[Metal Detectors] How it Works

Features - Feature

A closer look at one of the most important pieces of quality assurance equipment in your plant

October 24, 2007

If all metals were magnetic, finding them in your product and removing them would be a breeze. Unfortunately there are a number of metals that contain little or no iron or magnetic properties that can contaminate product and pose food safety hazards, thus making detection a much more complex process. Just how do metal detectors “see” those tiny particles of copper, lead or stainless steel baked inside a loaf of bread, drifting within a flow of chocolate or stuck inside a meat carcass? In 1990, Andrew Lock, founder of Safeline, put together a guide book for the industry on this subject. “He felt that most people didn’t understand metal detection technology, yet it is so important to food safety,” said Oscar Jeter, national sales manager for Safeline, Tampa, Fla. The resulting publication, “The Guide to Reducing Metal Contamination in the Food Processing Industry,” has been updated a number of times since then, most recently in September of this year, and is available from Safeline.

We used Lock’s Guide and a paper called “Metal Detectors for Food Processing” by Tim Bowser, an associate professor of biosystems engineering at Oklahoma State University and a food process engineer for food and agricultural products, to bring you this explanation of How it Works.

1. Nonferrous metal detectors implement a balanced three-coil system:

  • Three coils are wound exactly parallel on a non-metallic frame.
  • The center coil is connected to a high-frequency radio transmitter.
  • Two transmission-receiving coils sit on either side of the center coil.
  • Because the two outer coils are identical and the exact same distance from the center, they receive the same signal and produce an identical, balanced output voltage.

When a particle of metal passes through the system:

  • The high-frequency field is disturbed beneath one coil, changing the voltage and disrupting the balance.
  • The output fluxes from zero, producing a signal alerting the system to the presence of metal.
  • Depending on the specifics of the system, a rejection mechanism is generally activated, with the ideal result of removal of 100 percent of the metal and a minimum amount of salable product.

2. The opening through which the product passes has a great deal of effect on the sensitivity of the system.

  • The smaller the opening, or aperture, the more sensitive the detection will be. This is because the larger the opening gets, the further apart the coils need to be, which creates a gradual decrease in sensitivity capability.
  • For a similar reason, the exact center of the system is the least sensitive area, because it is the furthest from both coils. This difference in sensitivity between the center and the corners is the “sensitivity gradient,” a factor that is dependent on the coil and frame design.

Metal detectors are created to be highly sensitive and can be affected by external factors. So:

  • The coil arrangement is mounted inside a metal case with an opening in the center through which the product passes.
  • This metal case protects the detection device from external influence and adds strength and rigidity to reduce the effect of external vibrations, such as those from motors, pulleys, nearby equipment and even the detector’s own auto-reject device.
  • When testing a metal detector to purchase, test it under plant-specific conditions.

3. All products are not created equal in the eyes of the metal detector. Product composition has a direct effect on metal detection:

  • Product effect can be the result of product density, amount of moisture, or even certain ingredients as such factors can affect the conductivity of the signal or set off a false positive.
  • Moisture affects conductivity. Wet products create an interference signal in the detector that needs to be cancelled out before metal detection can begin.
  • The amount, or even existence, of such ingredients as salt, acid, fat and vitamins can create false positives, setting off the metal alert signal if sensitivity has not been properly calibrated.
  • To counter this, the sensitivity of the detector can be gradually reduced until the signal from the product is no longer detectable; a lower operating frequency can be selected; or operator-adjusted electronic circuits can be added which amplify and filter detector signals by differing amounts according to their characteristics. It is generally a combination of the three techniques that will provide the optimum results.

4. Where is the best place to put the metal  detector?

  • Beginning of the line — For meat products that could come in with metal tags, knife blades or twist ties, the start of the line can keep plant equipment from being damaged by incoming metal. However, if this is the only metal detector used, the plant can leave itself open to contamination later in the production process.
  • End of the line — Many plants prefer this placement as it is the final check before product is sent to the consumer. It can be a good option, Bowser said, if the final product is not too large or too dense, which could affect the detector’s degree of sensitivity.
  • In-line — Detectors may be placed at a point along the line that the plant has identified as a  critical control point for the product, or to protect expensive equipment that could be damaged by metal.
  • More than one of the above — Metal contamination can come in at any point in the production process: with raw materials, during processing, or from maintenance activity.

5. Sensitivity is defined as the diameter of a metal sphere of a specific metal which is just detectable in the center of the detector’s aperture.

  • Plants generally set their own internal requirements for sensitivity based on product characteristics and known susceptibility. “Sensitivity is best determined by an actual test of the product,” said Safeline’s Jeter.
  • Because metal detectors are an electronic instrument and are very susceptible to environmental conditions, they need to be tested and, if needed, recalibrated, at fairly regular intervals.

“And that interval, only the customer can answer,” Jeter said. If you test at 7:00 a.m., then again at noon, but the test fails at noon, what do you do about all the product run since 7:00 a.m.? he asked.

Because it is potentially contaminated, you need to determine the period of time you are willing to recheck or rework. Often the greater the value of the product, the shorter the time between tests.

6. Metal spheres are used as a standard in determining detector sensitivity capabilities because:

  • A sphere is the only object without orientation, that is, it appears the same from every angle, and no matter which direction it goes through the detector, it should have the same effect. In addition, tests run with spheres are duplicable and measurable. Compare, for example, the difference between a ball bearing and a pencil. No matter how it rolls, the ball bearing will have the same orientation each time it is passed through the system; a pencil, on the other hand, could go through vertically, sideways or cocked to any angle in between, giving the detector a different “vision” and a different test-reading each time.
  • However, what may appear to be a minute, harmless object without orientation can have a drastically different effect from a comparative object with orientation. For example, a piece of ferrous wire can be very difficult for the system to detect when it is lying at a 90-degree angle to the direction of flow. It is much easier to detect when aligned along the conveyor belt; non-ferrous wire is just the reverse. So, while a 2-millimeter sphere may be approximately the size of a bullet point in this article, a piece of wire of a significant length but a lesser diameter could pass through undetected.
  • As Jeter noted, “Real-world contamination is different than those little metal test spheres.”

CONCLUSION. Even when all product effects are accounted for, sensitivity is at its optimum and testing is conducted regularly, a metal detector should not be considered as insurance against metal contamination, Oklahoma State’s Bowser said. “Rather, it is a diagnostic device that can help discover metallic contamination,” he said. “Plans and procedures should be in place to prevent contamination.” 

Editor’s note: In an effort to examine the way certain products function, QA magazine will profile other facets of food production in future issues.