NDT.net • May 2005 • Vol. 10 No.5

Thickness Measurement of Chromium Layers on Stainless Steel Using the Thermoelectric Method with Magnetic Readout (TEM)

Johann H. Hinken, Bastian Barenthin, Josef Halfpaap, Christoph Moebes, Herbert Wrobel, Christian Ziep;
Hochschule Magdeburg-Stendal (FH), University of Applied Sciences, Magdeburg, Germany
Michael Hekli;
Hartchrom AG, Steinach, Switzerland

Corresponding Author Contact:
Email: johann.hinken@et.hs-magdeburg.de, Internet: http://www.elektrotechnik.hs-magdeburg.de/...


The thickness measurement of thick chromium layers on stainless steel is difficult using standard methods such as eddy current testing or the X-ray fluorescent method. Therefore, a new procedure is under development, which uses the thermoelectric method with magnetic readout (TEM). First measurements are described using a stainless steel rod of circular cylindric coruscation coated with chromium.

1. Introduction

When thicknesses of metal coatings on metal substrates are to be measured this can be done using various physical principles. If only the substrate metal is ferromagnetic the magnetic force method can be used. If the electrical conductivities are different then eddy current testing can be used. And if the layer thickness is not too large then the X-ray fluorescent method is applicable. In the case of chromium layers on stainless steel however, the electrical conductivities are similar, so excluding eddy current testing. Also the substrate metal is not strong enough ferromagnetic so excluding the magnetic force method. And in the case of chromium layer thicknesses greater than a few tens of micrometers also the X-ray fluorescent method is not applicable.

Therefore it was to be tested if the Thermoelectric Method with Magnetic Readout (TEM), which in general is applicable for thickness measurements, [1], can be used also in this special case.

The TEM works as follows. By external means a proper temperature gradient is generated in the device under test (DUT). This causes sources of electrical voltages and currents at the interface of regions with different thermoelectric Seebeck coefficient. The resulting currents in the metallic DUT cause also magnetic fields outside the DUTs. Generally such magnetic fields are rather small and can only be detected by very special magnetic sensors. The currents and the magnetic fields are proportional to the conductances of the metallic parts. Therefore, in the case of a metallic layer a strong dependence of the magnetic field amplitude on the layer thickness can be expected.

More information on the TEM can be found in [2].

The mentioned voltages, currents and magnetic fields are also proportional to the difference of the absolute differential thermoelectric coefficient. In case of chromium (e ≈ +16µV/K) on stainless steel (e being only a few µV/K [3]), this difference is a significant value.

2. Measurements

Figure 1 shows the test rod

Fig.1: Partial view of the test rod

It is a circular cylindric rod of 10 mm diameter made from stainless steel and eletrolytically covered with chromium. There are six zones of different chromium layer thicknesses ranging from 0 µm to about 200 µm.

The measurement sequence using a semi automated system is as follows.
A non magnetic short brass cylinder (diameter 4 mm, height 6 mm) is heated to about 100 °C and then placed onto the test rod. The adjacent special magnetometer detects the magnetic field variation in its time dependence. The special magnetometer is similar to that described in [4] but is a version specialised for the present measurement task. This special magnetometer for example excludes a strong magnetic background field as is stems from the earth magnetic field and man-made magnetic noise.

Fig.2: Typical time dependence of voltage or magnetic field strength, respectively, of two single measurements

As an example, figure 2 shows the time dependence of the magnetic field strength during two single measurements following one after the other. The curves show a steep increase after placing the heat element on the rod followed by a smooth decrease. The height of this voltage pulse at the output of an amplifier or the recalculated magnetic flux density in Tesla was used as the characteristic signal. The height of these pulses are shown in figure 3 versus the layer thickness. This layer thickness was deduced from measurements with a micrometer gauge.

Fig.3: Measuring results

The vertical bars in figure 3 describe the min/max variation of five single measurements with the average values in the centre. This variation is maximally about +/- 10 % and can certainly be reduced by further improving the system.

Also the small offset, which can especially be seen at layer thickness zero micrometer, can be balanced to zero.

The measuring values are to a good approximation proportional to the layer thickness, which corresponds to our theoretical understanding of the process.

3. Conclusions

The TEM method has been used to measure thicknesses of chromium layers on stainless steel. This was a feasibility study. The results encourage to continue research and development on this subject towards measurement systems, which are applicable in industrial environment. This includes the generalisation of the method to DUTs of other geometry and other material. It can be expected that using this method can lead to the closing of some open application niches in the non-destructive measurement of metallic layer thicknesses.


  1. Patent DE 102004021450.6, filed 30 April 2004
  2. http://www.finoag.com/fitm/n6.html
  3. W. Morgner: Introduction to Thermoelectric Nondestructive Testing. Materials Evaluation, Sept. 1991, 1081-1087
  4. www.elektrotechnik.hs-magdeburg.de/Mitarbeiter/hinken/news/n12.htm

© NDT.net |Top|