirements set forth by: _______________________________
activate to test the entire system. This can include: (Company Name)

*Stability. Environmental factors and uncontaminated product

• T
 esting with contaminant at leading and Pass Fail
does not trip the metal detector

trailing edge.
*Sensitivity Check. Passing test wands no less than three (3)
Pass Fail times, making sure there is proper detection and rejection of the
• Testing successive packs. metal contaminants.

• Testing alternate packs.

Customer Signature Date
◊ The results of the test must be recorded.
Verification Completed By Date

Test Records **Make 2(two) copies of this Metal Detector Verification form. Give 1 (one) copy to the customer and include the other
copy in the email to Advanced Detection Systems (see Verification Service Procedure document).
Advanced Detection Systems
A sample of a test record is included for information. The “Metal Detection Experts”
Identifying and Meeting Unique Metal

format is not important but should include:

Detection Needs for Over 30 Yrs.

◊ Line or detector identification

◊ The date and time of the test Metal detector verification documentation
◊ The sample used
◊ Identification of the operator
◊ The result of the test (pass, fail, etc.)
◊ Corrective action taken if the result was failure.

Automatic Test Systems

Automatic test systems have been developed to
compliment, and in some applications replace, manual
test procedures. The capability of these systems should
be reviewed carefully to ensure that the testing is relevant
and feasible for the application. When installed effectively,
they can offer considerable savings through reduced
labor and material waste associated with manual test
procedures when a facility has difficult to access gravity
drop or a high number of metal detectors in service.

8
 Sources of Interference
Environmental conditions may affect the performance
of the metal detector, particularly where high levels of
sensitivity are to be achieved. Wherever possible, the
detector should be positioned to avoid or minimize the
effect from such conditions. These can be generated by a
number of sources:

◊ A
 irborne electrical interference – static, radio, earth
loops, etc.
◊ Vibration – moving metal
◊ Temperature fluctuation – ovens, freezers, etc.

While the detector may be capable of filtering some of

this interference out through such features as “Automatic
Balance,” in many cases the only option is to reduce the
sensitivity level. This is an important consideration when
comparing the capabilities of detectors.

Applications
Conveyor System– End of Line Design Issues
End of line metal detection is generally the preferred The conveyor’s purpose is to move product through
option as at this point product has been packaged and the detector’s aperture and successfully reject
there is no further risk of contamination. This is not always contaminated product. To achieve optimum
feasible for many reasons including: performance from the system, several design issues
should be considered.
◊ Physical restrictions – no space available
◊ Packaging type/material – foil lids or trays The conveyor design must avoid interfering with the
detector in any way. Both the conveyor and reject
◊ Package size – too large for detection standard
device affect the performance of the metal detector.
◊ C
 ritical points – machinery protection or raw Eddy current loops (often called “ground loops”) and
ingredients static electricity build-up can occur unless precautions

Metal detector with Welded Conveyor System. Metal detector with Bolt-Together Conveyor System.
Ideal for Bulk Product or Wet Process Conditions Ideal for Packaged Product.

9
are implemented during the design phase. These can testing procedures, designing an easy-to-clean system,
interfere with the detector and reduce the sensitivity of and following regulatory guidelines are essential when
the unit. sanitary concerns apply.

Metal detectors create eddy currents loops by emitting Detector Performance

high frequency signals in the aperture. Those frequently
“spill over” to the surrounding conveyor construction. Manufacturers will generally supply recommendations for
They do not affect the detector if they remain constant. achieving the best performance from the metal detector.
They can briefly change if the conveyor has an Always apply these recommendations;
intermittent joint of variable resistance, and then the
eddy currents create a large interference signal, which the ◊ Isolated rollers – to prevent ground loops
detector can read. ◊ H
 igh quality belt – metal free, interlocked finger joints,
plastic modular belt
Eddy current loops are typically created by any metal-
to-metal contact (e.g. bolted items, pulley shafts and ◊ Low vibration and static-free area
bearings, chain drives and guards, reject supports and ◊ Adequate metal free zone.
metal clamps, etc.) Rusting joints or changes in bearing
lubrication can cause problems to increase over time.
Reliable performance is enhanced by fully-welded Reliable Reject Device
structures, correct metal free zones, and properly isolated The selected reject device must be suitable for:
rollers, pulleys, cross-structures, and detector head
mounts. Conveyor belting must be metal-free as well as ◊ Package size, weight, and shape
having appropriately contaminant-free joints. Avoid anti- ◊ Presentation pitch, line speed, and belt width.
static belting material because it can create signals which
reduce detector performance. The following table provides general guidelines as to
which reject device will be best suited to your application.
When these factors are not given adequate consideration The figures may vary slightly with the manufacturer.
during design, the common outcome is a gradual increase Diagrams of the reject devices are in Appendix B.
in false triggers and false rejects. This is often “fixed”
by reducing the detector’s sensitivity. This “fix” can Consider the following additions to ensure reliable
violate customers’ inspection requirements and be more rejection of contaminated product:
expensive than appropriate corrective action.
◊ Registration photo eye – to ensure correct timing

Sanitary Issues ◊ E
 nclosed area from detector to reject device – to avoid
pack removal
All metal detection systems’ design must consider the
operating environment and cleaning routines they ◊ L
 ockable bin – to ensure contaminated packs are
are likely to encounter. Eliminating dirt and bacterial quarantined
traps, sealing hollow sections, minimizing horizontal
surfaces, providing easy access for maintenance and

Type Suited for Max Wt Notes

Light consumer packs (i.e. biscuits, 2 lbs. Unsuitable for loose product, boxes, curved
Air blast
chocolate bars, etc.). High throughput. (1 kg.) surfaces, and some bagged product.

Medium to light packs. Medium 10 lbs. Product generally enters bin diagonally–
Divert arm
throughput. (5 kg.) must ensure it will fit!

15 lbs.
Push arm Medium packs. High throughput. Unsuitable for loose or fragile product.
(7 kg.)

Large bags or boxes, hand fed or bulk 50 lbs. Requires an operator to remove
Stop on detect
material. Slow throughput. (25 kg.) contaminated product.

Retracting
Small product in lines or irregular shape. 5 lbs. Dimensions are for whole line or batch of
pulley or
Medium throughput. (2 kg.) products.
Flop gate

10
Reject Device Testing
Area of aperture
Complete system functionality should be tested as well required (in2) = 0.024 X Flow rate (#/hr)
as ensuring the metal detector is functioning correctly.
This involves ensuring that the reject device is operating Bulk density (#/ft3)
correctly and contaminated product is correctly handled.
Example: A product with a flow rate
Reject device testing is generally proved by observing
of 20,000 #/hr and a bulk
contaminated product that is quarantined after:
density of 18 # /ft3 will require a
calculated pipe area of .0.024 X
◊ P
 lacing the test piece at the leading edge of the
20,000 / 18 = 26.7 in2.
product
◊ P
 lacing the test piece at the trailing edge of the
product
Round versus Rectangular Pipes
◊ Passing successive test packs
Round pipe will utilize the pipe area required for product
Passing alternate test packs. flow more efficiently than a square or rectangular pipe
and therefore the flow capacity of rectangular pipes
Other Considerations should be de-rated by at least 20%.

Additional fail-safe methods can be incorporated into

the system design, in addition to regular manual testing,
including:
◊ Reject confirmation/Bin full sensors
◊ Air pressure failure sensors
Non-Conductive
◊ Fault/Shutdown.
Static-Free Pipe

Gravity Drop/Drop-Through
Metal Detectors
On the surface, the gravity drop application of metal Gravity Drop
Metal Detector
detector is very simple. However, care must be taken Head
in the initial design to avoid having to make major
modifications after installation. Many design concerns
(and solutions) for systems with conveyors are applicable
for all systems. Consider all of the following steps when
designing a gravity drop detection system.
Support Stand
Design Issues
Gravity drop detection systems are ideally suited to
inspecting dry, free-flowing products such as grains,
flours, cereals, rice, nuts, sugar, plastic pellets and flakes, as
well as in vertical packaging applications, liquids, pastes,
and slurries. The product must remain free flowing Automatic
and not back up into any part of the system. Support Fast Acting
structures and reject devices can reduce metal detector Reject Valve
effectiveness if not designed properly.

Determine Internal Pipe Dimension

Existing piping often determine the pipe and detector
size. This formula is useful if you know the peak expected
flow rate of the product and its density:

11
Overall System Length Automatic Testing
After establishing the proper pipe size, the overall system Perform manual tests upon initial installation and at
length can be considered. The bigger the pipe’s internal reasonable time intervals afterward. An automatic test
diameter (“ID”) the longer the system must become. The system can offer both consistent and relevant testing, if
detector through dimension must increase as aperture correctly designed into the system.
size increases, the valve height will also increase due to
the increased stroke, and the required distance between Other Considerations
the valve and the detector must increase. The latter is due
to the larger valve taking more time to reach full divert Static
position and therefore it must be located further from the
detector. All falling dry powders and granules generate static
electricity. Some are more prone to this than others, and
The relationship between the valve response time, environmental conditions (lack of humidity) can increase
product free fall height, and system length is very static electricity levels. Consider the following measures to
important. Correct design relies on knowing the product reduce static damage and interference;
pipe size (see above), the valve response time (from
the manufacturer – bigger valves require more time), ◊ G
 round all metal near the detector system
and product free fall distance from initial drop to the (pipes, flanges, structural supports, etc.) to avoid
centerline of the detector. accumulating static electric charges.
◊ W
 rap conductive shields around plastic parts
If the free fall height is increased, the distance between (product tubes, etc.) to help dissipate static electric
the detector and reject device must be increased in charges. Grounding standard (non-conductive) will
order to maintain adequate time for the reject device to not eliminate static. Some conductive plastics are
respond. Small changes in the detector-to-reject device available for food use but may interfere with the
distance can substantially affect the maximum allowable detector.
free fall distance because the product accelerates at 32 ft/
sec2 as it falls. ◊ U
 se a major single point ground for the detector
itself (consult the manufacturer for their
For example, given a reject device response time of 40 recommendations). Detectors using remote power
ms and a detector-to-reject device distance of 10 in., the supplies may be more susceptible to damage.
calculated maximum free fall would be 19 in.. If the reject
◊ An ionizing anti-static device may also be useful.
device was moved 3 in. closer to the detector (to 7 in.), the
maximum free fall would be 14 in.
Excellent sensitivity and early warning of product
contamination can be obtained by gravity drop metal
Testing detectors through careful design, accurate information,
and consideration of critical parameters early in the
Gravity drop detectors can be very difficult to test. Testing
design phase of the system.
can be performed quickly and reliably if testing access and
recovery are designed into the system. The test procedure
must confirm the detector’s performance as well as the Metal Detector Head Support Stand
response of the reject device.

A test access port is required for test access. This port

allows the introduction of a test sample (a plastic ball with
a metal sample embedded) where the product free fall
starts. The port should allow the test sample to fall such
that the test sample speed matches the speed of the
product during production.
Product Tube
A test sample safety retrieval gate is required for test
recovery. This port should be inserted into the normal
product flow below the valve where “good” product
flows following testing. This allows for safe recovery if the
detector fails to detect the sample or the reject device fails Automatic
to perform properly. Reject Valve

Well-designed systems allow test samples to be quickly

inserted into product flow during testing and removed
from product flow after testing.

* Gravity Drop metal Detector System

12
Pipeline Metal Detectors For example, if a system uses a 2 in. pipe with a maximum
flow rate of 80 GPM, the average product speed will be
Pipeline applications involve the installation of a non- 10 ft/sec.. If the response time for a 2 in. valve is 0.25 sec.,
metallic pipe section that sends product through a metal the valve must be at least 2½ ft. from the detector, but,
detector for testing. Pipeline applications are used where to allow for laminar effects, increasing the distance to 3 ft.
the extra sensitivity of a small aperture outweighs the would be recommended.
benefit of final package inspection. An example would
be testing soup before it reaches a packaging line using Selecting the Valve Style
metal cans.
Product temperature and viscosity influence valve
Pipeline systems are ideally suited to inspecting liquid, selection. Some valves are best suited for low viscosity
slurries, or paste products that can be pumped through a products such as juices, etc.
pipe. Such products would include sauces, dairy products,
meat slurries, juices, etc Selecting the Non-Metallic Pipe
Pipe selection depends on the required connection style,
Design Issues product temperature and, especially, the expected pipe
Designing a successful pipeline detection system relies on pressure. Careful design is required to ensure the non-
many critical factors, including the following: metallic pipe will not receive added pressure from the
stainless steel piping.
◊ Internal diameter (“ID”) of pipe
◊ Pipe clamp connection style (tri-clamp, I-line, etc.) Testing
◊ Product flow rate (GPM) Pipeline detectors can be very difficult to test. Testing can
be performed quickly and reliably if testing access and
◊ Product viscosity recovery are designed into the system. The test procedure
◊ Product temperature range must confirm the detector’s performance as well as the
response of the reject device.
◊ Product pressure
A test access port is required for test access. This port
◊ E
 xpected cleanup procedures allows the introduction of a test sample (a plastic ball with
(wash down, pipe pig, etc.) a metal sample embedded) upstream from the detector.
The port should allow the test sample to travel such that
Determine System Length the test sample speed matches the speed of the product
during production.
The position of the reject valve relative to the detector
depends on the product speed. The distance between A test sample safety retrieval gate is required for test
the valve and the detector must be increased in direct recovery. This port should be inserted into the normal
proportion to the product speed and valve response product flow below the valve where “good” product
time, since the valve has a minimum divert response flows following testing. This allows for safe recovery if the
time. Estimated product speed is measured in gallons detector fails to detect the sample or the reject device fails
per minute (“GPM”). An appropriate safety margin must to perform properly. Installing a similar port on the reject
be added to the speed calibration, due to laminar flow output will allow easier recovery of rejected test samples.
characteristics of liquid in a pipe, and the system must Test samples can be quickly inserted into product flow
be rigorously tested at full production rates to ensure the during testing and removed from product flow after
valve can respond in time. For example, if a system uses testing in well-designed systems.

Metal Detector
Head
Metal Free
Product pipe

Reject Valve

13
Automatic Testing Inverse Detection
As with gravity drop applications, testing in pipeline Detecting metal can also mean ensuring something
systems does not allow placement of the test sample you want is in a package. Reversing the reject device
down the centerline of the detector’s aperture, the setup causes the system to reject product with no metal
least sensitive point. Consequently, test results may be in it while accepting metal-containing product. Care is
inconsistent. required to ensure that unwanted metal is not included
before or between the point the desired metal item is
Automatic test systems can perform short-interval testing added and the detector.
of the detector and valve reaction without operator
involvement. Reject devices can employ a valve position
switch, which returns a confirming signal to the detector,
which can then monitor, or log, the reject device response
time.

Good Manufacturing Practice – Guidelines

Every packaging and processing company must perform When metal detection is impractical to perform on the
a hazard analysis for each product it produces to assess finished package, alternative control systems must be
the risk of metal contamination. The analysis should created. When good product is manually removed from
identify potential sources of contamination and the the conveyor to be hand-packed, guarding should cover
types of metals involved. This information identifies the and extend along the conveyor and reject system to
needs around which the metal detection system must be prevent packers from taking packages before they pass
designed. the reject device.

A producer of biscuits, for example, may identify potential

risk in the following areas: Prevention
Personal effects and operational consumables present
◊ Raw materials arriving from grain processing plants a contamination risk where poor awareness and lack
◊ Combining raw materials together for mixing of work practices exist. Minimizing contamination risk
requires time spent on identifying potential risks, defining
◊ Cutting machines forming the biscuits good work practices, and providing proper equipment.
◊ Conveyors transporting product Continually implement, update, and communicate
operating policies to ensure everyone remains informed
◊ Packaging. and supportive of these practices. Extend these practices
back along the supply chain to vendors.
Contamination allowed to continue through the The following are some examples of what constitute
production process can pose a threat to processing good manufacturing practices (“GMP”). Other specific
equipment downstream and/or may break into smaller, control measures may be relevant to specific industries,
more difficult to detect, pieces. This can increase scrap companies and manufacturing processes, however,
costs because processing costs have been incurred on they effectively demonstrate risks which may easily be
product after it has been contaminated. overlooked:

Establishing a GMP program in a ◊ Cover product-holding containers at all times

processing environment minimizes the ◊ C
 over conveyor lines carrying open containers until
number of contaminants entering the they are closed or capped
production line. ◊ D
 o not use paper clips or staples on documents in
production areas
Metal detection should be part of the normal product
flow whenever possible to avoid potential confusion ◊ Do not use pins on notice boards in production areas
regarding what has been inspected, and prevent ◊ A
 llow no hair clips, watches, or jewelry in production
unintentional bypass of inspection processes. The end of areas
each production line should be considered a “CCP,” and
the optimum inspection point is usually immediately after ◊ Avoid protective clothing with outside pockets
final packaging, or as close as possible to final packaging ◊ U
 se only “metal detectable” pens, hairnets, hearing
in the normal product flow. protection, and ancillary equipment so lost items can
be detected.

14
Testing (based on signal strength) may lead to additional testing
which may eliminate questions about sensitivity and
repeatability.
Establishing operating sensitivity
performance At least three tests per material type and aperture
position are generally considered realistic for most testing
Sensitivity performance should be maintained without
purposes. When good detection capability has been
need of operation attention. Unstable units which require
established during implementation, fewer tests may be
constant attention, or reject good product due to false
required as acceptable practice.
reject signals from plant vibration or other influences do
not provide value.
Statistical importance ultimately determines the number
of tests for each sample type to satisfy requirements
The best achievable sensitivity depends on product type,
within the manufacturer’s quality program, regulatory
size, and packaging materials, and should be selected
requirements, or customer demands.
in consultation with metal detector’s manufacturer, in
combination with test results of the product(s) being
inspected. Rejected Product Handling
Packaged product may contain detectable metal, even ◊ I solate and re-screen potentially contaminated
when it is free of staples or other metallic fasteners, product on test failure.
and has been determined to be the cause of false reject ◊ Investigate source of contaminant – trained personnel,
signals. off-line and within reasonable time.
◊ Identify the source of repeat detections contaminant.
Best achievable sensitivity should be established for each ◊ Stop production when multiple detections occur.
product being inspected. If a test sample’s signal is large
compared to that required to trigger detection (visible Documenting the Program
on the display panel of some detectors), there is a large
safety margin for detection. This information can be used Document the metal detection program as a set of
to determine how frequently the system needs to be internally controlled policies and procedures supported
verified. by all employees in contact with product. The scope
and detail of the procedures should be commensurate
Maintain records of sensitivity tests, and the results for with the size, complexity, and lines of communication
each product, during technical, operational, and sanitary within the organization. The most effective programs
audits of the system. are established, documented, operated, and maintained
within the framework of a structured food safety
Access to sensitivity adjustment controls should be limited management system supported by the management
to properly trained employees. Prevent access by others activities of the organization.
through password protection or lockout of the appropriate
controls. Relevant, meaningful documentation can be critical if
your company is investigated because of a customer
complaint. Documentation provides needed evidence of
Documenting the sensitivity standard compliance with production practices in effect.
Sensitivity standards should be expressed as the
minimum detectable ball size and indicated by the Minimal Requirements and Verification
nominal spherical diameter and material (e.g. 2.3 mm
diameter type 316 stainless steel). The minimal detectable Adopting a metal detection program should be a strategic
sphere size should be qualified in relation to the height decision for the company to avoid losing its importance
of the metal detector’s aperture, the product type, and and reducing support for its effective maintenance. The
application. needs and objectives of the company, the products made,
the manufacturing processes used, and the size and
The sensitivity standard should be properly documented, structure of the company should govern the system’s
effectively communicated throughout the organization, design and implementation. The program should aim
and readily available to appropriately trained verification to control production through the supply chain to the
personnel. final customer and support needs after shipment. The
program needs to be proactive, and should help prevent
contamination rather than merely detecting when it
Number of tests happens.
The quantity of samples required to test the metal
detector system should be derived from confidence Your metal detector will require annual documented
levels established when the system was designed and verifications conducted by a 3rd party. This helps ensure
implemented. The system’s capability to accurately and the purity of a company’s product and provide protection
repeatedly detect metal should match those results is of a company’s brand. Verifications are required by
being tested. Tests indicating only marginal detection government regulations, auditors, Hazard Analysis Critical
Control Point (HACCP) plans and most food retailers.

15
 Appendix List

Appendix A – Typical Processing Line

Grinders,
Raw Material Metal
Choppers,
Input Detector
Cutters, Mixers

Cookers Cookers
Formers etc. Formers etc.

Coolers, Slicers, Coolers, Slicers,

Dicers Dicers

Packagers, Packagers,
Cartoners, Cartoners,
Checkweigher Checkweigher

Metal Metal
Detector Detector

Load Out Palletizer Case Packer

16
 Appendix B – Reject Devices

Sweep Arm Divert Arm

Air Blast Push Arm (Ram)

Retracting Pulley
Running Retracted

Flop Gate

Running Retracted

17