Issue 7.3, December 2003
Information within this issue may be outdated. Click here to view the most recent issue

Standardized Testing of Metal Detectors

Stemming from initiatives to promote standardization within humanitarian mine action, a European Committee for Standardization (CEN) Workshop has been established to standardize the testing of metal detectors. This has produced a CEN Workshop Agreement (CWA) that gives principles, guidelines and procedures for detector testing.

by T.J. Bloodworth and A.J. Sieber, JRC

Introduction

Metal detectors are an essential part of the toolkit of a humanitarian deminer. In recent years, demining end-users have required testing and evaluation of the available metal detectors prior to making procurement decisions. Many trials of the capabilities of metal detectors have been performed. However, the lack of an agreed standard for testing these instruments has limited the value of this work to the end-users. It is difficult to make cross-comparisons between instruments to determine which are best suited to any particular needs.

In response to this problem, CEN Workshop 7, “Humanitarian Mine Action—Test and Evaluation—Metal Detectors” (CW07) was established. CW07 has the objective of developing specifications for the testing and evaluation of metal detectors used in humanitarian mine clearance. The background to CW07 is presented here, as is a summary of the tests that have been specified.

History of CW07

Following a mandate given to CEN by the European Commission (EC), the Working Group CEN BT/WG 126 was set up with the aim of making recommendations and initiating standardization within humanitarian mine action. One of the first decisions of CEN BT/WG 126 was that a CEN Workshop be started in order to standardize the testing and evaluation of metal detectors. It was proposed that the EC Joint Research Centre (JRC) be the “driving force” of this process. One great advantage of a CEN Workshop is that it is open to all, not restricted to those from CEN member states.

In addition, the International Test and Evaluation Program for Humanitarian Demining (ITEP) gave its support to the idea of standardization of metal detector test and evaluation and requested that the JRC initiate the CEN Workshop. Support has also been given by the United Nations Mine Action Service (UNMAS) and by the Geneva International Centre for Humanitarian Demining (GICHD), which is responsible for International Mine Action Standards (IMAS). Close cooperation has been maintained with the GICHD.

CW07 was launched on 8 November 2001 in Brussels, with the approval of the Business Plan.1 It was agreed that JRC provide both the chairmanship and the secretariat, with standardization support from CEN via the Italian National Standards Body (UNI), the Italian CEN member. The aim of CW07 was to produce a CEN Workshop Agreement (CWA) giving principles, guidelines and testing procedures for the testing and evaluation of metal detectors.

Full meetings of the Workshop took place at JRC, Ispra, Italy, in December 2001, April 2002 and December 2002 at which the decisions were made about the tests that should be performed, how they should be done and how the test specifications should be written in the CWA. Between the April and December 2002 meetings, a small Drafting Working Group met twice—at Defence Research and Development Canada (DRDC) Suffield in June 2002 and in Ispra in September 2002—to make faster progress in producing the CWA.

Following the full meeting of CW07 in December 2002, a final version of the CWA was prepared. The final version was submitted to CEN in May 2003 for approval and publication. The Agreement was issued by CEN as CWA 14747:2003.2

Experience Applied in CW07

In formulating the standardized test procedures for the CWA, extensive use has been made of the test procedures developed and followed during the International Pilot Project for Technology Cooperation (IPPTC) for commercial off-the-shelf (COTS) metal detectors.3 Previous standardization work on demining testing has also been useful in the preparation of the CWA (e.g., the International Test Operations Procedures (ITOPs)4, 5, 6). Studies of metal detector responses7 and tests used in previous metal detector trials8, 9 gave valuable information. In addition, an existing U.S. military Performance Specification10 for metal detectors and a standard for metal detectors used for detection of concealed weapons and contraband in the U.S. penal system11 were considered in CW07.

The most important contribution was the combined experience of the members of CW07 that was brought to the Workshop. Manufacturers, testing laboratories, researchers into metal detection and those with experience of using and testing detectors in the field all contributed to creating the test specifications.

General Principles of CWA 14747

Figure 1: Geometry for testing metal detectors.

CWA 14747 establishes the general principles for detector testing. One of the most important of these is that all of the tests of detection capability are based on the idea that the only output that the detector gives is a yes/no alarm signal. This means that all of this testing is based upon a correct alarm/no alarm judgement. For some detectors, this is clear; for others, less so. A criterion is therefore defined; detection has occurred when a repeatable, non-intermittent alarm indication is produced.

To quantify detection capability, the maximum detection height or depth of a target is used. The distance is always measured from the top of the target concerned. When testing in air, the height of the sensor from the top of the target is measured. In soil, the depth of the top of the target below soil level and the sweep height of the sensor above the soil are measured. Figure 1 shows this convention schematically.

In-Air Tests

Figure 2: Minimum target detection capability curves of three metals.

The first tests defined in the CWA are tests of detection capability in air. The benchmark test of detection capability is defined as the in-air maximum detection height of a chrome steel ball that is 10 mm in diameter. This test is used to check whether detection capability changes with sweep speed over a target, whether it is repeatable on setup and whether it drifts. These tests are intended to be performed in controlled conditions.

Specifications are given to measure the way that detection capability changes as a function of sensor height above target. This is actually done by measuring the maximum detection height of metal balls. The results are then expressed in terms of a minimum target that is detectable at a given height. Metal balls do not closely resemble most of the metal components found in mines, but this exercise is a way of quantifying detection capability and gives a reference (albeit an arbitrary one) against which other targets can be compared. The arguments for using metal balls are given elsewhere.12

Figure 3: Manual measurement of detection capability curve.

Chrome steel balls are chosen as reference standards, but tests for other metals are also included. These give an indication of the relative detection capability for these metals. The results are reported as minimum target detection curves as shown in Figure 2. Figure 3 shows how the measurement is done.

Two forms of a test to measure the sensitivity profile (often called the “footprint”) of a detector are included. One of these requires that the alarm output level is recorded in some way—this is the only part of the CWA where this is needed. Figure 4 shows an example of this type of sensitivity profile. The other test uses the maximum detection height principle already established to define detection contours.

Figure 4: Sensitivity profile of a metal detector at three different heights above a target.

There are tests of the immunity to environmental and operational conditions of the detector performance. The detection capability, as measured by maximum detection height in air, is always used to give a quantifiable measure of any variation. Tests are given for the effect of temperature extremes and moisture on the sensor head, for example.

 

 

 

Continue ->