Last week one of my overseas contacts rang and asked about checking if eddy currents could be used to check for fake/forged Gold (Au) bars. This is apparently becoming an endemic problem as the price of gold rises in response to global financial instability. The current method used by criminals is to plate Tungsten with a thin layer of gold (about 0.1mm). They do this because Tungsten is a very similar density to gold but worth 1% of the value. By doing this the criminals get something that weighs the same as a gold bar and also defeats the current favoured non-destructive testing method of X-Ray Fluorescence (XRF) equipment as this only tests the material surface. Drilling a hole to take a sample for an assay analysis is not permitted as of course that would remove valuable material. So then some other non-destructive test is required to test for this condition.
So I worked out in theory this was possible at 60kHz with a conductivity meter and informed my contact of this. The reason for this is that the conductivity of Gold is 45 Mega Siemens per m (MSm-1) (or if you prefer 77.6% IACS or resistivity 2.2 mp.cm) whereas Tungsten is 18 Mega Siemens per m (MSm-1) (or if you prefer 31.3% IACS or resistivity 5.5 mp.cm). However there is of course the skin effect to overcome (this is a phenomena that limits the eddy current concentration to the material surface). Well the good news here is that at 60kHz the traditional frequency for conductivity measurement the skin depth (average depth of penetration) in gold is 0.3mm and the gold plate is no more than 0.2mm.
A couple of days latter my contact phoned and reported that they had done some tests on forged and real gold bars and yes it worked very well.
One XRF supplier is very interested in contacting adding EC Conductivity functionality to their XRF system probably a special for Gold Bullion use.
We have also added the capability to the SigmaCheck software (should it be necessary in the future) to penetrate thicker plating.
John Hansen 2nd Dec 2011